- I used to do competitive coding back in my undergrad (expert in codeforces, 5-star in codechef). I participated in ACM ICPC India Regionals in 2017 and 2018. I still ocassionally like to ponder over some tough problems.
- I love movies - especially the ones with spectacular visuals and/or slick action and/or high concept.
- Being a Gujarati, I love Garba.
- A few more memorable and fun projects:
- Writing a compiler and interpreter from scratch for a made-up language called flatB: github
- Content-aware image resizing: github
- AI-agent to play ultimate Tic-Tac-Toe: github. It was fun to play different agents against each other.
- Training RL agents on Atari games: github
- Implementing a super-fun game with C++ OpenGL: github
Chaitanya Patel
CS PhD student at Stanford University
chpatel <at> cs <dot> stanford <dot> edu
I am a Computer Science PhD student at Stanford University, advised by Ehsan Adeli and Juan Carlos Niebles. I completed my undergrad with Gold Medal from IIIT Hyderabad, majoring Computer Science and Engineering. My research interest spans 3D computer vision, self-supervised learning and machine learning. I have been fortunate to work with Shih-En Wei, Varun Gulshan, Gerard Pons-Moll and Avinash Sharma as well.
Experience
Research Scientist InternMeta Reality Lab, Pittsburgh, USAJun 2022 - Sep 2022Jun 2023 - Sep 2023
Worked with Shih-En Wei on fast registration of photorealistic 3D avatars for VR facial animation.
Our work was accepted at ECCV 2024.
Predoc ResearcherGoogle Research, IndiaJan 2020 - Aug 2021
Worked with Varun Gulshan on Self-supervised learning with large-scale satellite data for earth observation applications like landcover mapping.
Research InternMPI Informatics, GermanyMay 2019 - Nov 2019
Worked with Gerard Pons-Moll to develop a data-driven garment model which is realistic, differentiable and easy to animate.
Our work TailorNet was accepted for Oral presentation at CVPR 2020.
Honours StudentIIIT Hyderabad, IndiaMay 2017 - Apr 2019
Worked under the guidance of Avinash Sharma. Contributed to the development of 3D Surface Capture system using Microsoft Kinect depth sensors. Co-authored a publication HumanMeshNet at ICCV 2019 Workshop 3DRW.
Publications
UniEgoMotion: A Unified Model for Egocentric Motion Reconstruction, Forecasting, and Generation
ICCV 2025
Chaitanya Patel, Hiroki Nakamura, Yuta Kyuragi, Kazuki Kozuka, Juan Carlos Niebles, Ehsan Adeli
AdaVid: Adaptive Video-Language Pretraining
CVPRW 2025
Chaitanya Patel, Juan Carlos Niebles, Ehsan Adeli
Fast Registration of Photorealistic Avatars for VR Facial Animation
ECCV 2024
Chaitanya Patel, Shaojie Bai, Te-Li Wang, Jason Saragih, Shih-En Wei
Evaluating Self and Semi-Supervised Methods for Remote Sensing Segmentation Tasks
Chaitanya Patel*, Shashank Sharma*, Valerie J. Pasquarella, Varun Gulshan
TailorNet: Predicting Clothing in 3D as a Function of Human Pose, Shape and Garment Style
CVPR 2020 (ORAL)
Chaitanya Patel*, Zhouyingcheng Liao*, Gerard Pons-Moll
HumanMeshNet: Polygonal Mesh Recovery of Humans
ICCV-3DRW 2019
Abbhinav Venkat, Chaitanya Patel, Yudhik Agrawal, Avinash Sharma
* Equal contribution
Teaching
Stanford - Spring 2025
CS231N (Head CA) - Deep Learning for Computer Vision
Stanford - Spring 2024
CS231N - Deep Learning for Computer Vision
Stanford - Winter 2024
CS228 - Probabilistic Graphical Models
Stanford - Fall 2023
CS157 - Computational Logic
IIITH - Spring 2019
CSE578 - Computer Vision
IIITH - Monsoon 2018
CSE471 - Statistical Methods in AI
IIITH - Spring 2018
ICS103 - Data Structures
IIITH - Monsoon 2017
ICS101 - Computer Programming
Other Stuff