Uday Agarwal

I'm currently a Visiting Researcher at the Computer Vision and Machine Learning Group, National University of Singapore advised by Prof. Angela Yao. My research interests lie in how visual observations can serve as a signal for building contextual AI systems that understand people and, learning from continuous video streams.

Previously, I served as a Research Assistant at the Vision, Language and Learning Group, Indian Institute of Technology Jodhpur advised by Prof. Anand Mishra. My work there centered on fine-grained video understanding, including moment retrieval, video retrieval using multi-modal queries, and self-supervised representation learning. During this time, I was fortunate to be mentored by Dr. Manish Gupta (Microsoft Bing) on fine-grained video analysis.

I am also privileged to be mentored by Dr. Ankush Gupta (Google DeepMind). Under his guidance, I built my foundations rigorously - from core computer science principles to training neural networks from scratch and regular discussion of research papers in seminar-style sessions.

I graduated with a Bachelors in Computer Science (specialization in Artificial Intelligence and Machine Learning) in September, 2023.

I am grateful to Mr. Anup Gupta (Nexus Venture Partners) and Mr. Anurag Gupta (EY) for their invaluable guidance on my career path.

Email  /  LinkedIn  /  CV  /  Google Scholar  /  Github

Image

Recent Activities

Selected Publications
PontTuset PatientVLM Meets DocVLM: Pre-Consultation Dialogue Between Vision-Language Models for Efficient Diagnosis
K Lokesh*, Abhirama Subramanyam Penamakuri*, Uday Agarwal, Apoorva Challa, Shreya K Gowda, Somesh Gupta, Anand Mishra
AAAI, 2026 (Main Track) (*equal contribution)
paper / project page

PontTuset Aligning Moments in Time using Video Queries
Yogesh Kumar*, Uday Agarwal*, Manish Gupta, Anand Mishra
ICCV, 2025 (*equal contribution)
paper / code & dataset / slides

PontTuset CHAPVIDMR:Chapter-based Video Moment Retrieval using Natural Language Queries
Uday Agarwal*, Yogesh Kumar*, Abu Shahid*, Prajwal Gatti, Manish Gupta, Anand Mishra
ICVGIP, 2024   (Spotlight) (* equal contribution)
paper / code & dataset

Miscellanea

I have been learning and practicing the violin for the past 18 years and am skilled in both Indian Classical and Western Styles of Music. My musical journey has been guided by the esteemed Mr. Gulzar Hussain. I also hold a completely objective and non-negotiable opinion that Roger Federer is the greatest to ever hold a tennis racquet.


Template design taken from here.