Susim Roy

I'm a first year PhD student in the Electrical and Computer Engineering Department at the Johns Hopkins University where I am a member of the VIU Lab, advised by Prof. Vishal Patel.
Previously, I received my MS in Computer Science and Engineering at the University at Buffalo where I worked on privacy-preserving ML and controllable image generation at the TRAIL Lab under Prof. Nalini Ratha. Prior to my graduate studies, I was an undergraduate researcher under Prof. Richa Singh and Prof. Mayank Vatsa at the IAB Lab during my time at IIT Jodhpur, where I worked on adversarial machine learning.

Email / CV / LinkedIn / Scholar / Github

Research Interests

I'm broadly interested in computer vision with focus on controllable foundation generative models and vision foundation models. Currently, my research focuses on general object re-identification and image and video generative models.

News

[Aug 2026]: Joined Johns Hopkins University as a PhD student
[May 2026]: Awarded the Patricia J. Eberlein Master's Thesis Award at UB.
[May 2026]: Two papers are accepted to ICIP 2026.
[Jan 2026]: One paper is accepted to ICASSP 2026.
[Aug 2025]: Awarded the Broadening Participation Grant for ICCV 2025.
[July 2025]: One paper is accepted to ICCVW 2025.

Publications

Please refer to my Google Scholar profile for the complete list of publications.
Representative papers are highlighted.

	CoreView: Compact Yet Complete Video Representation Susim Roy, Arjun Ramesh Kaushik, Nalini Ratha, Venu Govindaraju IEEE International Conference on Image Processing (ICIP), 2026 arXiv \| Code
	Efficient and Secure Convolutions on Encrypted Data Susim Roy, Bharat Chandra Yalavarthi, Nalini Ratha, IEEE International Conference on Image Processing (ICIP), 2026 arXiv \| Code
	Multimodal Privacy-Preserving Entity Resolution using Fully Homomorphic Encryption Susim Roy, Nalini Ratha IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2026 Oral Presentation Paper \| Slides \| Code
	TAIGen: Training-Free Adversarial Image Generation via Diffusion Models Susim Roy, Anubhooti Jain, Mayank Vatsa, Richa Singh ICCV Workshops 2025 Paper \| Supp \| Slides \| Code
	Discerning the Chaos: Detecting Adversarial Perturbations while Disentangling Intentional from Unintentional Noises Anubhooti Jain, Susim Roy, Kwanit Gupta, Mayank Vatsa, Richa Singh IEEE International Joint Conference on Biometrics (IJCB), 2024 Oral Presentation Paper \| Slides \| Poster \| Code

Awards and Academic Services

Awards

Patricia J. Eberlein Award | UB CSE Dept. | May 2026
Master's Thesis award for students who display academic excellence, with preference given to female students.
ICCV BP Award | IEEE/CVF BP Council | Oct 2025
Broadening Participation Award for ICCV 2025.
Best Student Paper | IEEE Rochester Section | Nov 2024
Best Student Paper Award at the 2024 IEEE Western New York Image Processing Workshop.
DEI Travel Grant | IEEE Biometrics Council | Sep 2024
Rising Star in Diversity, Equity and Inclusion advocacy travel grant for IJCB 2024.
MITACS Research Internship | MITACS | May 2023 – Aug 2023
MITACS scholarship for an onsite AI research internship at the University of Alberta.

Academic Services

Aug 2026: Serving as a reviewer for BioSig 2026
June 2026: Serving as a reviewer for BMVC 2026
Apr 2025: Serving as a reviewer for CVPRW 2025
[Jan - Apr] 2024: Teaching assistant for Deep Learning
[Jan - Apr] 2023: Teaching assistant for Pattern Recognition and Machine Learning

Source code from John Barron.