Index

Muhammad Shahzad
Associate Professor
Department of Computing

School of Electrical Engineering and Computer Science (SEECS)
National University of Sciences and Technology (NUST)
NUST Campus H-12, Islamabad
Tel: 05190852400
Email: muhammad.shehzad@seecs.edu.pk
Web: http://seecsnust.seecs.edu.pk/faculty/muhammadshahzad.html

Education:
PhD - Radar Remote Sensing, 3-D Computer Vision & Image Analysis, Point Cloud Processing - Technical University of Munich (TUM), Germany - Germany - 2016

Dr. Muhammad Shahzad received B.E. degree in electrical engineering from the National University of Sciences and Technology, Islamabad, Pakistan, M.Sc. degree in autonomous systems (robotics) from the Bonn Rhein Sieg University of Applied Sciences, Sankt Augustin, Germany, and his PhD degree on radar remote sensing & image anaylsis at the department of Signal Processing in Earth Observation (SiPEO), Technische Universität München (TUM), Munich, Germany in 2004, 2011 and 2016 respectively. His Phd topic was automatic 3-D reconstruction of objects from point clouds retrieved from spaceborne synthetic-aperture-radar (SAR) image stacks. During his PhD, he proposed several novel and sophisticated methodologies and achieved several “firsts” in the international community, e.g. the first façade models from interferometric SAR (InSAR) data, its first reconstruction of an entire city, the first 2D building footprint models from InSAR and the first reconstruction of even individual trees. His work was closely linked to the TerraSAR-X and TanDEM-X satellite missions, the biggest German Earth observation endeavors ever, with both scientific as well as commercial applications. Besides, he also attended twice two weeks professional thermography training course at Infrared Training Center (ITC), North Billerica, Massachusetts and Portland, Oregon, USA in 2005 and 2007 respectively. Moreover, he worked as a Visiting Research Scholar at the Institute for Computer Graphics and Vision, Technical University of Graz, Austria in 2015/16. Since 2016, he is working as Associate Professor (promoted from Assistant to Tenured Associate Professor in Feb 2022) at the School of Electrical Engineering and Computer Science (SEECS), National University of Sciences and Technology (NUST), Islamabad. Furthermore, he is also a Co-Principal Investigator of recently established Deep Learning Laboratory (DLL) under the umbrella of National Center of Artificial Intelligence (NCAI), Islamabad. His research interests include application of deep learning for processing unstructured/structured 3-D point clouds, optical RGBD data, and very high-resolution radar images.

1 .
Transformer based UAV Geo-Localization:

In this research, we aspire to utilize the acquisition method as it becomes more flexible then previous and produce high resolution images for navigating and localizing Unmanned Aerial Vehicle (UAV). Unmanned Aerial Vehicles (UAVs), also known as aerial drones, are pervasive in enormous fields, including photography, aerial mapping, agriculture, surveillance, search and rescue, and parcel deliver...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2022
Status: Completed
Advisor
Muhammad Imran Malik [DoC]
Committee Members
Safdar Abbas Khan [DoC] Muhammad Khuram Shahzad [DoC]
Students
MAHRUKH KHALIL [MSDS-1]

2 .
A Multi-Modal Architecture for Visual Question Answering using Knowledge Graphs:

Visual Question Answering (VQA) is the task of answering the questions based upon the image. Many of the previous techniques used CNNs, RNNs LSTM, GCN to acquire visual contents from the image. However, information other than objects that are visually seen is not catered. Some recent techniques used knowledge graphs to acquire external knowledge related to images using knowledge bases to answer t...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2022
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Arsalan Ahmad [DoC]
Students
IFRAH MAQSOOD [MSCS-9]

3 .
Enhancing Cotton Crop Mapping in Pakistan: Integrating Transfer and Active Learning with Remote Sensing Technologies:

Cotton cultivation is pivotal to Pakistan's economy, contributing significantly to its GDP and accounting for a substantial portion of its export earnings. However, the country currently faces a steep decline in cotton production due to challenges such as reduced sowing area, adverse weather conditions, and pest attacks. This underscores the urgent need for enhanced agricultural data to support e...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2022
Status: Completed
Advisor
Zuhair Zafar [DoC]
Committee Members
Muhammad Moazam Fraz [DoC] Syed Imran Ali [DoC] Fahad Ahmed Satti [DoC]
Students
Noor Us Sabah [MSDS-2K20]

4 .
National Water Reservoirs Monitoring through Time Series Remote Sensing:

Efficient management of water resources plays a critical role in countries characterized by high aridity levels and vulnerability to floods, as seen in the case of Pakistan. However, challenges arise, largely stemming from the scarcity of resources required for effective water management. Fortunately, Remote Sensing field has emerged as a significant player, providing time series data, and allowi...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2022
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Zuhair Zafar [DoC] Fahad Ahmed Satti [DoC]
Students
Rukhsana Zafar [MSDS-2K20]

5 .
Multi-year Wheat Mapping from Satellite Time Series Leveraging Superpixel-Based Active Learning for Data Sparse Regions:

Given the escalating levels of food insecurity, particularly in underdeveloped nations,
automated solutions of agriculture land distribution management need to be put forward. A recent surge in the generation of high-resolution satellite data has improved the automated analysis of agriculture land distribution, providing valuable information regarding crop health monitoring, agro-meteorological ...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2022
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Zuhair Zafar [DoC] Fahad Ahmed Satti [DoC] Nazia Perwaiz [DoC]
Students
Vaneeza Mehmood [MSDS-2K20]

6 .
Vehicle re-identification for Visual Surveillance:

Intelligent video surveillance systems are becoming an increasingly important major area of artificial intelligence as the use of motor vehicles in today's transportation networks grows. In recent times re-identifying vehicles over a surveillance camera network have been proven to be the most effective method of efficiently controlling traffic, upholding the law, gathering information, and progra...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2022
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Zuhair Zafar [DoC] Ahsan Saadat [DoC]
Students
HASAN ALI ASGHAR [MSCS-9]

7 .
Towards Large-scale Unsupervised Face Recognition in Videos:

Video content is ubiquitous in the modern world and there is a growing need for automated methods to extract information from videos. Face-based person retrieval is a particularly interesting task in this domain which involves the use of face recognition to track the appearances of people in video data. It is useful in a variety of applications, from video analytics and indexing to video surveill...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2022
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Zuhair Zafar [DoC] Rai Sajjad [HU]
Students
Atif Khurshid [MSCS-2K20]

8 .
Biasness of World Leaders: A Sentiment Analysis of World Leaders’ Twitter Data:

The United Nations' Sustainable Development Goals (SDGs) are a set of 17 goals that were adopted in 2015 as a blueprint for a better and more sustainable future. The goals are interconnected and address diverse global challenges such as poverty, inequality, climate change, peace, and justice. World leaders play an integral role in creating awareness and establishing policies that support attainin...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2022
Status: Completed
Advisor
Shah Khalid [DoC]
Committee Members
Bilal Ali [DoC] Hasan Ali Khattak [DoC]
Students
SAMIR KHAN [MSCS-9]

9 .
Hyper-spectral Document Image Segmentation:

Separating overlapping text and non-text from images is one of the challenging tasks in
document analysis and segmentation. It was previously difficult to separate text and
non-text due to a lack of information available in existing datasets consisting of
RGB Images. A hyper-spectral image depicts pixel information into a number of channels than can overcome the constraint of limited pixel inf...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2021
Status: Completed
Advisor
Muhammad Imran Malik [DoC]
Committee Members
Muhammad Shahzad [DoC] Faisal Shafait [DoC]
Students
Hasan Irtaza Mirza [MSCS-8]

10 .
Self-Supervised Crop Type Classification using Multi-Spectral Remote sensing:

Accurate and efficient crop classification using multispectral remotely sensed data is essential for crop yield estimation and agricultural management. However, one of the major challenges in this task is the limited availability of labeled data, which hinders the ability to achieve good classification results. In our study, we propose a two-step approach to address this challenge and improve cro...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2021
Status: Completed
Advisor
Zuhair Zafar [DoC]
Committee Members
Muhammad Moazam Fraz [DoC] Fahad Ahmed Satti [DoC]
Students
SYEDA FAIZAN FATIMA [MSCS-9]

11 .
Tapping the power of clinical data to identify multidimensional Irritable Bowel Syndrome (IBS) endo-phenotype:

The pathophysiology of irritable bowel syndrome (IBS) is complex and remains poorly understood. A heritable component is long known in IBS, however gene-evaluatory efforts have so far been scarce and no unequivocal IBS risk locus has yet been identified. This is partly because the phenotypic heterogeneity of IBS doesn’t just encompass GI symptoms, but also a wide range of co-morbidities and the p...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2021
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Rabia Irfan [DoC] Yasir Faheem [DoC]
Students
USWAH AHMAD KHAN [MSCS-9]

12 .
Uncovering Natural Clinical Data-Driven Clusters of Irritable Bowel Syndrome (IBS) Endo-phenotypes Using Machine Learning Techniques:

The physiopathology of Irritable Bowel Syndrome (IBS) is complex and multidimensional that has continued to baffle both researchers and clinicians. IBS has long been correlated with genetic factors, indicating a hereditary element and the efforts to identify these genetic risk factors for IBS have been restricted and inconclusive.

In recent years, there has been considerable interest in the a...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2021
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Rabia Irfan [DoC] Fahad Ahmed Satti [DoC]
Students
AHSAN IJAZ [MSDS-1]

13 .
Automatic Covid-19 Detection Using Deep Features:

Coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2 (SARS CoV-2), first diagnosed in late 2019, spread globally, infected millions with high mortality rate. Pharmaceutical industry fails to provide the essential infrastructure to detect the suspected cases. Medical health care systems are in urgent need of timely and precise identification of COVID-19 in...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2021
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Arsalan Ahmad [DoC] Muhammad Imran Malik [DoC]
Students
MUHAMMAD YASIR ALI KHAN [MSCS-7]

14 .
Vision Based 3-D Localization of UAVs Using Deep Image Matching:

Vision Based 3-D Localization of UAVs Using Deep Image MatchingDrones and UAVs are used both professionally and privately and they rely heavily on onboard GPS, altimeter, survey meter along with other navigational equipment for completing their automated flight paths. These systems are prone to failure as well as jamming, hence they require a backup plan to complete their flight. We focus on usin...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2021
Status: Completed
Advisor
Hasan Ali Khattak [DoC]
Committee Members
Muhammad Moazam Fraz [DoC] Muhammad Imran Malik [DoC]
Students
Mansoor Khurshid [MSCS-9]

15 .
Squeeze-and-Residual-Excitation in Holistic Attention Network for Improved Super Resolution in Remote Sensing Imagery:

Super-resolution (SR) provides an excellent approach of improving applications related to remote-sensing imagery, hence, tasks like object detection, classification and segmentation can greatly benefit from high performing SR models. Extensive research is being done in the field of SR for both natural and remote-sensing imagery where deep convolutional neural networks (CNN) have achieved tremendo...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2021
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Zuhair Zafar [DoC]
Students
BOSTAN KHAN [MSCS-9]

16 .
CLOUD: Clustering Large Online Unrecognized Detections:

This thesis idea presents a variety of face verification, recognition and other face analytics tasks. Multiple thesis will be created from this idea

Large-Scale Face Search for Visual Analytics
Although the face recognition has matured enough, still searching a large collection of unconstrained face images remains a difficult problem. This thesis will aim to search faces in a large customized...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2021
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Muhammad Jawad Khokhar [ECM]
Students
MUHAMMAD EHSAN UL HAQ [MSCS-9]

17 .
Machine Learning-based Predictive Maintenance framework for an Industrial Internet of Things:

The Industrial Internet of Things (IIoT) is an industrial framework comprised of machines, devices, and robots collaborated through the wireless communication channel for better utilization of available resources. In the industrial revolution, devices are getting intelligent, making autonomous decisions. Thus, such a situation demands regular maintenance to avoid system failure, financial loss an...


Project Level: PG
Project Type: Thesis Knowledge Area: Parallel and Distributed Computing
Year: 2021
Status: Completed
Advisor
Asad Waqar Malik [DoC]
Committee Members
Muhammad Shahzad [DoC] Anis Ur Rahman [DoC]
Students
TUBA ILYAS [MSIT-19]

18 .
Crop Type Classification using Multi-temporal Satellite Imagery:

Remote sensing data is available free of cost with an ever-increase in the number of satellites. This satellite imagery can be used as raw input from which cultivated/non-cultivated and crop fields can be mapped. Previous trends included the use of traditional ML techniques and standard Convolutional Neural Networks (CNNs), Long Short Term Memory (LSTM) for such mappings. In this research work, w...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2021
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Zuhair Zafar [DoC] Nazia Perwaiz [DoC]
Students
ASIM HAMEED KHAN [MSCS-9]

19 .
Multi-lingual Traffic Signboard detection and Recognition in Natural Scene Images:

Self-driving vehicles can establish their driving tasks i.e. path planning, navigation, and motion control by understanding the road traffic signboards. Though, convolutional neural networks have demonstrated better applicability in scene-text detection and recognition. However, real-time scene text detection and understanding of traffic signboards is a challenging task. This study focuses on mul...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2021
Status: Completed
Advisor
Faisal Shafait [DoC]
Committee Members
Muhammad Shahzad [DoC] Adnan Ul-Hasan [ECM] Muhammad Imran Malik [DoC]
Students
MUHAMMAD ATIF BUTT [MSCS-9]

20 .
Deep Transformer-based Semantic Segmentation using Remote Sense Data:

Geological surveys, the management of urban resources, and the monitoring of natural disasters all make extensive use of semantic segmentation for remote sensing images (RSIs). Transformer-based models and CNN-based models have been used recently to solve remote sensing segmentation problems. Several CNN-based models have been proposed and performed SOTA on different benchmarks. Transformers are ...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2020
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Muhammad Moazam Fraz [DoC] Arsalan Ahmad [DoC]
Students
SYEDA FARHAT UL AIN [MSCS-8]

21 .
MIST computing Framework For Delay Sensitive Medical Applications:

Smart cities play an important role in betterment of life standards of human beings by providing
effective urban environment. This idea come into existence based of Internet of things (IOT) as it is
predicted that Internet of Things (IOT) will have 50 billion devices connected and a data of 507.5
zettabytes in a decade. A subset of Internet of things (IOT) is Internet of medical things (IOMT)....


Project Level: PG
Project Type: Thesis Knowledge Area: Parallel and Distributed Computing
Year: 2020
Status: Completed
Advisor
Asad Waqar Malik [DoC]
Committee Members
Muhammad Shahzad [DoC] Anis Ur Rahman [DoC]
Students
ASFANDYAR KHAN [MSIT-18]

22 .
Orientation Aware Object Detection in Visual Data:

Machine learning has achieved great successes in the area of computer vision, especially in object recognition or classification. One of the core factors of the successes is the availability of massive labeled image or video data for training, collected manually by human. Labeling source training data, however, can be expensive and time consuming. Furthermore, a large amount of labeled source dat...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2020
Status: Transferred to Non-Thesis
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Sohail Iqbal [DoC]
Students
Fareeha Ikram [MSCS-8]

23 .
A Step-One Visual Learning App for children leveraging Knowledge Aware Visual Question Answering framework:

VQA models perform very well when asked questions within the visual domain of the image but they are not able to answer the questions that also require external knowledge about the contents of the image. In order to address this problem, we need domain-specific datasets to train and benchmark the performance of future models on the questions, dependent on external knowledge. Creating large VQA da...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2020
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Omar Arif [DoC]
Students
AHMED JAMSHED [MSCS-8]

24 .
Crop Identification Using Remote Sensing technology:

Crop identification can be done by many ways. One way is to use remote sensing technology in which data is collected remotely via sensors. Satellites are the most sophisticated example of remote sensing technology. Satellites collect data in the form of bands and each band serves a separate purpose. We can use publicly available data to identify crop types using machine learning or deep learning ...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2020
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Muhammad Moazam Fraz [DoC] Arsalan Ahmad [DoC]
Students
SYED NAQVI [MSCS-8]

25 .
HistoSeg : Quick attention with multi-loss function for multi-structure segmentation in digital histology images:

According to Cancer Research UK, there are around One Million cancer deaths every year between 2016-2018. Cancer accounts for more than a quarter (28%) of all deaths during 2018. Medical experts distinguish over 100 different types of cancer. However, it is increasingly accepted that every single type of cancer is a heterogeneous disease requiring ‘personalized and precise’ treatment tailored to ...


Project Level: PG
Project Type: Thesis Knowledge Area: Computational Science
Year: 2020
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Rafia Mumtaz [DoC]
Students
SAAD WAZIR [MSCS-8]

26 .
Large-Scale Face Detection in the Wild: A Comparison Study:

Various applications like face analysis, recognition, reidentification exist where the use of Face Detection is necessary as their preprocessing algorithm in the pipeline. There have been extensive studies done in the domain of Face Detection in the past, and various robust algorithms have been proposed and evaluated on different datasets. Such techniques are also deployed in various applications...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2020
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Muhammad Moazam Fraz [DoC] Muhammad Jawad Khokhar [ECM]
Students
HAFIZ SYED AHMED QASIM [MSCS-7]

27 .
Identifying human emotions from single walk step using a body-mounted inertial sensor:

Emotion recognition through gait analysis is a key area of research due to its significant academic and commercial potential. With the recent developments in hardware technology, most of the modern digital devices such as smart bracelets, smart watches, smart phones, smart insoles etc. are equipped with inertial sensors. These sensors can be used in a body mounted fashion to analyze human motion ...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2020
Status: Completed
Advisor
Qaiser Riaz [DoC]
Committee Members
Muhammad Shahzad [DoC] Muhammad Moazam Fraz [DoC] Dr. Muhammad Zeeshan [DoC]
Students
HAMNA KALEEM [MSDS-1]

28 .
Soil Moisture Mapping Using Geospatial Technologies for Precision Agriculture:

Soil moisture content plays an important role in the growth of the crop. It is important to determine the level of moisture precisely for the effective development and yield of the crop. Conventional farming techniques rely on farmer’s knowledge base for crop irrigation, which results in wastage and non-uniform distribution of water. There are several factors involved in determining the moisture ...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2020
Status: Transferred to Non-Thesis
Advisor
Rafia Mumtaz [DoC]
Committee Members
Muhammad Shahzad [DoC] Omar Arif [DoC]
Students
SAADAT ALI [MSCS-9]

29 .
Classification and Segmentation of 3d Point Clouds based on deep learning:

Deep learning methods are being extensively used to analyze 3D objects inspired by the successes of these techniques in 2D images and 1D texts. Earlier work such as pointnet++ and ACNN presents a methodology which wisely captures the local neighborhood with increasing contextual scale. However features learnt at different scales need to be treated more appropriately keeping in view the proximity ...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2020
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Muhammad Moazam Fraz [DoC] Asif Ali [DoC]
Students
RABBIA HASSAN [MSCS-8]

30 .
Deep emotion recognition by analyzing inertial data of human gait:

The Internet of Things (IoT) anticipates a future in which commonplace products will be incorporated with sensors. Sensors can detect both living environs and the human body, allowing for easier contact with the physical world. Sensors, such as inertial sensors, are found in devices such as smartphones, smartwatches, and fitness bands. There is a need for deep learning models that can work for th...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2020
Status: Completed
Advisor
Qaiser Riaz [DoC]
Committee Members
Muhammad Shahzad [DoC] Dr. Muhammad Zeeshan [DoC] Muhammad Moazam Fraz [DoC]
Students
HAMZA IMRAN [MSCS-8]

31 .
Medical Image Captioning:

Image captioning refers to the process of generating textual description from an image based on the objects and actions in the image. This project will focus on medical image captioning. For example, given a cardiovascular ultrasound dynamic images, the object will be to automatically generate diagnostic report.


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2020
Status: Transferred to Non-Thesis
Advisor
Omar Arif [DoC]
Committee Members
Muhammad Shahzad [DoC] Muhammad Moazam Fraz [DoC]
Students
RANA WALEED MANZOOR [MSCS-8]

32 .
Energy-efficient data dissemination for large-scale smart farming using reinforcement learning:

Smart farming is essential to increase the crop production for which precision in agricultural parameters
is crucial for the growth of crops. To achieve this, modern technology is required to enable accuracy in
fertilizing, watering and adding pesticides to the crops as well as monitoring the conditions of
environment. Now a days, more and more sophisticated sensors are developed but to achiev...


Project Level: PG
Project Type: Thesis Knowledge Area: Networking and Communications
Year: 2020
Status: Completed
Advisor
Asad Waqar Malik [DoC]
Committee Members
Muhammad Shahzad [DoC] Muazzam A Khan Khattak [DoC] Anis Ur Rahman [DoC]
Students
MUHAMMAD YASIR ALI [MSIT-18]

33 .
Price & Load Forecasting in a Deregulated Energy Market using a Hybrid Deep Neural Network Approach:

In many countries including Pakistan, the electric power industry has been undergoing
deregulation through the introduction of competition. Accurate System Marginal Price
and Load forecasts play an important role in an economic power dispatch, system
reliability and planning.  In a competitive wholesale Energy market, individual
generating stations submit their pricing bids on a day-ahead bas...


Project Level: PG
Project Type: Thesis Knowledge Area: Algorithms and Complexity
Year: 2020
Status: Completed
Advisor
Muhammad Shahzad [EE]
Committee Members
Muhammad Shahzad [DoC] Faisal Shafait [DoC] Wajahat Hussain [EE]
Students
MUHAMMAD FURQAN AZAM [MSEE-9]

34 .
Evaluating the privacy features of privacy preserving maps:

3D maps reveal the geometry of the scene. This is a privacy violation. One can detect which objects are present in the scene. Furthermore, once can also locate them. These 3D maps are required for multiple applications, e.g., localization of agents, AR applications etc. Is it possible to achieve localization without revealing the 3D geometry of the environment? Recently a privacy preserving mappi...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2020
Status: Completed
Advisor
Wajahat Hussain [EE]
Committee Members
Muhammad Shahzad [DoC] Syed Taha Ali [EE] Muhammad Latif Anjum [EE]
Students
MUHAMMAD WALID QAMAR [MSEE-8]

35 .
Attention based bidirectional GRU hybrid model for inappropriate content detection in Urdu language:

Broadcast media is monitored by the organizations who wish to keep track of the items being aired such as Pakistan Electronic Media Regulatory Authority (PEMRA). Of late, the broadcasting content is increasingly available online in the form of text and videos as well. Timely identification of unsuitable or banned content broadcasts can speed up the process of taking legal action for the violation...


Project Level: PG
Project Type: Thesis Knowledge Area: Information Management
Year: 2020
Status: Completed
Advisor
Rabia Irfan [DoC]
Committee Members
Muhammad Ali Tahir [DoC] Faisal Shafait [DoC] Muneer Ahmad [DoC]
Students
EZZAH SHOUKAT [MSIT-19]

36 .
AI enabled framework for classification of Windows portable executable malware using static feature analysis:

Malware refers to malicious software perpetrators dispatch to infect individual computers or an entire organization’s network. It exploits target system vulnerabilities, such as a bug in legitimate software (e.g., a browser or web application plugin) that can be hijacked. A malware infiltration can be disastrous — consequences include data theft, extortion or the crippling of network systems.
Ma...


Project Level: PG
Project Type: Thesis Knowledge Area: Information Assurance and Security
Year: 2020
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Muhammad Khuram Shahzad [DoC] Arsalan Ahmad [DoC]
Students
WARDA ASLAM [MSCS-8]

37 .
Deep learning based human age estimation using inertial data of normal walk :

Human gait enables information about soft biometrics features in multiple scenarios where the environment and visual cues are not known. This characteristic of human gait has numerous application in the field of surveillance, advertisement and monitoring and provides a practical approach in numerous real world applications such as motion sensor, alarms and detections. Moreover, since human gait s...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2020
Status: Completed
Advisor
Qaiser Riaz [DoC]
Committee Members
Muhammad Shahzad [DoC] Dr Hasan Tahir [DoC] Muhammad Moazam Fraz [DoC]
Students
KHAWAJA ALI ARSHAD [MSCS-7]

38 .
Localisation of Abnormalities in EEG signal:

Automatic analysis of EEGs provides a clinical decision support tool that can reduce time to diagnosis and assist clinicians in real-time monitoring applications. In this study, a high-performance automated EEG analysis system based on principles of deep learning is proposed.


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2020
Status: Completed
Advisor
Faisal Shafait [DoC]
Committee Members
Muhammad Shahzad [DoC] Adnan Ul-Hasan [ECM] Muhammad Ali Tahir [DoC]
Students
TAYYABA FAISAL [MSCS-8]

39 .
Measuring the effects of Inquiry-based Pedagogy in a Web Enhanced Learning Environment at Secondary Level:

Traditional history classroom relies heavily on rote memorization of “facts” narrated by a few scholars. This study will use the concepts of “doing history” to help the students to think critically about their past. Inquiry based content will be developed, giving students access to multiple online resources to corroborate and interpret. The study will assess the effect on the perceptions of stude...


Project Level: PG
Project Type: Thesis Knowledge Area: Social Issues and Professional Practice
Year: 2019
Status: Completed
Advisor
Safdar Abbas Khan [DoC]
Committee Members
Muhammad Shahzad [DoC] Erum Afzal [DoC] Tahira Lashari [DoC]
Students
IQRA MALIK [MSITL-6]

40 .
Enhancing Intrapersonal Intelligence for Strong Vocational Identity:

This research will focus on developing self-awareness in the learners through an intervention. Students usually go through a lot of anxiety and confusion while deciding their college major/career. This happens due to the lack of career awareness and low self-concept. The significance of self-knowledge is highlighted in various phases of an individual’s life but it is especially needed while takin...


Project Level: PG
Project Type: Thesis Knowledge Area: Social Issues and Professional Practice
Year: 2019
Status: Completed
Advisor
Safdar Abbas Khan [DoC]
Committee Members
Maajid Maqbool [DoC] Erum Afzal [DoC] Jaudat Mamoon [DoC]
Students
HAREEM ALI [MSITL-6]

41 .
Investigating student’s achievement using equitable assessment practices in secondary school classrooms:

The assessment phase plays a very important role during the learning processes since from this phase the knowledge acquired by them is validated. Various assessment procedures are used whereby the strengths and weaknesses of students are detected. However, to do so, the assessments designed by teachers in our education system do not always respond to the needs and characteristics of the students....


Project Level: PG
Project Type: Thesis Knowledge Area: Information Management
Year: 2019
Status: Completed
Advisor
Safdar Abbas Khan [DoC]
Committee Members
Maajid Maqbool [DoC] Tahira Lashari [DoC] Erum Afzal [DoC]
Students
MISBAH JAVED [MSITL-6]

42 .
Towards efficient resource utilization in collaborative fog environment using multi-armed bandit:

Vehicles are connected to roadside units (RSUs) that help in task offloading and
provide support in the storage and management of tasks. Cloud provides storage and connection to the vehicles but the latency-sensitive vehicles need fast response. During a small time frame, a vehicle has to decide where to send the task for execution i.e. to execute it locally, to forward it to RSU, or to ask for ...


Project Level: PG
Project Type: Thesis Knowledge Area: Networking and Communications
Year: 2019
Status: Completed
Advisor
Asad Waqar Malik [DoC]
Committee Members
Muhammad Shahzad [DoC] Anis ur Rahman [DoC] Muazzam A Khan Khattak [DoC]
Students
RABIA SALEEM [MSCS-8]

43 .
Effectiveness of digital technology on engagement for enhancing English language skills among primary school students :

Primary Education is always considered very essential for any society because it guarantees learners' cognitive, affective, social, psychomotor developments as well as their readiness not only for elementary but for the higher levels of education (Ahmad, 2019). It produces the sound basis for the education system. The developed and the developing nations are offering free and compulsory education...


Project Level: PG
Project Type: Thesis Knowledge Area: Social Issues and Professional Practice
Year: 2019
Status: Completed
Advisor
Tahira Lashari [DoC]
Committee Members
Muhammad Shahzad [DoC] Jaudat Mamoon [DoC] Maajid Maqbool [DoC]
Students
HAFIZ MUHAMMAD ABDULLAH [MSITL-6]

44 .
Learning Category Aware Aggregation for Large Scale Point Cloud Semantic Segmentation:

This research proposed a method of semantic segmentation on large scale point clouds which addresses the feature aggregation of joint areas of objects in the scene. Previous Point Cloud semantic segmentation methods lack the ability to fully capture the feature representation of the overlapping areas of different classes. This research has applied an aggregation strategy which processes features ...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Muhammad Moazam Fraz [DoC] Qaiser Riaz [DoC] Asif Ali [DoC]
Students
AMNA EJAZ [MSCS-8]

45 .
Non-Intrusive Load Monitoring (NILM) using a LSTM with socio-economic parameters:

Nonintrusive load monitoring (NILM) deconstructs aggregated electrical usage data into individual appliances. The dissemination of disaggregated data to customers raises consumer awareness and encourages them to save power, lowering 𝐶𝑂2 emissions to the environment. The performance of NILM systems has increased dramatically thanks to recent disaggregation methods. However, the capacity of these...


Project Level: PG
Project Type: Thesis Knowledge Area: Algorithms and Complexity
Year: 2019
Status: Completed
Advisor
Asad Waqar Malik [DoC]
Committee Members
Safdar Abbas Khan [DoC] Omar Arif [DoC] Arsalan Ahmad [DoC]
Students
AHMED TAIMOOR [MSCS-8]

46 .
Segmentation of Satellite Imagery using Deep Learning:

This thesis aims at designing a deep learning based algorithm for segmentation of water reservoirs from satellite imagery and performing various analytics tasks on the segmented data.The solution will assist Pakistan water management authorities in estimating the scarcity of water.


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Imran Malik [DoC]
Committee Members
Muhammad Shahzad [DoC] Muhammad Ali Tahir [DoC] Arsalan Ahmad [DoC]
Students
KHOLA NASEEM [MSCS-8]

47 .
Crop Type Classification Using Deep Learning:

Crop type mapping at the field level is necessary for a variety of applications in agricultural monitoring and food security. As remote sensing imagery continues to increase in spatial and temporal resolution, it is becoming an increasingly powerful raw input from which to create crop type maps. Still, automated crop type mapping remains constrained by a lack of field-level crop labels for traini...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Muhammad Moazam Fraz [DoC] Qaiser Riaz [DoC] Arsalan Ahmad [DoC]
Students
MUHAMMAD DANISH [MSCS-8]

48 .
Semantic Segmentation of 3D Point Clouds Using Deep Learning :

This thesis presents a new fine-grained analysis of large-scale point cloud semantic segmentation through a mechanism that fuses a local and global context-aware feature representation strategy into a semantic segmentation model. Real point cloud scenes can acquire intrinsic details of the object present in the scene, but due to the raw structure of point cloud data, some important details can be...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Muhammad Moazam Fraz [DoC] Qaiser Riaz [DoC] Asif Ali [DoC]
Students
RAO MUHAMMAD UMAIR WAHEED [MSCS-8]

49 .
PyraContrast- Unsupervised Pre-training on 3D Point Clouds:

Arguably, one of the deep learning’s greatest achievements is transfer learning. It is proven that if we use a rich source dataset to pre-train a network (ImageNet in 2D) and after we use a smaller target dataset to be fine-tuned, it can help to boost performance. It has been used in many applications including language and vision. In 3D scene understanding, there have been few works using this m...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Muhammad Moazam Fraz [DoC] Qaiser Riaz [DoC] Asif Ali [DoC]
Students
MUHAMMAD HASNAT MANZOOR [MSCS-8]

50 .
Multi Stage Quasi-labeled Vehicle Re-Identification Using Branched Convolutional Neural Networks:

Vehicle re-identification has become an important field of interest in computer vision, especially after the rise of importance of smart cities, and also with the evolution of Convolutional Neural Networks it has seemed possible to take advantage of technology to perform vehicle re-identification, Due to vehicles being an important object in the formation of Smart cities many scholars have applie...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Muhammad Imran Malik [DoC] Arsalan Ahmad [DoC] Muhammad Moazam Fraz [DoC]
Students
ALI HAIDER [MSCS-8]

51 .
Vehicle Re-Identification Using Shifted Window Transformer:

Vehicle re-identification has become a key area in computer vision which has important applications in areas like vehicle tracking, commodity tracking, security tracking, etc. But Vehicle re-identification inherits two challenges; there are large variations between the similar classes, two different vehicles may look the same . Various CNN-based Vehicle Re-ID have been proposed but CNN faces two ...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Muhammad Imran Malik [DoC] Arsalan Ahmad [DoC] Muhammad Moazam Fraz [DoC]
Students
FAISAL IMRAN [MSCS-8]

52 .
Semi-Supervised Vehicle Re-identification via Front & Rear Detection:

Vehicle reidentification is an important part of intelligent traffic monitoring systems, whose application doesn’t just limit to vehicle monitoring or surveillance. Having an efficient vehicle reidentification procedure allows a system to accurately and timely detect/track a vehicle, which can play an important part in doing forensic analysis as well. The procedure that will be based on deep neur...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Muhammad Imran Malik [DoC] Arsalan Ahmad [DoC] Muhammad Moazam Fraz [DoC]
Students
SHEHZAD AMJAD [MSCS-7]

53 .
Automated Signature Analysis & Verification:

Handwritten signature analysis is an active area of research where the current project would aim at solving the problem of extraction and verification of signatures from multifarious document images.


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Imran Malik [DoC]
Committee Members
Muhammad Shahzad [DoC] Faisal Shafait [DoC] Muhammad Ali Tahir [DoC]
Students
NOOR UL AIN IBRAHIM [MSIT-19]

54 .
Critical Analysis of SERU-ICT in Teacher Educational Programs at Pakistan.:

Technology is changing the everyday life of a common man. However, educational sector is still based on old traditional model of "chalk and talk".  Research studies found out that it is mainly due to teachers are not trained and ready to adopt digital technologies in their classrooms.  
This study will focus on analyzing the curriculum of teacher education program at Pakistan with respect to int...


Project Level: PG
Project Type: Thesis Knowledge Area: Human-Computer Interaction
Year: 2019
Status: Transferred to Non-Thesis
Advisor
Nasir Mahmood [EE]
Committee Members
Muhammad Shahzad [DoC] Muazzam A Khan Khattak [DoC] Maajid Maqbool [DoC]
Students
NAWAL ATHAR [MSITL-6]

55 .
Multiple Object tracking in the WILD:

Object tracking is one of the most important problems in the field of computer vision with applications ranging from surveillance and human-computer interactions to
medical imaging. Given the initial state (e.g., position and extent) of a target object in the first image, the goal of tracking is to estimate the states of the target in the subsequent frames. Although object tracking has been stud...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Transferred to Non-Thesis
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Arsalan Ahmad [DoC] Muhammad Muneeb Ullah [DoC]
Students
USMAN IFTIKHAR [MSCS-8]

56 .
Modeling and Forecasting of Electricity Generation from Hydro Power Plants:

Electricity usage planning is a main concern for electricity stakeholders in a country. The exhaustible resources are not enough to address energy demand in our country. It comes with varied problems such as price, environmental hazards and availability of resources. Renewable energy resources, mainly hydropower energy is an ideal solution to energy deficiency in Pakistan. However, to meet the co...


Project Level: PG
Project Type: Thesis Knowledge Area: Information Management
Year: 2019
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Arsalan Ahmad [DoC] Omar Arif [DoC]
Students
UMER JAVED [MSCS-8]

57 .
Integrated Science (IS) for Middle School Via Interactive Digital Book:

The scientific rules and laws of the natural world are very complex, wherein partial knowledge is not enough to comprehend the complications of the real world. Interdisciplinary learning is the best way to understand and perceive natural phenomena in the real world in science because scientific literacy is a central component of education. It is a great approach to teach 21st century skills with ...


Project Level: PG
Project Type: Thesis Knowledge Area: Human-Computer Interaction
Year: 2019
Status: Completed
Advisor
Nasir Mahmood [EE]
Committee Members
Muhammad Shahzad [DoC] Anjuman Shaheen [ECM] Erum Afzal [DoC]
Students
KANWAL IMTIAZ [MSITL-6]

58 .
Automated detection of weapons in surveillance data:

Between 2012 and 2014, the crime rate caused by firearms is very worrying in many parts of the world, especially in countries where their possession is legally allowed. The latest statistics from the United Nations Office on Drugs and Crime (UNODC) reveal that the rate of firearm killings per 100,000 population is 62.0 in Venezuela, 15.7 in Mexico, 3.9 in the United States and 0.6 in Spain. In ad...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Arsalan Ahmad [DoC] Muhammad Khuram Shahzad [DoC]
Students
TUFAIL HASHMI [MSCS-8]

59 .
Application of AR for Improving Spatial Thinking Skills in Primary Level Mathematics:

Spatial thinking is an ability to mentally visualize and manipulate 3D objects, identify patterns, and to describe relationships among objects in a setting. These skills are highly predictive for success in engineering, architecture, graphic design, computer science, and
other fields of education.
In our education system at school’s level teachers do not focus on development of above mentioned...


Project Level: PG
Project Type: Thesis Knowledge Area: Graphics and Visualization
Year: 2019
Status: Completed
Advisor
Tahira Lashari [DoC]
Committee Members
Muhammad Shahzad [DoC] Nasir Mahmood [EE] Anjuman Shaheen [ECM]
Students
TASHFA AFZAL [MSITL-6]

60 .
Application of VR Based Approach to Help Secondary Students in Performing Practical Activities:

Smart devices are providing various real-time services due to the advancement in Internet of things. If we combine IOT with virtual reality (VR) it will provide experience of real world by using virtual environment. One of the main advantages of combining VR with IOT is to fill up the gap between real and virtual world. The education sector is expected to have significant change by using these te...


Project Level: PG
Project Type: Thesis Knowledge Area: Human-Computer Interaction
Year: 2019
Status: Completed
Advisor
Nasir Mahmood [EE]
Committee Members
Muhammad Shahzad [DoC] Muazzam A Khan Khattak [DoC] Umer Nasim [DoC]
Students
MAHAM SAJJAD [MSITL-6]

61 .
3D Building Reconstruction Using 3D Point Clouds:

3D Building Reconstruction Using 3D Point Clouds


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Active
Advisor
Muhammad Shahzad [DoC]
Committee Members
Muhammad Moazam Fraz [DoC] Muhammad Imran Malik [DoC] Arsalan Ahmad [DoC]
Students
NAVEED GHAFFAR [MSCS-8]

62 .
Text Summarization from Judicial Records using Deep Neural Machines:

The judiciary is the branch of the government whose task is the administration of justice. The courts are generating a large amount of data as legal proceedings. The legal documents are in the form of cases and their judgments. A judgment is a long, and detailed document. In order to prepare for a case, a lawyer has to read through hundreds of legal documents to find out the relevant judgments. I...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Faisal Shafait [DoC]
Committee Members
Adnan Ul-Hasan [ECM] Seemab Latif [DoC] Rabia Irfan [DoC]
Students
AYESHA SARWAR [MSCS-8]

63 .
CBR-based Similar Case Retrieval from Judicial Records:

A system that can retrieve similar cases will not only lessens human effort but also reduce the amount of time it would take to retrieve similar cases.
The techniques in Case Based Reasoning (CBR) can be used to index, find, retrieve, and rank the cases that are most similar to a newly registered case. The dataset used will be the judgments from the courts of Pakistan.


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Faisal Shafait [DoC]
Committee Members
Adnan Ul-Hasan [ECM] Seemab Latif [DoC] Rabia Irfan [DoC]
Students
MEHREEN GUL [MSCS-8]

64 .
Information Extraction from Court Room Records:

Information extraction is one of the most difficult tasks when it comes to retrieving information especially from semi-structured documents. But on the other side, it is most useful when information is required for indexing and analyzing the data. The judgments found in the different courts of Pakistan do not follow a standard format overall rather many different types of formats have been found....


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Faisal Shafait [DoC]
Committee Members
Adnan Ul-Hasan [ECM] Seemab Latif [DoC] Rabia Irfan [DoC]
Students
NIDA AHMED [MSCS-8]

65 .
STEMIFIED MATHEMATICS: Measuring the engagement and learning outcome of primary students using math focused STEM projects on scratch.:

In this study, we measure the improvement in the student’s engagement and academic achievement by using stemified activities along with projects on scratch software in mathematics. There is a dire need of new methodologies for students’ understanding of the mathematics. To achieve this goal, this study was set out with the aim of developing interactive videos on scratch software: Math in STEM as ...


Project Level: PG
Project Type: Thesis Knowledge Area: Social Issues and Professional Practice
Year: 2019
Status: Completed
Advisor
Safdar Abbas Khan [DoC]
Committee Members
Muhammad Shahzad [DoC] Adnan Aslam [BS] Tahira Lashari [DoC]
Students
TAYYABA NOREEN [MSITL-6]

66 .
Epilepsy disease detection using Deep learning Techniques :

Epilepsy is a neurological disorder in which brain activities becomes abnormal causing uncontrollable jerking movements of legs and arm, temporary confusion and loss of awareness. Electroencephalogram (EEG) is used to read brain’s behavior by doctors but according to some research EEG fails to detect the epilepsy disease in epileptic patient because it happens due to a sudden generation of abnorm...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Qaiser Riaz [DoC] Muhammad Moazam Fraz [DoC] Arsalan Ahmad [DoC]
Students
QURAT UL AIN [MSIT-17]

67 .
Wheat Crop Health Monitoring Using Deep Learning on Satellite Images:

There are several advantages of remote sensing in the field of agriculture. Each year Pakistan loses tons of its crop production due to poor farming practices. WSNs and Low altitude imagery adds up the costs as the area of field grows. On the contrary, remote sensing helps in timely monitoring and provide an accurate picture of the agricultural sector with high accuracy and high revisit frequency...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Transferred to Non-Thesis
Advisor
Rafia Mumtaz [DoC]
Committee Members
Muhammad Shahzad [DoC] Asad Ali Shah [DoC] Sana Khalique [DoC]
Students
USAMA MASOOD [MSCS-8]

68 .
An intelligent framework for support ticket resolution actions recommendation:

Nowadays, customer support systems are one of the key factors in maintaining any big company's reputation and success. These systems are capable of handling a large number of tickets systemically and provides a mechanism to track/logs the communication between customer and support agents. Companies invest huge amounts of money in training support agents and deploying customer care services for th...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Arsalan Ahmad [DoC] Sharifullah Khan TI [DoC]
Students
SYED ALI ZAIDI [MSCS-8]

69 .
Liveness Detection:

Know Your Customer (KYC) is a vast field potentially sought after every business in the category of B2C. Knowing the customers induces confidence in dealing with the customer and ensures minimal fraud. Although, with current situations of global pandemic and general technological movement, both companies and customers seek an easy way out of verification. This project will potentially create a wh...


Project Level: PG
Project Type: Thesis Knowledge Area: DSSP
Year: 2019
Status: Completed
Advisor
Safdar Abbas Khan [DoC]
Committee Members
Iftikhar Ahmad Rana [EE] Muhammad Khuram Shahzad [DoC] Ibrar Hussain [BS]
Students
TAUSEEF MEMON [MSCS-8]

70 .
Natural Image Synthesis Using Scene Graphs:

A scene graph is a general data structure commonly used by vector-based graphics editing applications and modern computer games, which arranges the logical and often spatial representation of a graphical scene.

A scene graph is a collection of nodes in a graph or tree structure. A tree node may have many children but only a single parent, with the effect of a parent applied to all its child no...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Arsalan Ahmad [DoC] Anis Ur Rahman [DoC]
Students
MAHGUL MALIK [MSCS-8]

71 .
AI based Dengue Fever Prediction using Deep Learning:

Dengue epidemic is a major public threat in Pakistan since 2005, following millions of people at risk. To account for the human suffering caused by the epidemics of dengue, the project aims to develop a system that provides a cost-effective and rapid solution for diagnostic screening of dengue. Artificial Intelligence powered system accompanied by a medical chatbot, providing an easy-to-interact ...


Project Level: PG
Project Type: Thesis Knowledge Area: DSSP
Year: 2019
Status: Completed
Advisor
Faisal Shafait [DoC]
Committee Members
Muhammad Shahzad [DoC] Adnan Ul-Hasan [ECM] Muhammad Imran Malik [DoC]
Students
MAHNOOR HANEEF [MSEE-9]

72 .
Orientation Aware Weapons Detection in Visual Data:

Automatic detection of weapons is significant for improving security and well being of individuals, nonetheless, it is a difficult task due to large variety of size, shape and appearance of weapons. View point variations and occlusion also are reasons which makes this task more difficult. Horizontal object detection methods have greater background information along with object thus it will have l...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Arsalan Ahmad [DoC] Muhammad Khuram Shahzad [DoC]
Students
NAZEEF UL HAQ [MSCS-8]

73 .
Crop Monitoring Via Remote Sensing:

The increasing world population is generating higher food demand that needs effective cultivation methodologies to meet them, especially for the developing countries that have a higher dependence on the agricultural sector. Crop type classification is part of crop monitoring which can help to plan crops effectively and meet the demandsupply chain. Our study had mainly two objectives, acquisition ...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Qaiser Riaz [DoC] Rafia Mumtaz [DoC] Faisal Shafait [DoC]
Students
ZUBAIR IQBAL [MSCS-7]

74 .
Applications of Low Dimensional Manifold Method in MRI and Depth Image Reconstruction:

Recently (Osher et al. 2017) proposed a low dimensional manifold model (LDMM) for image inpainting and denoising. The basic assumption of LDMM is that the image patches sample a manifold whose intrinsic dimensions are much lower than the high dimensional ambient space. Based on this assumption, the image is recovered by keeping the dimension of the patch-manifold as small as possible. LDMM has b...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Omar Arif [DoC]
Committee Members
Muhammad Shahzad [DoC] Muhammad Moazam Fraz [DoC] Faisal Shafait [DoC]
Students
SAIM ABDULLAH [MSCS-8]

75 .
Multimodal Knowledge Reasoning for Enhanced Visual Question Answering:

Visual Question Answering (VQA) is an Artificial Intelligence task in the field of Natural Language Processing (NLP) and Computer Vision (CV) in which an image and a question are given to a VQA system which is responsible to give the answer a given question can be open-ended or multiple-choice. The VQA system is used for a variety of realworld applications, such as providing situational informati...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Muhammad Khuram Shahzad [DoC] Arsalan Ahmad [DoC]
Students
AFZAAL HUSSAIN [MSCS-8]

76 .
Deep Multi-Labelling based Unsupervised Person Re-identification:

Person re-identification is the problem of finding the given person in the non-overlapping cameras at a given time or finding the person that has appeared in the same camera but at different time stamps. The re-identification problem plays vital role in the field of visual surveillance. In real world scenario, the problem is not as easy as it seems because of the various obstacles including but n...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Qaiser Riaz [DoC] Muhammad Moazam Fraz [DoC] Muhammad Imran Malik [DoC]
Students
ZARMEENA HASSAN [MSCS-8]

77 .
Recursive Skip Attention Network with Mask Refinement for Nuclei Instance Segmentation and Classification in Histology Images:

Automatic Nuclei instance segmentation and classification within Hematoxylin \& Eosin stained images can be considered a preliminary step in developing advanced medical systems which can provide diagnosis and prognosis to deadly diseases like cancer. However, the heterogeneous and crowding nature of the cells (with large inter and intra class variability), combined with the fact that there are cu...


Project Level: PG
Project Type: Thesis Knowledge Area: Computational Science
Year: 2019
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Qaiser Riaz [DoC] Arsalan Ahmad [DoC]
Students
NIMRA AMIN [MSCS-8]

78 .
Informative Attention Based Deep Network for Person Re-identification:

Person re-identification (ReID) is the most essential task in visual based surveillance. The recent success of deep learning in extracting unique and powerful discriminatory features has made immense shift in this era. The existing limitations in major challenges of person re-identification like pose variations, body misalignment, patch based learning are not comprehensively addressed. We propose...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Muhammad Khuram Shahzad [DoC] Arsalan Ahmad [DoC]
Students
RAO FAIZAN [MSCS-8]

79 .
Distance Regression and Classification with Attention Network for Nuclei Instance Segmentation:

Nuclei instance segmentation and classification in histology plays a major role in routine pathology image examination, which enable morphological features analysis that further facilitates streamlined diagnosis and prognosis quantification. However, accurate nuclei instance segmentation and classification remains a challenging task. The nuclei in the tissue images obtained from different human o...


Project Level: PG
Project Type: Thesis Knowledge Area: Computational Science
Year: 2019
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Qaiser Riaz [DoC] Arsalan Ahmad [DoC]
Students
GHULAM DOGAR [MSCS-8]

80 .
Detection of Student behavior in Classroom:

This research aims for scrutinization of students in a classroom environment and can be used as an alternate for monitoring an attention of students in real-time. Detection of attentiveness and not attentiveness of students studying in school, collages and university could help to identify the weaknesses and strengths of any educational institute. Human factor is the main and the most critical pa...


Project Level: PG
Project Type: Thesis Knowledge Area: DSSP
Year: 2019
Status: Completed
Advisor
Faisal Shafait [DoC]
Committee Members
Muhammad Shahzad [DoC] Adnan Ul-Hasan [ECM] Muhammad Imran Malik [DoC]
Students
NAQASH GERARD [MSITL-6]

81 .
Large Scale Fine-Grained Visual Recognition:

Large scale fine-grained visual recognition is a very challenging area as models must find discriminative features between sub-categories of the objects. The use cases include identifying animal breeds, car models or hair colors in stores. The current best methods include deep learning-based methods which are complex, resource-intensive and slow to train. The aim of this research is to use teach...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Muhammad Imran Malik [DoC] Adnan Ul-Hasan [ECM] Faisal Shafait [DoC]
Students
OMER USMAN UPPAL [MSCS-7]

82 .
Vehicular Traffic Flow Parameter Estimation Using Deep Neural Networks:

To overcome the challenges regarding the fixed place camera installation we introduce a drone-based real-time traffic flow parameter estimation. With the recent convenience of UAV (Unmanned aerial vehicles) videos begin a new smart transportation application. We Captured the live vehicles traffic data using DJI Phantom 4 Pro drone from 8 AM - 6 PM. We manually annotated the vehicle instances of a...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Faisal Shafait [DoC] Muhammad Moazam Fraz [DoC] Muhammad Imran Malik [DoC]
Students
MUHAMMAD RAEES [MSCS-7]

83 .
Deep learning based segmentation of multiple tissue structures in histology Images for predictive cancer analytics:

According to a survey of the Global Health Metrics, Cancer is the reason behind every sixth death, making it the second leading lethal disease around the world. This immensely growing issue has dragged much attention of researchers and pathologists over the last few years. The analysis of nerves and vessels density within affected histopathology images is an important step in determining the rate...


Project Level: PG
Project Type: Thesis Knowledge Area: Computational Science
Year: 2019
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Qaiser Riaz [DoC] Arsalan Ahmad [DoC]
Students
AFIA RASOOL [MSCS-8]

84 .
Estimation of QoT using machine learning techniques in optical networks:

With the increase in global IP traffic it is becoming important in optical networks to estimate the Quality of Transmission (QoT) of new lightpaths before their establishment. The traditional approaches used to estimate the QoT of new lightpaths are coupled with high margins which leads to inaccurate QoT predictions. Accurate prediction of QoT is significant so that network can work efficiently, ...


Project Level: PG
Project Type: Thesis Knowledge Area: Networking and Communications
Year: 2019
Status: Completed
Advisor
Arsalan Ahmad [DoC]
Committee Members
Muhammad Shahzad [DoC] Salman Abdul Ghafoor [EE] Syed Taha Ali [EE]
Students
MAHNOOR KHAN [MSEE-9]

85 .
Quality of transmission prediction using machine learning models in future optical networks:

With the increase in global IP traffic in core optical networks, the accurate estimation of the Quality of Transmission (QoT) of new lightpaths before their establishment has become an important issue. The traditional approaches used to estimate the QoT of new lightpaths are coupled with high margins which leads to inaccurate QoT predictions. Accurate prediction of QoT is significant so that netw...


Project Level: PG
Project Type: Thesis Knowledge Area: Networking and Communications
Year: 2019
Status: Completed
Advisor
Arsalan Ahmad [DoC]
Committee Members
Muhammad Shahzad [DoC] Salman Abdul Ghafoor [EE] Syed Taha Ali [EE]
Students
MEHEK SIDDIQUI [MSEE-9]

86 .
Deep Learning for Forensic Examination of Handwritten Text:

Forensic Handwriting Examiners (FHEs) often have the scenarios where they are provided with a piece of text handwritten by an author in a different style than those texts which are present in the dataset and were written by the same author. This is a usual situation when the same person, for example, writes threat letters to more than one persons and tries to change her/his handwriting style ever...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Withdraw
Advisor
Muhammad Imran Malik [DoC]
Committee Members
Muhammad Shahzad [DoC] Faisal Shafait [DoC] Muhammad Ali Tahir [DoC]
Students
MUHAMMAD MOHSIN [MSCS-8]

87 .
Abstractive Text Summarization of Judicial Cases in Supreme Court of Pakistan:

Due to the huge amount of legal information availability on the internet, as well as other sources, it is important for the research community to do more extensive research on the area of legal text processing, which can help us make sense out of the vast amount of available data. This information growth has compelled the requirement to develop systems that can help legal professionals as well as...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Faisal Shafait [DoC]
Committee Members
Muhammad Shahzad [DoC] Adnan Ul-Hasan [ECM] Muhammad Imran Malik [DoC]
Students
MUNEEB AHMED ANWAR [MSIT-18]

88 .
Analyzing Emergency Evacuation Strategies For Large Buildings Using Crowd Simulation Framework:

An emergency evacuation is defined as the immediate act of egress or escaping of pedestrians from an open public gathering place or a multistoried building containing a man-made or natural threat (Jim Burtles 2016). Evacuation plans are devised beforehand and are implemented using different techniques. Pakistan is ranked amongst the top 10 in the list of countries with the highest number of death...


Project Level: PG
Project Type: Thesis Knowledge Area: Software Engineering
Year: 2019
Status: Completed
Advisor
Imran Mahmood Hashmi [DoC]
Committee Members
Muhammad Shahzad [DoC] Sidra Sultana [DoC] Muhammad Moazam Fraz [DoC]
Students
TALHA NADEEM [MSIT-18]

89 .
Detailed Image Description Using Human Apparels:

Image captioning is a process that deals with the generation of captions of an image in human understandable language .A lot of work has been already done in this field using various techniques like neural talk ,show and tell ,show, attend and tell and various models have been proposed to make the captioning better than before .In this work, a new model has been proposed that will generate the de...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Arsalan Ahmad [DoC] Qaiser Riaz [DoC] Muhammad Moazam Fraz [DoC]
Students
MADIHA SAMAD [MSIT-18]

90 .
An Efficient Framework for GPU Virtualization using API Remoting:

The aim of this research is to provide a GPU Virtualization framework using API remoting technique for CUDA (Compute Unified Device Architecture) applications. CUDA is a parallel computing platform and programming model developed by Nvidia for general purpose computing on NVIDIA GPUs (Graphics Processing Units). CUDA enables developers to speed up compute-intensive applications by harnessing the ...


Project Level: PG
Project Type: Thesis Knowledge Area: Parallel and Distributed Computing
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Mian Muhammad Hamayun [DoC] Muhammad Muneeb Ullah [DoC] Anis ur Rahman [DoC]
Students
ALI SHEHARYAR [MSCS-7]

91 .
Person Identification through Analysis of Gait Using Deep Learning and Genetic Programming:

In the recent years, with burgeoning of terrorist activities, measures have been taken to curb or
prevent/minimize them by designing stringent screening systems. One of the many deployed systems
involves authentication or identification of a person through his biometrics. Biometric classification
comes under two categories; physiological and behavioral.
Physiological biometrics such as face r...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Imran Malik [DoC]
Committee Members
Muhammad Shahzad [DoC] Gul Muhammad Khan [ECM] Khawar Khurshid [EE]
Students
SABA GUL [MSEE-9]

92 .
One-Shot Vehicle Re-Identification:

Vehicle reidentification is an important part of intelligent traffic monitoring systems, whose application doesn’t just limit to vehicle monitoring or surveillance. Having an efficient vehicle reidentification procedure allows a system to accurately and timely detect/track a vehicle, which can play an important part in doing forensic analysis as well.
The procedure that will be used is Siamese N...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Faisal Shafait [DoC] Muhammad Moazam Fraz [DoC] Muhammad Imran Malik [DoC]
Students
HASSAN FAROOQ [MSCS-7]

93 .
Adversarial Attacks on Deep Neural Networks:

Deep neural networks are increasingly being used for classification tasks. Image classfication is the area using DNN extensively. These classification models are deployed in real time and laboratory environments. In some applications these networks are vital to the security of access points using these networks. In order to check the robustness of the networks, adversarial attacks are being carri...


Project Level: PG
Project Type: Thesis Knowledge Area: DSSP
Year: 2019
Status: Completed
Advisor
Faisal Shafait [DoC]
Committee Members
Muhammad Shahzad [DoC] Adnan Ul-Hasan [ECM] Muhammad Shahzad [EE]
Students
SALAH UD DIN [MSEE-9]

94 .
Crop classification though low altitude remote sensing platforms:

Remote sensing has been found to be a valuable tool in evaluating, monitoring and management of land, water and crop resources. Satellite imagery and aerial imagery have wide applications in the field of agriculture, monitoring snow cover trends, wild fire trends, water level rising and forestry. In agriculture, crop health monitoring, yield estimation, classification of crops based on land cover...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Rafia Mumtaz [DoC]
Committee Members
Muhammad Shahzad [DoC] Safdar Abbas Khan [DoC] Faisal Shafait [DoC]
Students
NAVEED IQBAL [MSIT-18]

95 .
Deep Learning based Supervised Classification of Water Bodies from High-resolution Multispectral Sentinel-2 Imagery:

The rapid growth of technology in the fields of remote sensing, image acquisition and data storage has led to an enormous increase in volume of available high-resolution satellite image data. One of the most challenging tasks faced by scientists these days, is the efficient and precise use of satellite images for earth observation. Due to its high resolution and multispectral characteristics, sat...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Muhammad Imran Malik [DoC] Adnan Ul-Hasan [ECM] Faisal Shafait [DoC]
Students
RAFIA MALIK [MSCS-7]

96 .
How do people look at crowds? A psychophysics perspective :

Crowds of people offer multitudes of stimuli for attracting gaze and attention. It is a fact that during normal social interactions, faces are the most important stimuli. We are interested in exploring if this statement holds in conflation with crowd density (the number of people per square area). To this end, we would like to understand this relationship between human gaze and varying crowd dens...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Omar Arif [DoC]
Committee Members
Anis ur Rahman [DoC] Faisal Shafait [DoC] Muhammad Muneeb Ullah [DoC]
Students
SOBAS MEHBOOB [MSCS-7]

97 .
Predictive visual saliency model for crowd videos :

What do people look at first, instantaneously and inadvertently? The answer would be: They look at what is salient. Visual Saliency is an ongoing area of research interest to figure out what captures bottom-up attention. This is an important question because visual saliency provides the link between the perception of external stimuli and human cognition. Understanding Visual Saliency means to und...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Omar Arif [DoC]
Committee Members
Anis ur Rahman [DoC] Faisal Shafait [DoC] Muhammad Muneeb Ullah [DoC]
Students
MEMOONA TAHIRA [MSCS-7]

98 .
Pansharpening of Multispectral images using dual attention based two stream network:

With the advent of technology there are large number of earths observing satellites having sensors, delivering different type of images that can provide a varying useful information about different surfaces of earth and those areas that are not reachable by human. Most popular of these images are panchromatic (PAN) image having the geometric details and multispectral (MS) image that contains refl...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Anis ur Rahman [DoC] Faisal Shafait [DoC] Muhammad Muneeb Ullah [DoC]
Students
SYEDA ROSHANA ALI NAQVI [MSCS-7]

99 .
Scene Understanding for Map Compression:

In Robotics, Map plays an important role for the navigation and localization of Robot. Current 3D maps are highly accurate. Google Street View is one such example of a rich 3D Map. Can robots use such a map with limited storage capability? Perhaps not.
In this thesis, We have implemented the novel approach to compress a large scale vision based map by implementing Convolution neural networks for...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Wajahat Hussain [EE]
Committee Members
Muhammad Shahzad [DoC] Omar Arif [DoC] Muhammad Latif Anjum [EE]
Students
Ahmad Faraz Khan [MSEE-8]

100 .
Applying different techniques of localization on drone imagery related to agriculture:

Pakistan's major economy depends on agriculture, but we have seen many problems with automation and technology due to a lack of information and lost a lot of crops. In order to get high yield and more production along with the savage of extra labor time and money, we need automation in this era. Automated and high-resolution imagery (temporal and spatial) acquired from low-cost drones provides an...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Wajahat Hussain [EE]
Committee Members
Muhammad Shahzad [DoC] Muhammad Shahzad [EE] Muhammad Latif Anjum [EE]
Students
IQRA AKRAM [MSEE-9]

101 .
Optical Music Recognition using Deep Learning:

Optical Music Recognition is a field of research that investigates how to computationally
decode music notation from images. Despite the efforts made so far, there are hardly any end to end complete solutions to the problem which is dealing with a comprehensive set of notation symbols. Using a neural model that combines the capabilities of CNN, which work on the input image, and RNN, which deal ...


Project Level: PG
Project Type: Thesis Knowledge Area: Computational Science
Year: 2018
Status: Completed
Advisor
Faisal Shafait [DoC]
Committee Members
Muhammad Shahzad [DoC] Adnan Ul-Hasan [ECM] Muhammad Imran Malik [DoC]
Students
MUHAMMAD FAROOQ [MSCS-6]

102 .
Spatio-temporal detection of human actions in video:

Automated analysis of human actions in video is an active research area in computer vision, having practical applications in surveillance, video indexing, and entertainment, etc. In this project, we aim to develop a learning algorithm which can localize different human actions both in space and time in videos. A potential application could be the localization of robber(s) in CCTV videos.


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2018
Status: Active
Advisor
Muhammad Muneeb ULLAH [DoC]
Committee Members
Muhammad Shahzad [DoC] Omar Arif [DoC] Wajahat Hussain [EE]
Students
MUHAMMAD UZAIR AZHAR [MSCS-7]

103 .
Deep learning based person re-identification:

Surveillance cameras have been widely used in different scenes. Accordingly, a demanding need is to recognize a person under different cameras, which is called person re-identification. This problem can be defined from two aspects that is matching of the same person across multiple non-overlapping cameras and matching of a person along the same camera but with different point of time.


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Asad Waqar Malik [DoC] Qaiser Riaz [DoC] Muhammad Moazam Fraz [DoC]
Students
IMAD GOHAR [MSCS-6]

104 .
Smart Surveillance through Periocular Biometrics:

Face recognition has paved its way to success in the recent years and has been intelligently merged with modern technologies. Nowadays, face recognition techniques are extensively used by law enforcement agencies worldwide. In our country, there is a significant need of a system which forms an integrated unit for security surveillance by identifying and recognizing persons through their facial ch...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Muhammad Imran Malik [DoC]
Committee Members
Muhammad Shahzad [DoC] Muhammad Ali Tahir [DoC] Faisal Shafait [DoC]
Students
MUHAMMAD EHSAN IRSHAD [MSCS-6]

105 .
Context Aware Real Time Task Distribution Strategies for Mobile Devices:

Due to advancement in technology, many routine applications can be executed on mobiles. However, energy is still the main concern for compute intensive applications. Therefore, to overcome this issue, we proposed peer to peer management framework for real-time task distribution over mobile devices. In this thesis, we will discuss two different techniques to offload computation task i.e. i) Centra...


Project Level: PG
Project Type: Thesis Knowledge Area: Parallel and Distributed Computing
Year: 2018
Status: Active
Advisor
Muhammad Shahzad [DoC]
Committee Members
Asad Waqar Malik [DoC] Sohail Iqbal [DoC] Muazzam A Khan Khattak [DoC]
Students
MUHAMMAD NAEEM QAYYUM [MSIT-17]

106 .
Data on Wheels: A Machine Learning Framework for Large-Scale Data Transfer:

The vehicular network has become the emerging paradigm for data dissemination due to its dynamic topology and predictable mobility. It supports intercommunication between data centers, as well as data aggregation from different data sources by allowing vehicles to communicate with each other and with the infrastructure. The main problem in big data dissemination is the aggregate energy consumptio...


Project Level: PG
Project Type: Thesis Knowledge Area: Parallel and Distributed Computing
Year: 2018
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Asad Waqar Malik [DoC] Imran Mahmood [DoC] Asad Ali Shah [DoC]
Students
MARIA KANWAL [MSIT-17]

107 .
Object Recognition and Segmentation Using Unstructured 3D Point Clouds:

Point Clouds provide an important type of scalable geometric data structure representation which is used for numerous applications of computer graphics. Point clouds are in unstructured form, due to irregularities in structure the shapes are not organized in a pre-defined manner and any change in their ordering does not affect the spatial distribution of points. The only usage of point clouds was...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Muhammad Moazam Fraz [DoC] Safdar Abbas Khan [DoC] Arsalan Ahmad [DoC]
Students
MEHAK SHEIKH [MSCS-6]

108 .
Deep Learning Based Segmentation of 3D Point Clouds:

3D point clouds are becoming increasingly popular to represent the 3D real world. It consists of an ordered set of 3D data points in a particular coordinate system and typically used to represent the exterior surface of the object of interest in a discretized/sampled form. A 3D point cloud is very helpful in understanding of an illuminated scene as it provides the necessary information with an ad...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Muhammad Moazam Fraz [DoC] Safdar Abbas Khan [DoC] Arsalan Ahmad [DoC]
Students
SABA MEHMOOD [MSCS-6]

109 .
Segmentation and Classification of 3D Point Clouds Using Deep Pointwise Residual Networks:

Deep learning in computer vision domain plays an important role in the last few years. Its performance surpasses previous approaches based on hand-crafted features. Deep learning approaches yield better performance in the scene understanding tasks. Scene understanding tasks have various applications e.g., autonomous driving, medical diagnosis and robot navigation for the indoor and outdoor enviro...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Muhammad Moazam Fraz [DoC] Safdar Abbas Khan [DoC] Arsalan Ahmad [DoC]
Students
SAIRA ARSHAD [MSCS-6]

110 .
Estimation of ground contacts from human gait by a wearable Inertial Measurement Unit:

IMU based ground contacts estimation is an active area of research in robotics, character animation, and smart healthcare. The basic idea of this thesis is to analyze the inertial data of human gait and to estimate both stance and swing phases of each foot along with other interesting features such as stride length, heel touch, toe touch, foot touch etc.

Important Literature:

https://www.s...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Qaiser Riaz [DoC]
Committee Members
Muhammad Shahzad [DoC] Hasan Tahir [DoC] Muhammad Zeeshan [DoC]
Students
MUHAMMAD JUNAID UMER [MSCS-6]

111 .
Vehicle Detection in Aerial Imagery:

Aerial surveillance is getting more attention with the increasing use of small drones for different social and military tasks, as a result more aerial data is being produced. Various object and pattern recognition tasks require this data to process, analyze and investigate for effective decision making. Vehicles are most prominent and important objects when observed from an aerial view and identi...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Faisal Shafait [DoC] Adnan Ul-Hasan [ECM] Muhammad Imran Malik [DoC]
Students
MUHAMMAD HAROON [MSCS-6]

112 .
VTALK – Smart Vehicles Through Inter and Intra Vehicular Communication:

Vehicle networks or vehicle-to-vehicle communication (V2V communication) is the wireless communication between vehicles either directly (peer-to-peer) or in-directly (vehicle-to-server-to-vehicle) forming a network of vehicles themselves. We assume all modern vehicles have on board sensors and computers and data is gathered via these sensors.

Data gathered can be transmitted and received to/fr...


Project Level: PG
Project Type: Thesis Knowledge Area: Software Engineering
Year: 2018
Status: Completed
Advisor
Hasan Arshad Nasir [EE]
Committee Members
Muhammad Shahzad [DoC] Omar Arif [DoC] Arsalan Ahmad [DoC]
Students
OMER RASHID [MSCS-6]

113 .
An HLA-based framework for simulation of geographically distributed Data centers:

Usually, the existing simulators provide limited scalability. Therefore, researchers can simulate their protocols and algorithms to a certain extent. In cloud simulators such as CloudNetSim++ and CloudSim are two most commonly used simulators are designed for geographically distributed data centers; however, the simulators cannot support execution on distributed systems. Thus, scalability is also...


Project Level: PG
Project Type: Thesis Knowledge Area: Parallel and Distributed Computing
Year: 2018
Status: Completed
Advisor
Muazzam A Khan Khattak [DoC]
Committee Members
Muhammad Shahzad [DoC] Asad Waqar Malik [DoC] Safdar Abbas Khan [DoC]
Students
BUKHTAWAR ELAHI [MSIT-17]

114 .
An intelligent context-aware recommendation system for real-estate:

Data science is being extensively used to recommend suitable products to end users and consumers. This thesis will focus on understanding a user requirements based on his browsing history as well as interaction with the website through a chatbot and recommend the products that the user will be interested in. The primary use case for this work will be in the domain of real-estate and the system wi...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Faisal Shafait [DoC]
Committee Members
Muhammad Shahzad [DoC] Adnan Ul-Hasan [ECM] Muhammad Imran Malik [DoC]
Students
Faiza Rehman [MSIT-16]

115 .
Gait based emotion recognition using inertial data :

Emotion recognition is an active area of research in the domain of human computer interaction. Computers can interact better with humans if they can explicitly understand human emotions. Most of the existing state of the art methodologies rely heavily on facial expression analysis which has its own drawbacks. In this research work, we present a set of hand-crafted features computed from the inert...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Qaiser Riaz [DoC]
Committee Members
Muhammad Shahzad [DoC] Hasan Tahir [DoC] Safdar Abbas Khan [DoC]
Students
MUHAMMAD ARSLAN HASHMI [MSCS-6]

116 .
Person Re-Identification based Visual Surveillance System:

Large networks of cameras are increasingly deployed in public places like airports, railway stations, college campuses and office buildings. These cameras typically span large geospatial areas and have non-overlapping fields-of-views (FOVs) to provide enhanced coverage. Such networks provide huge amounts of video data, which is either manually monitored by law enforcement officers or utilized aft...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Rabia Irfan [DoC] Arsalan Ahmad [DoC] Muhammad Muneeb Ullah [DoC]
Students
JANNAT KHATOON [MSCS-5]

117 .
Energy-aware Real-time Route Prediction Framework for Embedded Devices:

Internet of things (IOT) is becoming an important part of our lives as it integrates computing devices embedded in everyday objects and Vehicle of things is a major part of IOT. This research is focused on vehicular communication for optimal route prediction in developing countries like Pakistan where no specialized road sensors are available. Vehicular ad hoc network (VANET) is created based on ...


Project Level: PG
Project Type: Thesis Knowledge Area: Parallel and Distributed Computing
Year: 2018
Status: Completed
Advisor
Asad Waqar Malik [DoC]
Committee Members
Muhammad Shahzad [DoC] Anis ur Rahman [DoC] Omar Arif [DoC]
Students
AYESHA ZAHIR [MSCS-6]

118 .
Multi-Dimensional Challenges in Geo, Social and Economic Reachability: A Step towards Sustainable Development Goals:

Internet in Pakistan is becoming an area of exponential growth. The global rise in IT infrastructure demands efficient, fast, high-speed internet while offering alternative innovative technologies. In this skyrocket era of rapid development, the most trending communication medium internet has been mounting than before. With the introduction of 3G & 4G technology, it eliminates multi-dimensional c...


Project Level: PG
Project Type: Thesis Knowledge Area: Social Issues and Professional Practice
Year: 2018
Status: Completed
Advisor
Asad Ali Shah [DoC]
Committee Members
Muhammad Shahzad [DoC] Muhammad Muneeb ULLAH [DoC] Arsalan Ahmed [DoC]
Students
Asma Pirzada [MSIT-16]

119 .
Deep Learning based Classification of Retinal Vasculature in to Arteries and Veins:

Retinal imaging has rapidly grown within ophthalmology in the past twenty years. The availability of low cost fundus cameras to take direct images of the retina, fundus photography, makes it possible to examine the eye for the presence of many different eye diseases with a simple, non-invasive method.

The project aims to apply deep learning techniques to segment normal retinal anatomical stru...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Muhammad Moazam Fraz [DoC] Anis ur Rahman [DoC] Asad Waqar Malik [DoC]
Students
FARAZ AHMAD BHATTI [MSCS-6]

120 .
Inertial based terrain classification:

Ground surface or terrain classification is an active area of research in robotics and autonomous off-road vehicle driving. However, this problem has barely been studied with the inertial data of human gait. In this thesis, we propose a novel method of estimation terrain has been performed on which a person has walk. For this purpose we have recorded the gait patterns with inertial sensors of nor...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Qaiser Riaz [DoC]
Committee Members
Muhammad Shahzad [DoC] Muhammad Ali Tahir [DoC] Safdar Abbas Khan [DoC]
Students
MUHAMMAD ZEESHAN-UL-HASNAIN HASHMI [MSCS-6]

121 .
Information Extraction from Tables in Document Images using Deep Learning:

Table structure extraction is a challenging problem due to the diversity of layouts in which tabular information is presented. This problem is further complicated when a printed invoice is captured by a mobile phone camera, since the capture induces perspective and page curl distortions that result in distorting the alignment in table columns. This thesis aims at addressing these issues in a gene...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Faisal Shafait [DoC]
Committee Members
Muhammad Shahzad [DoC] Adnan Ul-Hasan [ECM] Muhammad Imran Malik [DoC]
Students
Muhammad Shahzad [MSCS-5]

122 .
Scene Understanding: Image Description in Natural Language:

Automatic description generation from natural images is a challenging problem that
has recently received a large amount of interest from the computer vision and natural language processing communities. The ability to associate images with natural language sentences that describe what is depicted in them is a hallmark of image understanding, and a prerequisite for applications such as sentence-ba...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Asad Waqar Malik [DoC] Omar Arif [DoC]
Students
IMRAN KHURRAM [MSCS-6]

123 .
Postal Address Parsing using Deep Learning:

Parsing of postal addresses is a standard procedure in postal automation. However, disambiguating between names and position is a challenging problem. Besides, non-standardized addresses are difficult to tackle with traditional approaches. This thesis aims at developing a deep learning based solution to address the challenges in postal address parsing.


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Faisal Shafait [DoC]
Committee Members
Muhammad Shahzad [DoC] Adnan Ul-Hasan [ECM] Muhammad Imran Malik [DoC]
Students
NOSHEEN ABID [MSCS-6]

124 .
Table Detection using Attention Based Networks:

This thesis aims at detecting tables in document images using attention-based deep networks. Table detection remains an unsolved problem due to the diverse visual appearance of tables and their complex structures.


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Faisal Shafait [DoC]
Committee Members
Muhammad Shahzad [DoC] Adnan Ul-Hasan [ECM] Muhammad Imran Malik [DoC]
Students
SAMAN ARIF [MSCS-6]

125 .
Incremental Learning of Object Detectors using Limited Training Data:

Object detectors in practice find it hard to generalize due to the wide variety of object categories present in real world environments. Incremental learning aims at providing more control to the end users of the object recognition technology so that they can add more object categories to be recognized without the intervention of the machine learning expert. This thesis will focus on indoor objec...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Faisal Shafait [DoC]
Committee Members
Muhammad Shahzad [DoC] Adnan Ul-Hasan [ECM] Muhammad Imran Malik [DoC]
Students
MUHAMMAD ABDULLAH HAFEEZ [MSCS-6]

126 .
Table structure extraction from document images:

The objective of this thesis will be to effectively extract structured information with good accuracy from tables present in the documents images.

As the industries are getting digitized now-a-days, optical character recognition is adding a huge value in the transformation process by extracting text information from document images. Tables contains information points for a large number of item...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Faisal Shafait [DoC]
Committee Members
Muhammad Shahzad [DoC] Adnan Ul-Hasan [ECM] Muhammad Imran Malik [DoC]
Students
FAISAL HUSSAIN SABIR [MSCS-6]

127 .
Automated Vehicle Detection in Aerial Imagery:

Detecting military vehicles and distinguish them out from non-military vehicles is a major challenge in the defence sector. An accurate and a faster detection of military vehicle could help to identify enemy’s move and early precautionary measures. In recent years, several deep learning based techniques have been proposed for object detection purposes. However, these were developed using such dat...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Adnan Ul-Hasan [ECM] Faisal Shafait [DoC] Muhammad Imran Malik [DoC]
Students
FARRUKH KAMRAN [MSCS-6]

128 .
Action detection and annotation in videos:

Action detection has been a hot topic for last many years. Despite the recent advances in computer vision especially deep learning techniques, action detection in realistic videos is still a problem far from being solved. The lack of availability of datasets makes it difficult to train and test various deep learning based models. This is because action detection datasets require precise spatio-te...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Committee Members
Anis ur Rahman [DoC] Asad Waqar Malik [DoC] Omar Arif [DoC]
Students
ANEEQA EJAZ [MSCS-6]

129 .
Emancipating the socio-economically marginalized youth through Freirean speed literacy and dialogue:

This research explores the potential of Freirean pedagogy to emancipate the socio-economically marginalized adults of rural Pakistan, through technology.
The oppressed adults have been domesticated with their behaviors for many years because of the country’s socio-political influence. They fail to realize the importance of education, work, and self sufficiency. Liberated adults are being looked...


Project Level: PG
Project Type: Thesis Knowledge Area: Social Issues and Professional Practice
Year: 2017
Status: Completed
Advisor
Maajid Maqbool [DoC]
Committee Members
Muhammad Shahzad [DoC] Jaudat Mamoon [DoC] Erum Afzal [DoC]
Students
NEHALA MUMTAZ [MSITE-4]

130 .
Vehicle re-identification for Visual Surveillance:

In an Intelligent traffic monitoring system vehicle re-id plays an import role as vehicle re-id is a process in which we are going to detect a vehicle and then re-identify the vehicle from our database of vehicles. This can play an important role in the field of live monitoring or tracking the vehicles or doing forensic analysis like finding patterns of different vehicles etc. This requires intel...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Safdar Abbas Khan [DoC] Anis ur Rahman [DoC]
Students
RAJA MUHAMMAD SAAD BASHIR [MSCS-6]

131 .
Person Re-Identification based Visual Surveillance System:

Large networks of cameras are increasingly deployed in public places like airports, railway stations, college campuses and office buildings. These cameras typically span large geospatial areas and have non-overlapping fields-of-views (FOVs) to provide enhanced coverage. Such networks provide huge amounts of video data, which is either manually monitored by law enforcement officers or utilized aft...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Shahzad Saleem [DoC] Asad Waqar Malik [DoC]
Students
WAJEEHA ANSAR [MSCS-6]

132 .
Weighted Hybrid Features for Person Re-Identification:

In video-surveillance, person re-identification is described as the task of recognizing distinct individuals over a network of cameras. It is an extremely challenging task since visual appearances of people can change significantly when viewed in different cameras. Many person re-identification methods offer distinct advantages over each other in terms of robustness to lighting, scale and pose va...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2016
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Shahzad Saleem [DoC] Asad Waqar Malik [DoC]
Students
Saba Mumtaz [MSIS-8]

133 .
Optimization of Person Re-Identification through Visual Descriptors:

Person re-identification is a complex computer vision task and plays a key role in providing authorities a great tool to maintain high level security. Human appearance is a critical part of information with high discriminating power. Many methods employ the combination of such types of visual features and solve one particular problem of re-identification. The combinationof Gaussian of Gaussian an...


Project Level: PG
Project Type: Thesis Knowledge Area: Intelligent Systems
Year: 2016
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Committee Members
Muhammad Shahzad [DoC] Shahzad Saleem [DoC] Asad Waqar Malik [DoC]
Students
Naima Mubariz [MSIS-8]

1 .
Cotton Crop Health Monitoring through Drone Imagery and IoT Sensors:

Pakistan is a prominent cotton producer and consumer and the economy depends profoundly on its cotton and textile industry. Despite the importance of the cotton crop to the economy, Pakistan is facing a significant decline in wide varieties of cotton crop production due to multiple challenges. The insect and pest attacks are the primary reason identified for the reduction of yield and the quality...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2022
Status: Completed
Advisor
Zuhair Zafar [DoC]
Co-Advisor
Muhammad Moazam Fraz [DoC] Muhammad Shahzad [DoC]
Students
Muhammad Usama [BEE-11]


2 .
Multimodal Estimation of Wheat Phenology using Remote Sensing Imagery:

Crop phenology describes the cycle of biological events during plant growth. These events include, for example, seedling emergence, leaf appearance, flowering, and maturity. Timing of growing seasons and their critical phases as well as estimates of them are increasingly important in changing climate. Matching the phenology of crop varieties to the climate in which they grow is critical for viabl...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2022
Status: Completed
Advisor
Zuhair Zafar [DoC]
Co-Advisor
Muhammad Moazam Fraz [DoC] Muhammad Shahzad [DoC]
Students
Asad Imtiaz [BSCS-9]


3 .
Rice Phenology Estimation and Enabling Yield Boosting using Remote Sensing and Drone Imagery:

As one of the most important sources of food for humans, rice is widely planted in Asia and its range has been extending gradually across Africa and Latin America. Obtaining accurate information regarding the phenological stages of rice helps support water, fertilisation, and crop-dusting management decisions, which in turn facilitate more efficient cultivation practices. Obtaining accurate knowl...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2022
Status: Completed
Advisor
Zuhair Zafar [DoC]
Co-Advisor
Muhammad Moazam Fraz [DoC] Muhammad Shahzad [DoC]
Students
Muhammad Qasim Khan [BSCS-9] Malik Zeeshan Ahmad [BSCS-9] Dawood Wasif [BSCS-9]


4 .
Rapid à la Mode Craze Scrutinizer (Trends Analyzer):

In today’s world, those who get a hold of trends and latest developments are the quickest are the winners. Facebook, YouTube, and other social media websites are hotspots for finding out what everyone's talking about. The idea is to create a tool which will mine these websites and get a hold of key developments taking place. Finding the right information quickly is crucial for understanding the l...


Project Level: UG
Project Type: FYP
KnowledgeArea: Information Management
Year: 2021
Status: Completed
Advisor
Farzana Jabeen [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Muhammad Uzair [BESE-9] Saud Kamran [BESE-9]


5 .
SOP Violation Detection in Fenceless areas through Live Videos:

In the current wake of a global pandemic and community spread of the virus, government officials and administrators have found restricting population movement and social interaction, a mammoth task. The high-priced schemes and lockdown containment measures are still questioning the authorities regarding the relentless and ill-informed actions of the public such as the omission of social distancin...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2021
Status: Completed
Advisor
Hasan Ali Khattak [DoC]
Co-Advisor
Muhammad Ali Tahir [DoC] Rafia Mumtaz [DoC]
Students
Mariam Ali [BESE-9] Ayesha Siddiqua [BESE-9]


6 .
Analysis of reforestation potential in Pakistan using Machine Learning and Satellite images:

The main idea behind this project is to identify under forested areas (due to various natural or human causes) in Pakistan and measure their reforestation potential. While forests play an important role in mitigating the effects of climate change, the process of reforestation is expensive, difficult to plan and even harder to execute and maintain. Therefore, it is important to understand the effi...


Project Level: UG
Project Type: FYP
KnowledgeArea: Graphics and Visualization
Year: 2021
Status: Completed
Advisor
Rafia Mumtaz [DoC]
Co-Advisor
Muhammad Ali Tahir [DoC]
Students
Muhammad Deedahwar Mazhar Qureshi [BESE-9] Amna Faisal [BESE-9]


7 .
Mentor 360:

This project is aimed at designing an Artificially Intelligent platform bridging the gap of knowledge, connecting experts with the future millionaires. The AI based algorithms will help the consulting businesses and experts to get their future leads. The platform will use the web technologies to create a social media platform that will engage the users in a 360 degree environment providing the me...


Project Level: UG
Project Type: FYP
KnowledgeArea: Software Engineering
Year: 2020
Status: Completed
Advisor
Arsalan Ahmad [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Unsha Aftab [BSCS-7] Muhammad Suhaib [BSCS-7]


8 .
Ridesharing with common interests using Artificial Intelligence:

Ridesharing/ hailing has seen little to no innovation ever since Uber introduced this model. At best, there have been various apps that modified some of the business aspects of Uber's ride-hailing model, but functionally all the apps remain the same.

We will be working on bringing an innovative approach by utilizing Artificial Intelligence to a greater extent. Although the industry leaders are...


Project Level: UG
Project Type: FYP
KnowledgeArea: Platform-based Development
Year: 2020
Status: Completed
Advisor
Arsalan Ahmad [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Asfand Shakeel [BSCS-7] Muhammad Nouman Imran [BSCS-7] Muhammad Zaki Ullah Usman [BSCS-7]


9 .
Indoor localization using deep learning techniques:

The project aims at locating unauthorized stationary or moving objects with the help of wireless communication devices using Radio tomographic imaging (RTI).
RTI is an emerging technology for imaging the attenuation caused by physical objects in wireless networks. Object localization is achieved by utilizing RSS values of the radio transceivers. Additionally, we will explore the effectiveness of...


Project Level: UG
Project Type: FYP
KnowledgeArea: Networking and Communications
Year: 2020
Status: Completed
Advisor
Arsalan Ahmad [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Usman Ishaq [BSCS-7] Jawad Haider [BSCS-7]


10 .
Automatic Survey of Reforestation Potential using Machine Learning:

This project aims to identify areas in Pakistan that are underforested due to various natural and human causes. Using satellite imagery, terrain and weather data as input to machine learning algorithms, the system aims to determine the potential vs actual vegetation cover. A survey of selected regions in Pakistan (especially semi-arid landscapes) will help automatically identify promising gegraph...


Project Level: UG
Project Type: FYP
KnowledgeArea: Graphics and Visualization
Year: 2020
Status: Completed
Advisor
Muhammad Ali Tahir [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Hamnah Khalid [BESE-8] Taha Abdullah [BESE-8]


11 .
Predictive modelling for future water resource management in Pakistan:

According to the International Monetary Fund (IMF), Pakistan ranks third amongst all surveyed countries in water scarcity and fourth in water consumption. This is due to climate conditions (Pakistan receives about 255 mm rainfall annually), excessive usage of water, and most importantly due to lack of proper water resource management-Pakistan dumps water of 70 billion USD economic value into the ...


Project Level: UG
Project Type: FYP
KnowledgeArea: Human-Computer Interaction
Year: 2020
Status: Completed
Advisor
Muhammad Imran Malik [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Asad Manzoor [BSCS-7] Muhammad Awais Adnan [BSCS-7]


12 .
Analyzing Multi Temporal Urban Development Using Deep Learning:

Introduction
Currently, all the techniques used to measure urban development and population census are manual and outdated. According to the World Bank estimates, 110 countries globally lack useful Civil Registration and Vital Statistics systems. CRVS are also fundamental to assessing progress in 67 of the 232 UN's SDG indicators. Inaccuracies in population estimation can lead to inadequate gove...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2020
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Arsalan Ahmad [DoC]
Students
Taha Waqar Alam [BSCS-7] Syed Osman Hussain [BSCS-7] Muhammad Sohaib Abid [BSCS-7]


13 .
A mobile application to convert American Sign Language to Speech:

A mobile application which takes input from the mobile camera, processes hand gestures using AI and convert them into readable sentences in real-time.


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2020
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Arsalan Ahmad [DoC]
Students
Haziq Usman Rana [BSCS-7] Alliya Fatima [BSCS-7] Moeed Ahmad [BSCS-7]


14 .
LYNX:

Re-structuring backbone networks is complex and expensive. Network operators’ needs to carry out careful network planning manually, designing networks topologies while keeping room for network upgrade in future. However, manual network planning brings with it tedious tasks of files maintenance. Our application provides network operators the ability to model networks from scratch or upgrade existi...


Project Level: UG
Project Type: FYP
KnowledgeArea: Networking and Communications
Year: 2020
Status: Completed
Advisor
Arsalan Ahmad [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Mohammad Saad Raza [BESE-8] Sami Mansoor Alavi [BESE-8] Muhammad Ali Taimoor [BESE-8]


15 .
Hyperspectral Image Analysis for Document Forgery Detection:

Hyperspectral Imaging based automatic analysis of scanned text documents is
one of the growing research areas these days. This FYP, in particular, focuses on the development of novel deep learning based models for bank checks and document forgery detection using hyperspectral imagery (consisting of more than two hundred bands).

Segmentation
of these scanned documents is a pre-requisite for ...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2020
Status: Completed
Advisor
Muhammad Imran Malik [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Muzamil Jillani [BSCS-7] Muhammad Hannan Faryad [BSCS-7]


16 .
Vehicle PathTracing Via Re-Identification:

Vehicle re-identification helps tracking vehicles with much ease. Manually tracking a visual feed is prone to error and is a tiring process.
In this project, vehicles are tracked using deep learning models. The model would be trained to detect the route of a car that passes through multiple cameras installed in city. Each car upon passing the camera would be given a unique id for tracking purpo...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2020
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Muhammad Moazam Fraz [DoC]
Students
Usama Bin Nadeem [BSCS-7] Usama Inam Paracha [BSCS-7] Muhammad Bin Jamil [BSCS-7]


17 .
Automatic Photo Editing Based on User Preferences:

Photo editing and adjustments require skill and is time consuming for many users
especially when it comes to huge batches. Moreover, the editing patterns vary depending upon the image content for example indoor and outdoor images may be edited differently by a single user.
This problem can be solved by learning editing patterns of a user with respect to the
content/features of the image via de...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2020
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Arsalan Ahmad [DoC]
Students
Ali Muhammad Chishti [BSCS-7] Zohaa Ali [BSCS-7] Ali Ather [BSCS-7]


18 .
HS3 : Visual analytics in Retail Store:

The goal of our project is to create an easy to implement face verification and recognition system for better customer analytics. Our proposed solution is to analyze in-store video footages and apply face verification and recognition techniques to identify new and frequent loyal customers, thus tracking their purchasing behavior, such as likes/dislikes and purchase patterns. Different statistical...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2020
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Ershad Hussain [BSCS-7] Muhammad Saad Tariq [BSCS-7] Osama Akhlaq [BSCS-7]


19 .
AI based HTML code generation from a Web Design Image:

Our project addresses the problem of converting a web design image template into HTML by reducing the time required to code it from scratch. This project targets a vast audience like web designers, developers and graphics designers. Our AI based solution would generate the code for the provided image template itself, thus reducing the time of web designing by a significant margin; as the web deve...


Project Level: UG
Project Type: FYP
KnowledgeArea: Graphics and Visualization
Year: 2020
Status: Completed
Advisor
Muhammad Ali Tahir [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Anas Shamoon [BSCS-7] Mohammad Abu Bakar [BSCS-7] Mudassir Hussain [BSCS-7]


20 .
Discerning Deep Fake Videos:

The spread of fake videos is rapidly increasing over the Internet.
The development of Deepfake, in particular, has
facilitated a massive abuse of fake videos. This not only affects the credibility of media but also raises human rights concerns at many levels.
The current FYP focuses on development of a novel approach to expose deepfake videos.


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2020
Status: Completed
Advisor
Muhammad Imran Malik [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Muhammad Anas Tahir [BSCS-7] Muneeb Ur Rehman [BSCS-7] Zaryab Muhammad Akram [BSCS-7]


21 .
Deep Fake Detection:

Deepfake is a technique that uses AI to swap a person’s face in a video very realistically. Recently number of deepfakes has been growing greatly and obviously this imposes significant impact on legitimacy of online videos. The fyp focuses on detecting if a video is a deepfake or real.

(ref: https://www.kaggle.com/c/deepfake-detection-challenge )


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2020
Status: Completed
Advisor
Omar Arif [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Daud Nasir Cheema [BSCS-7] Amna Hassan [BSCS-7]


22 .
Enhanced water quality monitoring and estimation by exploiting multi-modal approach :

With the advent of new technologies such as the Internet of Things (IoT), the water quality can be measured and reported in real-time. However, the limitation of this approach is that it provides point data which is insufficient to represent the entire water body quality unless the water samples are collected from a large number of points and for long durations. The collection of data from multip...


Project Level: UG
Project Type: FYP
KnowledgeArea: Software Engineering
Year: 2020
Status: Completed
Advisor
Rafia Mumtaz [DoC]
Co-Advisor
Faisal Shafait [DoC] Muhammad Shahzad [DoC]
Students
Taimoor Khan Mahsud [BESE-8] Aamir Farooq Khan [BESE-8] Muhammad Usama [BESE-8]


23 .
Chatbot based emergency hotline:

We are developing a chatbot that answers automatically to emergency calls such as 1122, and generates corresponding actions and alerts


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Ali Tahir [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Taimoor Ahmed [BSCS-6] Muhammad Tameem Rafay [BSCS-6]


24 .
Vehicle Recognition over Multiple Camera Footage:

The project, as the name suggests, is based upon the recognition of different vehicles over the video footages caught by different surveillance cameras. The aim of the project is to be used for security purposes where currently the concerned authorities require security personnel to manually look at the footages and identify each car individually which is a time consuming task. However, our propo...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Arsalan Ahmad [DoC]
Students
Abdullah Butt [BSCS-6] Muhammad Hamza [BSCS-6] Abdul Hadi [BSCS-6]


25 .
Automatic Target Recognition Using Synthetic Aperture Radar:

Automatic Target Recognition Using Synthetic Aperture Radar


Project Level: UG
Project Type: FYP
KnowledgeArea: Networking and Communications
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Abdul Basit [ECM]
Students
Khadija Kamran [BESE-7] Muhammad Uzair [BESE-7] Urvah Saikhani [BESE-7]


26 .
Change Detection in Land Cover Using Remote Sensing:

Land cover change detection is of great importance. It provides us with useful insights about how the earth is undergoing changes. This project has two phases: in the research phase, we aim to develop a novel deep learning model based on CNNs for detecting land cover changes given bitemporal aerial/satellite images; in the development phase, our goal is to build a web application powered by our m...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Madiha Hussain Malik [ECM]
Students
Hashir Ahmad [BSCS-6] Daud Akhtar Naveed [BSCS-6]


27 .
Lynx:

Re-structuring backbone networks is physically impractical and expensive. Network operators’ needs to carry out careful network planning manually, designing networks topologies while keeping room for network upgradability in future. However, manual network planning brings with it tedious tasks of files maintenance. Our application provides network operators the ability to model networks from scra...


Project Level: UG
Project Type: FYP
KnowledgeArea: Networking and Communications
Year: 2019
Status: Completed
Advisor
Arsalan Ahmad [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Muhammad Sharjeel [BSCS-6] Hassan Javed [BSCS-6] Saadullah Dar [BSCS-6]


28 .
RSSI based objects tracking:

The project aims at locating unauthorized stationary or moving objects with the help of wireless communication devices using Radio tomographic imaging (RTI).
RTI is an emerging technology for imaging the attenuation caused by physical objects in wireless networks. Object localization is achieved by utilizing RSS values of the radio transceivers. Additionally, we will explore the effectiveness of...


Project Level: UG
Project Type: FYP
KnowledgeArea: Networking and Communications
Year: 2019
Status: Completed
Advisor
Arsalan Ahmad [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Imama Jawad [BESE-7] Hiba Akram [BESE-7] Aena Hassan [BESE-7]


29 .
Deep Learning for Intelligent Surveillance through Gait Analysis:

Face recognition techniques are extensively used by law enforcement agencies worldwide. In our country, there is a significant need of a system which forms an integrated unit for security surveillance by identifying and recognising persons through their facial characteristics. Note that, in many cases, complete faces of intruders are not visible. The new dimension, which the current project aims ...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Imran Malik [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Uqqasha Ijaz [BSCS-6] Nauman Riaz [BSCS-6]


30 .
Detection of Illegal Banking Transactions using Machine Learning:

Market Abuse is a major concern for investors in stock markets of today, despite adequate responses from regulatory authorities. Existing practices within the industry to detect fraudulent activities in today’s stock market primarily rely on sets of rules derived using expert knowledge. The goal of the current project is to apply different machine learning strategies to identify suspicious transa...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Imran Malik [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Ali Ahmed Khan [BSCS-6] Muhammad Mustafa [BSCS-6]


31 .
SecureView - face recognition based suspicious person detection using mobile devices:

We propose to build a solution which identifies suspicious persons in real-time based on face recognition technology. A database of suspicious person's images will be maintained on a central server. A user such as security guard can point his mobile device at a person to scan his face in real time and identify whether he is a suspect.


Project Level: UG
Project Type: FYP
KnowledgeArea: Human-Computer Interaction
Year: 2019
Status: Completed
Advisor
Muhammad Ali Tahir [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Muhammad Saad [BESE-7]


32 .
ARSA - Augmented Reality Based Shopping Application:

Using Mobile Camera or Web Camera, real time Augmentation of jewellery items such as Necklaces, lockets ,
Rings, Earrings, bracelets or watches on selected body parts e.g. (Neck, Fingers, Ears or wrist). This App
enables the user to test and try virtually articles in real time before buying online.
Business domain: Jewellery / Watches
Other domains of application: Fashion / Shoes / Bags / Cap...


Project Level: UG
Project Type: FYP
KnowledgeArea: Platform-based Development
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Muhammad Oweis [ECM]
Students
Zaeem Habib [BESE-7] Yusra Masood [BESE-7]


33 .
Image based search on Shopware Ecommerce Store:

Online buyer can upload the image using mobile camera / web camera or image file to the respective store for search, based on the uploaded image, all the relevant/ matched articles must be displayed as a search result set.The algorithm must be capable for searching the images on a bigger dataset. Image based search module should be integrated in Shopware Backend and in Frontend.


Project Level: UG
Project Type: FYP
KnowledgeArea: Software Engineering
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Muhammad Oweis [ECM]
Students
Nadir Ali [BESE-7] Hafiz Muhammad Hassan Tariq [BESE-7] Faizan Shahid [BESE-7]


34 .
Unsupervised Forest Segmentation and Classification using Deep Learning:

Unsupervised Forest Segmentation and Classification using Deep Learning


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Faisal Shafait [DoC]
Students
Nouman Ahmed [BESE-7]


35 .
High Resolution Earth Geo-Observatory System for Sentinel-2 Imagery Using Deep Learning (HR-MARS):

High Resolution Earth Geoobservatory System for Sentinel-2 Imagery Using Deep Learning


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Faisal Shafait [DoC]
Students
Muhammad Roshan Mughees [BSCS-6] Awais Latif [BSCS-6] Abdul Razzaque Jawad [BSCS-6]


36 .
ShoppingAR - Indulging Augmented Reality with eCommerce:

Using Mobile Camera or Web Camera, real time Augmentation of jewellery items such as Necklaces, lockets ,
Rings, Earrings, bracelets or watches on selected body parts e.g. (Neck, Fingers, Ears or wrist). This App
enables the user to test and try virtually articles in real time before buying online.
Business domain: Jewellery / Watches
Other domains of application: Fashion / Shoes / Bags / Cap...


Project Level: UG
Project Type: FYP
KnowledgeArea: Platform-based Development
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Muhammad Oweis [ECM]
Students
Inshal Saleem [BESE-7] Muhammad Hamaad Latif [BESE-7] Hafiz M Asaad Iqbal [BESE-7]


37 .
Virtual Sale Person: AI based Chat Bot:

Online Chat Bot to answers possible all buyer real-time text based queries and questions.
Chat Bot implementation as a shopware plugin based on AI and Deep Learning.


Project Level: UG
Project Type: FYP
KnowledgeArea: Platform-based Development
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Muhammad Oweis [ECM]
Students
Farhana Shafi [BESE-7] Haniya Huda [BESE-7]


38 .
Virtual Try-on using 2d Images:

A generic Mobile App structure needs to be implemented, which works as a gateway to Shopware store, offering all the functionalities and features to the online user, which are being offered on the browser based version of the Shop.
This app will use Shopware as a base system, all the content, articles, articles listing etc will be fetch by App in real time should be able to handle 1 Million reco...


Project Level: UG
Project Type: FYP
KnowledgeArea: Platform-based Development
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Muhammad Oweis [ECM]
Students
Taimur Azhar [BESE-7] Rosheen Naeem [BESE-7] Hammad Aamer [BESE-7]


39 .
SpeakUp: Automated Conversional System for English Speaking Proficiency:

An automated system in which users can improve their English speaking proficiency by conversing with the system. English speaking proficiency is an unaddressed problem in our society because of which our students lose their confidence in speaking in English very early, so we are going to make a system so that the users can use to improve their English speaking. The user just have to sign in to th...


Project Level: UG
Project Type: FYP
KnowledgeArea: Human-Computer Interaction
Year: 2019
Status: Completed
Advisor
Muhammad Ali Tahir [DoC]
Co-Advisor
Arsalan Ahmad [DoC] Muhammad Shahzad [DoC]
Students
Abdul Basit [BSCS-6] Muhammad Abdullah Kazmi [BSCS-6]


40 .
Forest Change Detection Using Remote Sensing Data:

Forests are the lungs of our planet and their existence is vital to our survival on this planet. However, their existence is in danger due to several factors such as massive urbanization, industrialization, etc. Governments all over the world are striving to develop monitoring mechanisms to save the forests. Several programmes have been initiated by Pakistani government for forest cultivations, s...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Faisal Shafait [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Muhammad Rizwan Khalid [BSCS-6] Syed Abbas Hussain Naqvi [BSCS-6]


41 .
Person Re-identification using Multiple Cameras:

Recognizing the same human as he or she moves through a network of cameras with non-overlapping fields of view is an important and challenging problem in security and surveillance applications.Once the target has been tagged in one camera by a user or program, we want our system to learn the appearance of the target and recognize him/her by other cameras,


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Muhammad Moazam Fraz [DoC]
Students
Abdullah Makhdoom [BEEE-8] Salman Akhtar [BEEE-8] Shurjeel Fayyaz [BEEE-8]


42 .
Information Management, Visualization and Price Forecasting System for Wholesale Electricity Market:

With aging, bones and muscles begin to weaken and elderly people face strength and balance impairment which makes it difficult for them to perform their daily life activities. This leads to their dependency on other people.
Hence, a smart, exoskeleton wearable can be used to aid them in various movements and postures. Various Machine Learning algorithms will be used to analyze, predict and assi...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Muhammad Talha Abbas [BESE-7] Kafaite Zahra Hussain [BESE-7]


43 .
Crop Health Analysis Using Deep Learning:

Crop Health Analysis Using Deep Learning


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Faisal Shafait [DoC]
Students
Ahmad Amjad Mughal [BSCS-6] Saqib Ali Khan [BSCS-6] Faizan E Muhammad [BSCS-6]


44 .
Slum Mapping in Aerial Imagery Using Convolution Neural Networks:

Slum Mapping in Aerial Imagery Using Convolution Neural Networks


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Faisal Shafait [DoC]
Students
Syed Waleed Hyder [BESE-7] Hamna Moieez [BESE-7]


45 .
Autonomous Landing of fixed wing UAV on runway using computer vision techniques.:

Since there is no pilot onboard in UAV’s to land in emergency situations (like loss of link, navigation aid failure etc.). Thus an autonomous computer vision based solution can cater to the needs. In the project, state-of-the-art object detection algorithms based on deep learning will be optimized and implemented on Zynq FPGA based embedded platform. The output of the embedded platform will be a...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Faisal Shafait [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Muhammad Gohar Javed [BEEE-8] Minahil Raza [BEEE-8]


46 .
Automated Waste Sorting for Smart Waste Management:

Modern world meets a huge number of challenges and one of these
is to make recycling a fundamental part of sustainable economy. Effective
and efficient waste sorting is becoming one of the biggest issues in today’s
world. To be able to achieve this, it is important to use intelligent computer
systems instead of making use of manpower. In this regard, we are
proposing an automated waste sorti...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Muhammad Moazam Fraz [DoC]
Students
Hulma Naseer [BESE-7] Asad Awan [BESE-7] Muhammad Haris Latif [BESE-7]


47 .
Deployment of Deep Neural Networks on FPGAs:

Deep Neural Networks are computationally expensive computational machines. An emerging trend in this domain is to run these algorithms on stand-alone machines using FPGAs. Several FPGA vendors claims to run deep neural networks on their architectures. Current research aims to further develop this idea.


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Faisal Shafait [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Syed Tihaam Ahmad [BEEE-8] Abdullah Ashfaq [BEEE-7]


48 .
Person Re-Identification:

The process of person identification using images or videos that have been taken through different cameras with varying fields of views.


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2019
Status: Completed
Advisor
Muhammad Imran Malik [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Syed Saif Hasan [BSCS-5] Mishaal Ijaz [BSCS-5] Haseeb Ali Bhatti [BSCS-5]


49 .
Scribbles:

In the current era, time is regarded as a scarce and precious resource. Modern educational systems emphasize greatly on the efficient use of time such that students can achieve their overall objectives and goals. However, this resource called time is tough to manage when being overburdened with lengthy repetitive assignments of various calibers. This projects aims to deal with this issue by allow...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Arsalan Ahmad [DoC]
Students
Talha Mahmood [BESE-6] Basim Mannan [BESE-6]


50 .
Road Damage Detection Using Deep learning:

Idea is to detect damaged road using the deep learning concepts.


Project Level: UG
Project Type: FYP
KnowledgeArea: Algorithms and Complexity
Year: 2018
Status: Completed
Advisor
Muhammad Shahzad Younis [EE]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Assadullah [BSCS-5] Nathu Mal [BSCS-5] Aftab Hussain [BSCS-5]


51 .
Chaperone:

Chaperone


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Faisal Shafait [DoC]
Students
Umer Imran [BEEE-7] Muhammad Faisal Zahoor [BEEE-7] Muhammad Haroon [BEEE-7]


52 .
An Artificially Intelligent Chat Bot Designed Using Cognitive-Behavioral Therapy:

Depression is the leading cause of disability worldwide, and it can kill. Talking can help the depression patients— especially under the guidance of a licensed mental health professional. But therapy is expensive, inconvenient, and often hard to approach.
So, the purpose of our project is to build an artificially intelligent chatbot which would be designed using cognitive-behavioral therapy or C...


Project Level: UG
Project Type: FYP
KnowledgeArea: DSSP
Year: 2018
Status: Completed
Advisor
Ahmad Salman [EE]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Daniyal Ahmad Aftab [BEEE-7] Usman Haider [BESE-6]


53 .
Instance Segmentation for Autonomous Vehicles:

When you're driving, how important is it to be able to quickly tell the difference between a person and a stop sign? It is hugely important but typically very simple distinction that you would make reflexively, autonomous vehicles are not able to do this quite as effortlessly.

The aim of the project is to use/implement deep learning frameworks for semantic segmentation for the detection of var...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Ahmad Salman [EE] Hassan Aqeel [EE]
Students
Zaid Hassan Chaudhry [BEEE-7] Muhammad Asad Haider [BEEE-7]


54 .
Vision and Control Based Real-Time Autofocusing of an Infrared Camera:

Infrared (IR) Cameras are used in different defense applications to detect a target using its thermal signature. The more focused the target is, the better the thermal signatures are. Usually, IR cameras are equipped with manual motorized focus option and do not provide autofocus capability. Manual focus is tedious, time consuming and unpleasant task for the end user who is required to concentrat...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Ahmad Ali [ECM]
Students
Fahad Bin Khalid [BEEE-7] Muhammad Hamza [BEEE-7] Hamza Mahmood [BEEE-7]


55 .
Verification of In-Air Handwritten Sequences:

This project aims at creating, recognizing, authenticating, and visualizing documents in air. The focus would be on developing a novel algorithm to convert hand movements in air (captured by a Myo-armband worn by a user) into a sequence of x, y coordinates on a 2D Cartesian plane, and visualize them on a canvas. Existing sensor-based approaches either do not provide visual feedback or represent t...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Muhammad Imran Malik [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Rajnesh [BSCS-5] Usamah Abdur Rehman [BSCS-5] Adeen Fasihi [BSCS-5]


56 .
Forensic Examination of Hand Written Scripts for Writer Identification:

Forensic Handwriting Examiners (FHEs) often have the scenarios where they are provided with a piece of text handwritten by an author in
a different style than those texts which are present in the
dataset and were written by the same author. This is a usual
situation when the same person, for example, writes threat
letters to more than one persons and tries to change her/his
handwriting style...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Muhammad Imran Malik [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Syed Mohammad Jawwad Haider Hamdani [BSCS-5] Fatimah Nawaz [BSCS-5]


57 .
Automated Gait Recognition:

Biometric technologies are developing to cater for real-time, remote, distance-based, non-cooperative, and non-invasive surveillance. Gait recognition focuses on recognizing people by the way they walk. It offers non-invasive identification of threats at a distance where iris and/or face information is not visible, and provides long distance subject analysis. The project would primarily focus on ...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Muhammad Imran Malik [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Muhammad Asad Ali [BSCS-5] Muhammad Saad Sohail [BSCS-5]


58 .
Development of robust Face Recognition System by fusing IR and Visible Specta:

Development of robust Face Recognition System by fusing IR and Visible Specta


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Adnan Talib [ECM]
Students
Haseeb Raza [BEEE-7] Syed Zawar Shah [BEEE-7]


59 .
Deep Learning Based Forest Change Detection:

Pakistan is among one of the worst countries that are affected by forest degradation and deforestation and if appropriate measures are not taken, the country is likely to lose all of its forests in the next thirty to fifty years. Forest management faces many challenges in Pakistan. The Forestry department in Pakistan is ill-equipped to counter these challenges. It lacks human and financial resour...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Faisal Shafait [DoC]
Students
Annus Zulfiqar [BEEE-7] Osama Yaseen [BEEE-7]


60 .
Real-time color mapping of night vision imagery:

This projects aim at applying natural color mapping on the video feed from night vision camera. This involves the process of projecting RGB space to multiband night vision imagery. The end product will be software solution to transform night-time video to day-time colorization.


Project Level: UG
Project Type: FYP
KnowledgeArea: DSSP
Year: 2018
Status: Completed
Advisor
Ahmad Salman [EE]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Syed Talha Tirmizi [BEEE-7] Junaid Iqbal [BEEE-7] Muhammad Azzam Khan [BEEE-7]


61 .
AVITec (Augmented Virtually Intelligent Technology):

This project aims to achieve a VR based game that is interfaced with wearable equipment’s like electronic gloves. This project drastically enhances the user experience in games and doesn’t require KINECT. Thus, revolutionizing the gaming industry in Pakistan.
The greatest challenge to make an accurate 3-D modeling and control which works with high precision and accuracy.


Project Level: UG
Project Type: FYP
KnowledgeArea: DSSP
Year: 2018
Status: Completed
Advisor
Ahmad Salman [EE]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Syed Abdullah Shah [BEEE-7] Syed Abdul Hafeez [BEEE-7]


62 .
Hierarchical Softmax for fine-grained classification:

Purpose of the Project
To research and develop a training/prediction softmax layer / algorithm within tensorflow to have hierarchical categories.

Aim
We aim to add a layer in tensorflow i.e. hierarchical softmax which would train/predict classes of input image in hierarchical fashion, for e.g. Dog category would have sub classes of german-shephard, Labrador, etc. for reference, see YOLO v2 ...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Faisal Shafait [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Fatima Hassan [BSCS-5] Mohammad Uzair [BSCS-5]


63 .
Hardware Aware Training of Deep Networks:

Hardware Aware Training of Deep Networks


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Faisal Shafait [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Ussama Zahid [BEEE-7] Muhammad Umar [BEEE-7]


64 .
Deep learning for Drone Manuevering:

This project aims at exploring deep learning techniques for automatic maneuvering of drones.


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Faisal Shafait [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Sohaib Zahid [BESE-6] Muhammad Ali [BESE-6] Umar Khan [BESE-6]


65 .
Intelligent Gyser:

Gas water heating is responsible for more than 30% of household energy usage. Pakistan has population of about 200 millions and most of the houses use gas Gyser. Despite its prominence, users have very little knowledge and understanding of their own warm water consumption patterns, and how their behavior affects their energy (and water) bottom line.
In this project we will develop smart Gyser s...


Project Level: UG
Project Type: FYP
KnowledgeArea: Electronics
Year: 2018
Status: Completed
Advisor
Shakeel Alvi [EE]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Rahim Zahid [BEEE-7] Abdul Wahid [BEEE-7] Javeria Binte Hamid [BEEE-7]


66 .
Forest Monitoring Using Remote Sensing Imagery:

Pakistan is among one of the worst countries that are affected by forest degradation and deforestation and if appropriate measures are not taken, the country is likely to lose all of its forests in the next thirty to fifty years. Forest management faces many challenges in Pakistan. The Forestry department in Pakistan is ill-equipped to counter these challenges. It lacks human and financial resour...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Muhammad Imran Malik [DoC]
Students
Muhammad Talha Zubair [BSCS-5] Behzad Shafique [BSCS-5] Talha Saeed Khan [BSCS-5]


67 .
Vision Based Sense and Avoid system design for UAV :

The objective is to develop a prototypical vision based navigation system that may be used to automate the UAV to taxi back to its hanger after landing. The tentative workflow include,
- Adopt computer vision techniques to detect lines on the runway via conventional methods or advanced machine (deep) learning methods
- The self driving cars dataset will be used for development of the algorithm...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Jawaad Ullah Khan [ECM] Muhammad Imran Abeel [EE]
Students
Hafiz Muhammad Hassan Sadiq [BEEE-7] Muneeb Ahmad [BEEE-7] Ali Bahadur [BEEE-7]


68 .
Person Detection In Aerial Images:

Aerial monitoring serves as an important and essential ingredient for many applications such as surveillance, reconnaissance, disaster recovery and rescue procedures. Prior to scene analysis and interpretation, a common requirement for all such applications is the need of robust algorithms able to automatically detect relevant objects within the field-of-view of the optical sensor. Such automatic...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Qaiser Riaz [DoC]
Students
Muhammad Taha Mazari [BSCS-5] Muhammad Farah Adeed [BSCS-5] Zain-Ul-Abdin [BSCS-5]


69 .
CataNet: Deep Learning Model to Detect Breast Cancer Metastasis:

Automatic detection, localization, and classification of breast cancer metastasis in whole-slide images (WSI) of histopathological lymph node section is a prevailing problem. We propose a deep learning pipeline to address this problem.


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Qaiser Riaz [DoC]
Students
Khalid [BESE-6] Usama Baig [BESE-6] Muhammad Mahad Khan [BESE-6]


70 .
Location-based mobile augmented reality for a university campus:

Today's smartphones have enough computing resources and on-board sensors e.g. IMUs, GPS receivers, cameras etc. which can be used to detect, identify, and track markers and planar objects in real time. These computing abilities and on-board sensors make them good candidates for mobile augmented reality applications. The purpose of this study is to develop a framework which can be used to augmente...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Qaiser Riaz [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Muhammad Aasharib Nawshad [BESE-6] Muhammad Usama Ajmal [BESE-6] Muhammad Saleem [BESE-6]


71 .
Development of Vision Based Navigation system for Unmanned Aerial system:

The objective is to develop an image matching algorithm that may help to localize the UAV in case of GPS denial.
The tentative workplan include:
- Acquire UAV imagery of the region of interest
- Employ Computer Vision e.g., conventional feature extraction/matching via SIFT/SURF or machine (deep) learning based techniques for image matching
- Use the matched imagery and known height of the UA...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2018
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Jawaad Ullah Khan [ECM]
Students
Muhammad Hamza Mughal [BEEE-7] Maha Taqi [BESE-6] Khawaja Aneeq Ahmed Tariq [BEEE-7] Muhammad Talha Ali Bhatty [BEEE-7]


72 .
Natural language descriptions of images:

To develop a model/framework that generates natural language descriptions of images and their regions. A quick glance at an image is sufficient for a human to point out and describe an immense amount of details about the visual scene. However, this remarkable ability has proven to be an elusive task for our visual recognition models. We are working to develop methodology for automated image descr...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Muhammad Moazam Fraz [DoC]
Students
Raja Zain Abbas [BESE-5] Syed Wahab Aftab Chishty [BESE-5] Hamza Javed [BESE-5]


73 .
VeReID: Vehicle re-identification for Visual Surveillance:

Vehicle re-id is the process in which we are going to detect a vehicle and then re-identify the vehicle from our database of vehicles. This can play an important role in the field of live monitoring or tracking the vehicles from multiple cameras in urban surveillance or doing forensic analysis on the stored data for different tasks like finding patterns of different vehicles, clustering analysis ...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Muhammad Moazam Fraz [DoC]
Students
Muhammad Murtaza Shiraz [BESE-5] Asfa Mumtaz [BESE-5] Maimoona Khalid [BESE-5]


74 .
Hand drawn object recognition:

Recognizing a hand drawn object is a lot complicated than simple object recognition from images because of challenges like not so distinct features as well as difference in the drawn object with different subject. We will try to solve the problem using deep learning. We will develop a system able to recognize hand drawn objects.

Upon successful completion of this phase we will try to segment h...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Ahmad Salman [EE]
Students
Ali Akbar Rehman [BSCS-4] Muhammad Murad Malik [BSCS-4] Hassan Iftikhar [BSCS-4] Sarmad Awais [BSCS-4]


75 .
Signature Verification:

Forged signatures on bank cheques and legal documents, etc. leads to financial and legal fraud which can hinder justice or cause economic harm to individuals, organizations and/or states.

We will address this problem by applying various machine learning techniques to develop signature verification system. This system can then be deployed in different environments, such as banks and courts of l...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Muhammad Imran Malik [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Muhammad Saifullah Khan [BESE-5] Bilal Ahmad [BESE-5] Muhammad Mahad Tariq [BESE-5]


76 .
Virtual Vision for driving during fog:

This project will be sponsored by National Highway Authority. It involves development of a device for car drivers using Sensors and LCD Display. It will assist the drivers by constructing real time 3D map of the environment (road, obstacles, traffic movement) and display it on the LCD screen so that during foggy conditions and reduced visibility they can make use of sensors to drive safely.


Project Level: UG
Project Type: FYP
KnowledgeArea: Software Engineering
Year: 2017
Status: Completed
Advisor
Imran Mahmood [DoC]
Co-Advisor
Muhammad Shahzad [DoC] Muhammad Muddassir Malik [DoC]
Students
Syeda Uneeza Fareed [BESE-5] Amin Ur Rashid [BESE-5] Sanea Abid [BESE-5]


77 .
Smart Waste Management & Recycling:

Sweden recycles nearly 99 per cent of household waste. Even more waste is imported to have something to burn, to turn waste into energy or into some other useful form.
WHY CAN"T WE?

This project will be a part of International Research Collaboration (with possibility of support from Swedish Embassy).

It aims to solve the problem of waste management and recycling using Smart technologies. ...


Project Level: UG
Project Type: FYP
KnowledgeArea: Social Issues and Professional Practice
Year: 2017
Status: Completed
Advisor
Imran Mahmood [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Abdul Samad [BESE-5] Obaid-Ur-Rehman [BESE-5] Zain Imdad [BESE-5] Muhammad Waqar Ul Khaf [BESE-5]


78 .
ASuRE: Automated Surveillance Using Re-Id:

Large networks of cameras are increasingly deployed in public places like airports, railway stations, college campuses and office buildings. These cameras typically span large geospatial areas and have non-overlapping fields-of-views (FOVs) to provide enhanced coverage. Such networks provide huge amounts of video data, which is either manually monitored by law enforcement officers or utilized aft...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Muhammad Moazam Fraz [DoC]
Students
Hira Naveed [BSCS-4] Muhammad Fahad Rana [BSCS-4] Mukarram Ishaq [BESE-5]


79 .
AGRIS: Agriculture Remote Image Sensing :

AgriPred: Predicting Wheat Yield Prodcution Using Optical Remote Sensing


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Hammad Mehmood Cheema [ECM]
Students
Noor Ul Ain [BSCS-4] Kamila Zaman [BSCS-4] Gul E Fatima [BSCS-4]


80 .
Autonomous system for 3-D mapping of indoor environments:

The aim of this project is to make an autonomous system for 3-D mapping of indoor environments. The system has many uses including but not limited to assessing the inside of buildings during rescue operations and mapping interior of buildings for VR tours.


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Muhammad Moazam Fraz [DoC]
Students
Hamza Mahmood [BSCS-4] Hashir Salam Awan [BSCS-4] Muhammad Sohaib Tariq [BSCS-4]


81 .
Authenticate Quranic verses through imaging:

The project aims at authenticating Quranic verses. The objective is to develop an app that could take an image of a verse, validate it, and then returns its respective surah and verse number.


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Muhammad Moazam Fraz [DoC]
Students
Muhammad Hussain Farooq [BSCS-4] Rubina Batool [BSCS-4] Muhammad Hamza [BSCS-4]


82 .
Object detection and reconstruction using airborne UAV datasets:

In this work, we aim to develop deep learning based algorithms for object detection from RGB imagery or laser scanning 3-D point clouds acquired from airborne (drones) platform.


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Omar Arif [DoC]
Students
Mohammad Ismail Tirmizi [BSCS-4] Umer Ahmad Cheema [BSCS-4] M. Shoaib Zafar [BSCS-4]


83 .
Vehicle Detection and Recognition from Aerial Images:

Unmanned Aerial Vehicles (UAVs) or Drones are commonly used these days for surveillance and threat identification. However, with the current technology, there is a need of a human operator to constantly watch the video streams from these UAVs. This human operator involvement, which is prone to errors and negligence, may be reduced by having an automated system to detect threats from the aerial im...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Faisal Shafait [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Muhammad Noman Shafqat [BESE-5] Mahum Tariq [BESE-5] Hassan Mahmood [BESE-5]


84 .
Real Time Runway Detection from Drones using Computer Vision:

A major issue in UAVs / Drones is encountered when they loose signals from the base station and have to autonomously land. Computer vision algorithms can be used to detect roads or other regions where drone can land itself safely.


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Faisal Shafait [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Muhammad Gulzain Ali Khan [BESE-5] Muhammad Umer Javaid [BESE-5] Khadija Butt [BESE-5]


85 .
Incremental Learning of Object Detectors using Limited Training Data:

Object detectors in practice find it hard to generalize due to the wide variety of object categories present in real world environments. Incremental learning aims at providing more control to the end users of the object recognition technology so that they can add more object categories to be recognized without the intervention of the machine learning expert. This thesis will focus on indoor objec...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Faisal Shafait [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Khurram Javed [BESE-5] M. Talha Paracha [BESE-5]


86 .
3-D Facial Reconstruction with Emotion Detection and Interpretation (3D-FREDI):

This project is aimed at creating a 3D facial construct of an individual using a monocular camera. The input image will be modulated via machine learning algorithms, which will attempt to map feature points in an individual�s face onto a generic mask. This modified mask will in turn be fine tuned to match the individual concerned. Additionally, this mask will be generated in real time, to accommo...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Anis ur Rahman [DoC]
Students
Bilal Rafiq Musani [BSCS-4] Shaheer Hashim [BSCS-4] Muhammad Umair Khan [BSCS-4]


87 .
Analyzing Text Content in a Video:

This project is about extracting text content from the video and then analyzing that text using sentiment analysis. Extracting text from video involves image processing and sentiment analysis involves natural language processing and machine learning. This project will be built as an add-on in browser. When switched on, it will analyze the video in real time and generate warning if it contains neg...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Seemab Latif [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Hamza Ali Zafar [BSCS-4] Sadia Younis [BSCS-4] Muhammad Salman Ali [BSCS-4]


88 .
Stock Market Prediction Model Using Data Mining, Business Intelligence And BigData Techniques:

In today’s market, we have to incorporate various internal and external factors to analyze and predict the Trading Market’s (stock market) trends. Incorporating these factors requires handling of historical figures which amounts to massive amounts of data which is largely unformatted and irregular. Our project aims to use Data Analytics and Business Intelligence techniques to determine market tre...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Omar Arif [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Majid Ali Khan [BSCS-4] Affan Muddassir [BSCS-4] Hamza Jamshaid [BSCS-4]


89 .
Portable device to assist blind people:

The product will include a Raspberry Pi, a camera module, a power bank and a speaker/pair headphones.The device will take a snapshot of whatever is in view at the time and process the resulting image. It will then output a description of the snapshot, including objects present (cars, plants, people etc) and emotions that the people in the picture are showing. Additional features can include OCR (...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Omar Arif [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Mian Ihtisham Shah [BSCS-4] Mirza Muhammad Ali Baig [BSCS-4] Muhammad Abdullah Wajahat [BSCS-4]


90 .
Facial recognition based atendance system :

The project aims at automating the attendance of students in a class by developing a facial image based attendance system.


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Muhammad Moazam Fraz [DoC]
Students
Faaiz Asim [BSCS-4] Anas Tehseen [BSCS-4] Khizar Ijaz [BSCS-4]


91 .
Diagnostic Retinal Image Analysis:

Retinal imaging has rapidly grown within ophthalmology in the past twenty years. The availability of low cost fundus cameras to take direct images of the retina, fundus photography, makes it possible to examine the eye for the presence of many different eye diseases with a simple, non-invasive method.

We have focused on automatic early detection of Diabetic retinopathy (DR) from fundus photogr...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2017
Status: Completed
Advisor
Muhammad Shahzad [DoC]
Co-Advisor
Muhammad Moazam Fraz [DoC]
Students
Shoaib Sarwar [BESE-5] M. Rooshan Naeem [BESE-5] Mohammad Samiullah Bilal [BESE-5]


92 .
AutoDrive: Autonomous Driving System Prototype:

On the basis of a cost-effective embedded system, this work implements a preceding vehicle detection system by using computer vision technologies. The road scenes are acquired with a monocular camera. The features of the vehicle in front are extracted and recognized by the proposed refined image processing algorithm, and a tracking process based on optical flow is also applied for reducing the co...


Project Level: UG
Project Type: FYP
KnowledgeArea: Intelligent Systems
Year: 2016
Status: Completed
Advisor
Muhammad Moazam Fraz [DoC]
Co-Advisor
Muhammad Shahzad [DoC]
Students
Abdul Moiz Farooq [BESE-4] Husnain Arif Wahla [BESE-4] Muhammad Talha Munir [BESE-4]


Deep Learning

Synthetic Aperture Radar Image Processing, 3-D Point Cloud Processing, Object Reconstruction, Artifical Intelligence & Machine (Deep) Learning Applied To Digital Images, Image Analysis, Object Detection & Recognition, Remote Sensing and Satellite Image Processing. 

 

Research Experience

Research Assistant , Technical University of Munich (TUM), Munich, Germany

December 2011 - February 2016

Journal

(1.) A. Zulfiqar, M. M. Ghaffar, M. Shahzad, C. Weis, M. I. Malik, F. Shafait, N. Wehn , "AI-ForestWatch: semantic segmentation based end-to-end framework for forest estimation and change detection using multi-spectral remote sensing imagery", Journal of Applied Remote Sensing, Vol. 15 , No. 2, PP. 024518, May, 2021. IF: 1.36
(2.) Nazia Perwaiz, Muhammad Moazam Fraz, Muhammad Shahzad , "Stochastic attentions and context learning for person re-identification", PeerJ Computer Science, Vol. 7 , No. e447, Mar, 2021. IF: 3.09
(3.) Muhammad Hamza Mughal, Muhammad Jawad Khokhar, Muhammad Shahzad , "Assisting UAV Localization Via Deep Contextual Image Matching", IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (JSTARS), Vol. 14 , No. -, PP. 2445-2457, Jan, 2021. IF: 3.827
(4.) Saba Mehmood, Muhammad Shahzad, Muhammad Moazam Fraz , "DCARN: Deep Context Aware Recurrent Neural Network for Semantic Segmentation of Large Scale Unstructured 3D Point Cloud", Neural Processing Letters, Vol. 52 , No. 3, Oct, 2020. IF: 2.89
(5.) M. A. Hashmi, Q. Riaz, M. Zeeshan, M. Shahzad, M. M. Fraz , "Motion Reveal Emotions: Identifying Emotions from Human Walk Using Chest Mounted Smartphone", IEEE Sensors Journal, Vol. PP , No. -, Jun, 2020. IF: 3.076
(6.) M. Haroon, M. Shahzad, M. M. Fraz , "Multi-sized Object Detection Using Spaceborne Optical Imagery", IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (JSTARS), Vol. 13 , No. -, PP. 3032 - 3046, Jun, 2020. IF: 3.392
(7.) I. Khurram, M.M . Fraz , M. Shahzad, N. M. Rajpoot , "Dense-CaptionNet: A Sentence Generation Architecture for Fine-Grained Description of Image Semantics", Cognitive Computation, Vol. 12 , No. 2, PP. 1-31, Mar, 2020. IF: 4.287
(8.) I. Gohar, Q. Riaz, M. Shahzad, M. Z. U. H. Hashmi, H. Tahir, M. E. U. Haq , "Person Re-Identification Using Deep Modeling of Temporally Correlated Inertial Motion Patterns", Sensors, Vol. 20 , No. 3, PP. 949, Feb, 2020. IF: 3.275
(9.) M. Kanwal, A. W. Malik, A. U. Rahman, I. Mahmood, M. Shahzad , "Sustainable vehicle-assisted edge computing for big data migration in smart cities", IEEE Internet of Things Journal, Vol. 7 , No. 3, PP. 1857 - 1871, Dec, 2019. IF: 9.515
(10.) M. K. Ali, A. Rajput, M. Shahzad, F. Khan, F. Akhtar, A. Börner , "Multi-Sensor Depth Fusion Framework for Real-Time 3D Reconstruction", IEEE Access, Vol. 7 , No. -, PP. 136471 - 136480, Sep, 2019. IF: 4.098
(11.) M. Z. U. H. Hashmi, Q. Riaz, M. Hussain, M. Shahzad , "What Lies Beneath One’s Feet? Terrain Classification Using Inertial Data of Human Walk", Applied Sciences, Vol. 9 , No. 15, PP. 3099, Jul, 2019. IF: 2.474
(12.) R. M. S. Bashir, M. Shahzad, M. M. Fraz , "VR-PROUD: Vehicle Re-identification using PROgressive Unsupervised Deep architecture", Pattern Recognition, Vol. 90 , No. -, PP. 52-65, Jun, 2019. IF: 5.898
(13.) S. Arshad, M. Shahzad, Q. Riaz, M. M. Fraz , "DPRNet: Deep 3D Point based Residual Network for Semantic Segmentation and Classification of 3D Point Clouds", IEEE Access, Vol. 7 , No. 1, PP. 68892-68904, May, 2019. IF: 4.098
(14.) Q. Riaz, M. Z. H. Hashmi, M. A. Hashmi, M. Shahzad, H. Errami, A. Weber , "Move Your Body: Age Estimation Based on Chest Movement During Normal Walk", IEEE Access, Vol. 7 , No. -, PP. 28510-28524, Mar, 2019. IF: 4.098
(15.) N. Pervaiz, M. M. Fraz, M. Shahzad , "Person Re-Identification Using Hybrid Representation Reinforced by Metric Learning", IEEE Access, Vol. 6 , No. -, PP. 77334 - 77349, Dec, 2018. IF: 4.098
(16.) M.Shahzad, M. Maurer, F. Fraundorfer, Y. Wang, X. X. Zhu , "Buildings Detection in VHR SAR Images Using Fully Convolution Neural Networks", IEEE Transactions on Geoscience and Remote Sensing, Vol. 57 , No. 2, PP. 1100 - 1116, Jan, 2018. IF: 5.63
(17.) M. Shahzad, X. X. Zhu , "Automatic Detection and Reconstruction of 2-D/3-D Building Shapes From Spaceborne TomoSAR Point Clouds", IEEE Transactions on Geoscience and Remote Sensing, Vol. 53 , No. 2, PP. 1292-1310, Jan, 2016. IF: 5.63
(18.) M. Shahzad, X. X. Zhu , "Robust Reconstruction of Building Facades for Large Areas Using Spaceborne TomoSAR Point Clouds", IEEE Transactions on Geoscience and Remote Sensing, Vol. 53 , No. 2, PP. 752-769, Jan, 2015. IF: 5.63
(19.) M. Schmitt, M. Shahzad, X. X. Zhu , "Reconstruction of Individual Trees from Multi-Aspect TomoSAR Data", Remote Sensing of Environment, Vol. 165 , No. -, PP. 175-185, Jan, 2015. IF: 8.218
(20.) X. X. Zhu, N. Ge, M. Shahzad , "Joint Sparsity in SAR Tomography for Urban Mapping", IEEE Journal of Selected Topics in Signal Processing, Vol. 9 , No. 8, PP. 1498-1509, Jan, 2015. IF: 6.688
(21.) X. X. Zhu, M. Shahzad , "Facade Reconstruction Using Multiview Spaceborne TomoSAR Point Clouds", IEEE Transactions on Geoscience and Remote Sensing, Vol. 52 , No. 6, PP. 3541-3552, Jan, 2014. IF: 5.63

Conference

(1.) S. A. Khan, Y. Shi, M. Shahzad, X. X. Zhu , "FGCN: Deep Feature-Based Graph Convolutional Network for Semantic Segmentation of Urban 3D Point Clouds", Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, Jun, 2020, Seattle, Washington , USA.
(2.) N. Pervaiz, M. M. Fraz, M. Shahzad , "Hierarchical Refined Local Associations for Robust Person Re-Identification", Proceedings of the Proceedings of the International Conference on Robotics and Automation in Industry (ICRAI), Oct, 2019, NUST , Pakistan.
(3.) F. Kamran, M. Shahzad, F. Shafait , , "Automated Military Vehicle Detection from Low-Altitude Aerial Images", Proceedings of the Digital Image Computing: Techniques and Applications (DICTA), Dec, 2018, Canberra , Australia.
(4.) S. Batool, M. Z. Ali, M. Shahzad, M. M. Fraz , , "End to End Person Re-Identification for Automated Visual Surveillance", Proceedings of the third IEEE International Conference on Image Processing, Applications and Systems (IPAS 2018), Dec, 2018, Inria Sophia Antipolis , France.
(5.) R. M. S. Bashir, M. Shahzad, M. M. Fraz , , "DUPL-VR: Deep Unsupervised Progressive Learning for Vehicle Re-Identification", Proceedings of the 13th International Symposium on Visual Computing (ISVC), Nov, 2018, Las Vegas , United States.
(6.) I. Khurram, M. M. Fraz, M. Shahzad , , "Detailed sentence generation architecture for image semantics description", Proceedings of the 13th International Symposium on Visual Computing (ISVC), Nov, 2018, Las Vegas , United States.
(7.) W. Ansar, M M Fraz, M Shahzad , , "Two Stream Deep CNN-RNN Attentive Pooling Architecture for Video-based Person Re-Identification", Proceedings of the 23rd Iberoamerican Congress on Pattern Recognition, Nov, 2018, Madrid , Spain.
(8.) M.Shahzad, M. Maurer, F. Fraundorfer, Y. Wang, X. X. Zhu , , "Extraction of buildings in vhr SAR images using fully convolution neural networks", Proceedings of the IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Jul, 2018, Valencia , Spain.
(9.) M. Badar, M. Shahzad, M. M. Fraz , , "Simultaneous Segmentation of Multiple Retinal Pathologies Using Fully Convolutional Deep Neural Network", Proceedings of the Medical Image Understanding and Analysis, Jan, 2018, Southampton , United Kingdom.
(10.) Y. Sun, M. Shahzad, X. X. Zhu , , "Building Height Estimation in Single SAR image using OSM building footprints", Proceedings of the Joint Urban Remote Sensing Event (JURSE), Mar, 2017, Dubai , United Arab Emirates.
(11.) M. Schmitt, M. Shahzad, X. X. Zhu , , "Forest remote sensing on the individual tree level by airborne millimeterwave SAR", Proceedings of the IEEE Geoscience and Remote Sensing Symposium (IGARSS), 2016, Beijing, China, Jan, 2016, Beijing , China.
(12.) Y. Sun, M. Shahzad, X. X. Zhu , , "First prismatic building model reconstruction from TomoSAR point clouds", Proceedings of the XXIII ISPRS Congress 2016, Prague, Czech Republic , Jan, 2016, Prague , Czech Republic.
(13.) X. X. Zhu, N. Ge, M. Shahzad , , "Exploiting Group Sparsity in SAR Tomography", Proceedings of the 3rd International Workshop on Compressed Sensing Theory and its Applications to Radar, Sonar and Remote Sensing (CoSeRa) , Jan, 2016, Pisa , Italy.
(14.) X. X. Zhu, N. Ge, M. Shahzad , , "Group Sparsity in SAR Tomography – Experiments on TanDEM-X Data Stacks", Proceedings of the 16th International Radar Symposium (IRS), Jan, 2016, Dresden , Germany.
(15.) M. Shahzad, X. X. Zhu, R. Bamler , , "Façade structure reconstruction using spaceborne TomoSAR point clouds", Proceedings of the IEEE Geoscience and Remote Sensing Symposium (IGARSS), Jan, 2016, Munich , Germany.
(16.) M. Shahzad, X. X. Zhu , , "Reconstruction of building footprints using spaceborne TomoSAR point clouds", Proceedings of the ISPRS Geospatial Week 2015, ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume II-3/W5, 2015, Jan, 2015, La Grande Motte , France.
(17.) M. Shahzad, X. X. Zhu , , "Detection Of Buildings In Spaceborne TomoSAR Point Clouds Via Hybrid Region Growing And Energy Minimizatison Technique", Proceedings of the Joint Urban Remote Sensing Conference (JURSE), Jan, 2015, Lausanne , Switzerland.
(18.) M. Shahzad, M. Schmitt, X. X. Zhu , , "Segmentation and crown parameter extraction of individual trees in an airborne TomoSAR point cloud", Proceedings of the PIA15+HRIGI15 – Joint ISPRS conference, International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XL-3/W2, 2015, Jan, 2015, Munich , Germany.
(19.) X. X. Zhu, Y. Wang, M. Shahzad, R. Bamler , , "Spaceborne TomoSAR and Beyond: From SAR Image Stacks to Objects", Proceedings of the 10th European Conference on Synthetic Aperture Radar (EUSAR), Jan, 2014, Berlin , Germany.
(20.) M. Shahzad, X. X. Zhu , , "Reconstructing 2-D/3-D Building Shapes From Spaceborne Tomographic SAR Point Clouds", Proceedings of the Photogrammetric Computer Vision (PCV), International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Technical Commission III Symposium, Jan, 2014, , Switzerland.
(21.) M. Shahzad, X. X. Zhu , , "Automatic Large Area Reconstruction of Building Facades from Spaceborne TomoSAR Point Clouds", Proceedings of the IEEE Geoscience and Remote Sensing Symposium (IGARSS), Jan, 2014, Quebec , Canada.
(22.) M. Shahzad, X. X. Zhu , , "Reconstruction of Building Facades Using Spaceborne Multiview TomoSAR Point Clouds", Proceedings of the IEEE Geoscience and Remote Sensing Symposium (IGARSS), Jan, 2013, Melbourne , Australia.
(23.) X. X. Zhu, M. Shahzad, Y. Wang, R. Bamler , , "Tomographic Urban Imaging Using TerraSAR-X High Resolution Spotlight Data Stacks", Proceedings of the ESA Living Planet Symposium 2013, Jan, 2013, Edinburgh , United Kingdom.
(24.) M. Shahzad, X. X. Zhu , , "Robust Building Facade Reconstruction from Spaceborne TomoSAR Points", Proceedings of the CMRT13 - City Models, Roads and Traffic 2013 - ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Jan, 2013, Antalya , Turkey.
(25.) M. Shahzad, X. X. Zhu , , "Building façades reconstruction using multi-view tomosar point clouds", Proceedings of the Joint Urban Remote Sensing Event (JURSE), Jan, 2013, Sao Paulo , Brazil.
(26.) M. Shahzad, X. X. Zhu , , "From TomoSAR Point Clouds to Objects: Facade Reconstruction", Proceedings of the Tyrrhenian Workshop on Advances in Radar and Remote Sensing (TyWRRS), Jan, 2012, Naples , Italy.
(27.) M. B. Malik, A. Hasan, N. ur Rehman, M. Shahzad , , "State-space recursive least squares based object tracking", Proceedings of the 8th International Multitopic Conference INMIC 2004, Jan, 2004, , Pakistan.

Assistant Professor , School of Electrical Engineering & Computer Science (SEECS), National University of Sciences & Technology (NUST), Islamabad
October 2016 - February 2022

Details:

Associate Professor , SEECS, NUST
February 2022 - Present

Details:


Courses taught at SEECS
UG Courses:

CS370 Artificial Intelligence (3+1)

CS250 Data Structures & Algorithms (3+1)

EE433 Digital Image Processing (3+1)

CS474 Computer Vision (2+1)

 

PG Courses:

CS878 Deep Learning (3+0)

CS862 Advanced Image Processing (3+0)

RM898 Research Methodology (2+0)