News

• 02/2023: Our work on Multiscale Encoder-Decoder Video Transformer accepted to CVPR 2023

Research Area

My research is in the general area of Computer Vision with a special focus on Video Understanding. My demonstrated research experience includes using multiscale spatiotemporal attention mechanism for video segmentation, distributed feedback for scene segmentation, vehicle trajectory prediction from birds-eye-view videos for autonomous driving.

Selected Publications ( See Google Scholar for full list.)

• MED-VT: Multiscale Encoder-Decoder Video Transformer with Application to Object Segmentation
  Rezaul Karim , He Zhao, Richard P. Wildes, Mennatullah Siam
  IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023.
  Project Paper supplement code

• Distributed iterative gating networks for semantic segmentation
  Rezaul Karim, Amirul Islam, NDB Bruce
  IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2020.
  Paper

• Recurrent iterative gating networks for semantic segmentation
  Rezaul Karim, Amirul Islam, NDB Bruce
  IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2019.
  Paper

• Lossless Image Compression Using List Update Algorithms
  Arezoo Abdollahi, Neil Bruce, Shahin Kamali, and Rezaul Karim String Processing and Information Retrieval: 26th International Symposium (SPIRE), 2019.
  Paper Code

• CoMOGrad and PHOG: from computer vision to fast and accurate protein tertiary structure retrieval
  Rezaul Karim, Mohd Momin Al Aziz, Swakkhar Shatabda, M Sohel Rahman, Md Abul Kashem Mia, Farhana Zaman, Salman Rakin
  Nature Scientific Reports (Scientific Reports), 2015.
  Paper

Experience

• Autonomous Driving Team, Noah’s Ark Laboratory, Huawei Canada Corp.
  Associate Researcher, Intern
  • Conducted research in perception/prediction related problems in autonomous driving
• Vision Lab, York University
  Graduate Research Assistant with Professor Richard P. Wildes
  • Spatial temporal attention models for video understanding
  • Understanding the role of attention in transformer-based models
  • Unsupervised video object segmentation from unconstrained videos
  • Interpretability of temporal dynamics in video understanding
• Department of EECS, York University Teaching Assistant
  • Design and Analysis of Algorithms
  • Introduction to the Theory of Computation
  • Operating Systems
• Computer Vision Lab University of Manitoba
  Graduate Research Assistant with Professor Neil Bruce
  • Feedback and Gating in Deep Neural Networks
  • Adversarial attack and defence in ConvNet models
  • Semantic segmentation, scene parsing, panoptic segmentation

Education

• Ph.D. in Electrical Engineering and Computer Science, York University
  • Thesis: Spatial Temporal Attention Models for Video Understanding
  • Supervisor: Professor Richard P. Wildes
• M.Sc. in Computer Science, University of Manitoba
  • Thesis: Feedback and Gating in Deep Neural Networks
  • Supervisor: Dr. Neil D. B. Bruce

---