Video Surveillance Lab FAST-NUCES Karachi

About Us

Due to increased security risks in public places, especially in the metropolis Karachi, Pakistan, CCTV cameras installation and their monitoring has become crucial. According to the video surveillance market outlook, the burgeoning global surveillance is expected to reach $ 87, 361.86 million by 2025. At the moment, many security-based companies have employed manual surveillance which is cumbersome and prone to error. It, therefore, necessitates the use of technology such as computer vision and deep learning to automate crime detection.



At the Smart Video Surveillance Lab, we aim to research and develop solutions to address the security challenges of Pakistan. Working in tandem with the National Centre for Big Data & Cloud Computing (NCBC), Pakistan, the lab aims to develop deep learning models to extract and process appropriate information from a large number of video streams via IoTs in an automated manner to the benefit of local government agencies, hospitals, and educational institutions. 

Our Team

Meet the Principal Investigator as well the as Co – Principal Investigators of the Lab.

Dr. Muhammad Atif Tahir

Principal Investigator

Dr. Muhammad Atif Tahir received his PhD from School of Computer Science & Engineering at Queens University, Belfast, UK, MSc in Computer Engineering from King Fahd University, Dhahran, KSA, and BE in Computer Systems Engineering from NED University of Engg, and Tech., Karachi, Pakistan. He is also an academic fellow of UK higher education. He is currently working as Professor in the School of Computer Science, FAST University, Karachi Campus, Pakistan. Before joining FAST, he was working as Senior Lecturer at Northumbria University, United Kingdom.
Dr. Tahir also worked as Research Officer at the Centre for Vision, Speech and Signal Processing (CVSSP), University of Surrey on research projects involving interactive semantic video/audio search with a large thesaurus of machine-learned audio-visual concepts and face recognition on the uncontrolled environment. He has developed novel machine learning methods for concept detection/visual learning/face recognition. One of my methods has achieved the best performance and ranked first in prestigious international software competitions on visual category recognition (TrecVid 2009/2010, Pascal VOC 2010/2008, and ImageCLEF 2010). Dr. Tahir also worked as Research Fellow at the University of the West of England. His main research is in Machine Learning & Combinatorial Optimization Techniques with applications in image/video retrieval, cancer classification, surface inspection, bioinformatics, multi-label classification, and face recognition. He has authored and co-authored more than 60 publications in top-quality journals including IEEE Transactions on Pattern Analysis and Machine Intelligence, Pattern Recognition, Journal of Machine Learning Research, IEEE Transactions on Multimedia.

Dr. Muhammad Nouman

Co - Principal Investigator

Nouman M. Durrani received his Ph.D. degree from the FAST National University of Computer and Emerging Science, Karachi, Pakistan, in 2017. He is currently an Assistant Professor with the Department of Computer Science, FAST National University of Computer and Emerging Science. He is also a member of the Systems Research Laboratory, the Center for Research in Ubiquitous Computing (CRUC), and the Smart Video Surveillance Laboratory FAST-NUCES. His research interests include heterogeneous devices volunteer computing systems, computer vision, human computation, cloud computing, distributed systems, WSNs, and big data analytics.

















Research Domains

Theft Car Surveillance Application

Our system detects multi-level features, verifies the extracted model, color, and other aspects of the vehicle using a centralized database to identify if the car has been stolen.

Suspicious Activity Detection System

In order to automate the detection of notorious activity from a stream of CCTV videos, we have employed deep learning techniques such as anomaly detection.

Missing Child Surveillance System

We have developed, in collaboration with the Sarmi Burni Trust, using state-of-the-art facial recognition as well as age invariant algorithms, a surveillance system for finding and returning children to their rightful parents/guardians.

Person Re-Identification System

Person re-identification refers to finding a person from multiple images taken from cameras placed at various locations and times. Supervised re-id methods rely on a labeled dataset which is usually not available in real-world situations.

St-4, Sector 17-D، National Hwy 5, Karachi, Karachi City, Sindh
Our hours

10:00 AM – 06:00 PM
Monday – Friday

Contact us

Phone: 1 800 755 60 20