I am a DPhil student at the VGG Lab (Oxford)advised by Prof. Andrew Zisserman at the University of Oxford. I am interested in all kinds of video understanding, specially time-sensitive video understanding.

Previously, I finished my Master’s in AI at the University of Amsterdam. Along with the coursework, I did research on video understanding at the VISLab advised by Prof. dr. Cees Snoek and Dr. Makarand Tapaswi.

In the past, I’ve been a Research Fellow at Wadhwani AI where I primarily worked on estimating infant anthropometry from a monocular video. During the pandemic, I also worked on using cough as a biomarker to detect presence of CoVID-19.

Before that, I completed my Bachelor’s in Mathematics at IIT Kanpur where I worked with Dr. Piyush Rai and Dr. Swaprava Nath.

Selected publications

TARA: Simple and Efficient Time Aware RetrievalAdaptation of MLLMs for Video Understanding

Piyush Bagad, Andrew Zisserman

ArXiv 2025

ArXiv   |   Paper   |   Project page   |   Code   | Dataset   |   Poster   |   Video

Chirality in Action: Time-Aware Video Representation Learning by Latent Straightening

Piyush Bagad, Andrew Zisserman

NeurIPS 2025

ArXiv   |   Paper   |   Project page   |   Code   | Dataset   |   Poster   |   Video

The Sound of Water: Inferring Physical Properties from Pouring Liquids

Piyush Bagad, Makarand Tapaswi, Cees G. M. Snoek, Andrew Zisserman

ICASSP 2025

ArXiv   |   Paper   |   Project page   |   Code   |   Poster   |   Video

Test of Time: Instilling Video-Language Models with a Sense of Time

Piyush Bagad, Makarand Tapaswi, Cees G. M. Snoek

CVPR 2023

Also presented at Workshop on Understanding Foundation Models at ICLR 2023

ArXiv   |   Paper   |   Project page   |   Code   |   Poster   |   Video

How Severe is Benchmark-Sensitivity in Video Self-Supervised Learning?

Fida Mohammad Thoker, Hazel Doughty, Piyush Bagad, Cees Snoek

ECCV, 2022 (Tel Aviv, Israel)

ArXiv   |   Project page   |   Code

C-3PO: Towards Rotation Equivariant Feature Detection and Description

Piyush Bagad*, Floor Eijkelboom*, Mark Fokkema*, Danilo de Goede*, Paul Hilders*, Miltiadis Kofinas

VIPrior Workshop, ECCV, 2022 (Tel Aviv, Israel)

Paper   |   Project page   |   Code   |   Poster   |   Oral Presentation

Data-Sharing Economy: Value-Addition from Data meets Privacy

Piyush Bagad, Subrata Mitra, Sunny Dhamnani, Atanu R Sinha, Raunak Gautam, Haresh Khanna

WSDM Demos, 2021 (Virtual event, Israel)

Work done during Internship at Adobe Research (India) in 2018.

Paper   |   Video   |   Code

Selected projects

Reproducibility Study of "Counterfactual Generative Networks"

Piyush Bagad*, Paul Hilders*, Jesse Maas*, Danilo de Goede*

Best paper award

ML Reproducibility Challenge, ReScience Journal

Invited to present at Journal Showcase Poster Session @ NeurIPS 2022.

Paper   |   Code

Quantifying CLIP’s ability to Perform Cross-Modal Grounding Using Attention-Model Explainability

Piyush Bagad*, Paul Hilders*, Danilo de Goede*

Best poster award in the Interpretability course at UvA

Poster   |   Code   |   Demo

AI in the real-world

Cough Against COVID: Evidence of COVID-19 Signature in Cough Sounds

Piyush Bagad, Aman Dalmia, Jigar Doshi, Arsha Nagrani, Parag Bhamare, Amrita Mahale, Saurabh Rane, Neeraj Agarwal, Rahul Panicker

ArXiv, 2021 (Mumbai, India)

ArXiv   |   Project page   |   Code

Media credits: Wadhwani AI

Newborn Anthropometry: Estimating infant weight from a monocular video

Piyush Bagad, Aman Dalmia, Makarand Tapaswi, Sansiddh Jain, Aditya A Sarma, Namrata Deka, Jigar Doshi, Parag Bhamare, Amrita Mahale, Rahul Panicker (Not in order)

Unpublished, 2019-2021 (Mumbai, India)

ArXiv   |   Project page   |   Code

Media credits: Wadhwani AI

Synthetic Data Generation for Visual Weighing Machine

Piyush Bagad, Aman Dalmia, Siddharth Singh, Vishal Gupta, Aditya AS, Namrata Deka

Platform: Blender 2.8

Copyrights: Wadhwani AI

Deformable 3D Model for Newborn Babies

Piyush Bagad, Aman Dalmia, Rahul Kumar, Arsh Verma, Rahul Panicker

Underlying Model: SMIL (Nicholas Hesse et. al)

Copyrights: Wadhwani AI