Muhammad Ahmed Raza profile picture

M. Ahmed Raza

Assistant Professor

Department of Computer Science
Information Technology University (ITU)

Lahore, Pakistan

About me

I am an Assistant Professor in the Department of Computer and Software Engineering at Information Technology University (ITU), Lahore. Previously, I completed my PhD at the University of Edinburgh, supervised by Prof. Robert B. Fisher. My research focuses on Computer Vision and healthcare applications, with particular emphasis on action recognition and motion analysis.

Prior to this, I was a Computer Vision Engineerat Danu Robotics where I am also one of their Founding Member. During my PhD, I also completed an internship with Sony in Tokyo, where I worked on developing human centric deep learning based models. I completed my Master's in Electrical Engineering from Air University, Pakistan and my Bachelor's in Electrical Engineering from Pakistan Institute of Engineering and Applied Science (PIEAS), Pakistan.

I have published in reputable journals and conferences including NeurIPS, Image and Vision Computing, and Machine Vision and Applications. I actively serve as a reviewer for WACV, ECCV, ICPRS, and AIIH.

News

  • 2025: Joined ITU Lahore as Assistant Professor
  • 2024: Successfully defended PhD Thesis at University of Edinburgh
  • 2023: EatSense dataset paper published in Image and Vision Computing
Prospective Students

I am looking for motivated Master's and PhD students interested in Computer Vision, Action Recognition, and Healthcare AI. If you are passionate about research, please email me with your CV and research interests.

Research Interests

My research interests broadly include Computer Vision and healthcare applications. More specifically:

Motion Estimation & Quantification Behaviour & Activity Analysis Discriminative Modelling Action Recognition Temporal Segmentation Object Detection
Affiliations
ITU Lahore
Current
University of Edinburgh
PhD Alumni
Sony
Internship
Danu
SRL

Research

Projects

My publications grouped by research thread — from a multi-year program on vision-based health monitoring to earlier work in detection, recognition, and geometric vision.

EatSense
Featured · PhD Research 2021–2024

EatSense — Vision-Based Monitoring of Eating Behaviour & Musculoskeletal Function

My core PhD research program at the University of Edinburgh: building the EatSense dataset and a family of vision-based methods to quantify eating behaviour, upper-body motion quality, and simulated physical weakness. The work spans dataset creation, action recognition and temporal localization, motion quantification, and privacy-preserving analysis — aimed at applications in rehabilitation monitoring and early detection of musculoskeletal decline.

Object-Centric Scene Representation Learning

A generative spatial-temporal factorization model that learns object-centric representations from video without supervision, disentangling scenes into their constituent objects.

Aerial & Visual Object Detection

Small-scale object detection for UAV imagery, and classical object recognition using Bag-of-Visual-Words features with a neural network classifier.

Visual Odometry & Motion Estimation

Camera-based ego-motion and video stabilization: a simplified visual odometry pipeline using Graph Cut RANSAC, and a comparative study of SIFT-flow vs. optical-flow stabilization.