Pradyumna Chari

Email  /  Google Scholar  /  LinkedIn

profile photo

I am a first year PhD student at the Visual Machines Group, UCLA, advised by Prof. Achuta Kadambi. At VMG, we work on developing cutting-edge tools at the intersection of physics and artificial intelligence, to be applied to diverse problems in computer vision and computational imaging.

I completed my Bachelor's degree in Electrical Engineering from the Indian Institute of Technology, Madras, India, where I was the President's Gold Medal awardee for outstanding academic performance, for the year 2019. I worked on my undergraduate thesis with Prof. Kaushik Mitra at the Computational Imaging Lab in the domain of High Dynamic Range imaging.

  • Our paper, Visual Physics: Discovering Physical Laws from Videos, is up on ArXiv.
    Links: Paper, Webpage

  • Awarded the President of India Prize, Bharat Ratna M Visvesvaraya Memorial Prize and the Siemens Prize for outstanding academic performance from IIT Madras.
    Media and Links: IIT Madras, UCLA, The Hindu, Jagranjosh

  • Awarded the Graduate Dean's Scholars Award from the University of California, Los Angeles in 2019, in recognition of prior academic excellence.

My research interests lie at the intersection of applied deep learning and computer vision/computational imaging. My current interests and efforts are directed towards developing novel techniques to discover diverse physcial laws from videos, to be used for various vision and imaging applications. The field is nascent and holds great promise.

In the past, I have worked on more traditional imaging projects. As part of my undergraduate thesis, I worked on developing novel regimes for HDR imaging with new camera setups. I have also worked on minor projects in lensless imaging and camera calibration regimes for large camera networks.

* indicates equal contribution

Visual Physics: Discovering Physical Laws from Videos
Pradyumna Chari*, Chinmay Talegaonkar*, Yunhao Ba*, Achuta Kadambi
ArXiv, 2019
Project Webpage, Dataset

A novel pipeline that enables discovery of underlying parameters and equations from videos of physical phenomena.

Template credits: Jon Barron