Optical Imaging System
Laboratory
People
Muhammad Tayyab
B.S. Computer Information Science, Pakistan Institute of Engineering & Applied Sciences, (2012)
M.S. Computer Science, Lahore University of Management Sciences, (2014)
Ph.D. Computer Science, University of Central Florida, (2023)
I am deeply interested in the applications of machine learning in optics, particularly in solving complex problems that lie at their intersection. My work centers on leveraging machine learning techniques to advance computational imaging, which opens the door to high-resolution, real-time imaging systems that can operate under challenging conditions. I am particularly interested in exploring non-line-of-sight imaging, where machine learning algorithms can untangle and reconstruct hidden scenes from scattered light, and compressed sensing, where machine learning can be used to reconstruct the signal from compressed noisy measurements. Additionally, I also aim to incorporate diffusion models to understand light propagation, which has promising applications in phase recovery, hyperspectral imaging, and adaptive optics.
In my free time, I enjoy watching soccer matches and am an avid follower of Premier League teams like Manchester City, Liverpool, and Arsenal. Additionally, I have a keen interest in drone photography. I am particularly inspired by Aydin Buyuktas's flatland photography and strive to replicate his style using my own DJI Air drone.
[1] Sultan Daud Khan, Muhammad Tayyab, Muhammad Khurram Amin, Akram Nour, Anas Basalamah, Saleh M. Basalamah, and Sohaib Khan. Towards a Crowd Analytic Framework For Crowd Management in Majid-al-Haram. 17th Scientific Meeting on Hajj & Umrah Research, abs/1709.05952, 2017.
[2] Haroon Idrees, Muhammad Tayyab, Kishan Athrey, Dong Zhang, Somaya Ali Al-Maadeed, Nasir M. Rajpoot, and Mubarak Shah. Composition Loss for Counting, Density Map Estimation and Localizationin Dense Crowds. In European Conference on Computer Vision, 2018.
[3] Muhammad Tayyab, Fahad Ahmad Khan, and Abhijit Mahalanobis. Compressing Deep CNNs Using Basis Representation and Spectral Fine-Tuning. In 2021 IEEE International Conference on Image Processing (ICIP), pages 3537–3541, 2021.
[4] Muhammad Tayyab and Abhijit Mahalanobis. Simultaneous Learning and Compression for Convolution Neural Networks. In 2022 IEEE International Conference on Image Processing (ICIP), pages 3636–3640, 2022.
[5] Muhammad Tayyab and Abhijit Mahalanobis. Leveraging low rank filters for efficient and knowledge-preserving lifelong learning. In Workshop in European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD), 2023
