Rahul Venkatesh

I am a CS PhD student at Stanford starting Fall'21. Previously, I graduated from CMU with a Master's in Computer Vision where I was advised by Prof. Laszlo Jeni from the Robotics Institute and Dr. Maneesh Singh from Verisk AI. I also closely collaborated with Prof. Zico Kolter from the CS department at CMU. Prior to that, I was a research assistant at the Video Analytics Lab at the Indian Institute of Science, Bangalore, where I was advised by Prof. R Venkatesh Babu.

Email  /  Google Scholar  /  Resume

profile photo
Research

With a rich implementation and research skill set attained from my prior work and publications in the field of unsupervised learning and cognitively inspired computer vision, I have developed a focussed goal of improving data-efficiency and generalizability of machine learning systems.

3D Reconstruction

Deep Implicit Surface Point Prediction Networks
Rahul Venkatesh, Tejan Karmali, Sarthak Sharma, Aurobrata Ghosh, R. Venkatesh Babu, Laszlo Jeni, Maneesh Singh
ICCV, 2021

Deep Unsigned Distance Embeddings for Hi-Fidelity Representation of Complex 3D Surfaces
Rahul Venkatesh, Sarthak Sharma, Aurobrata Ghosh, Laszlo Jeni, Maneesh Singh
arXiv, 2020

Appearance Consensus Driven Self-Supervised Human Mesh Recovery
Jogendra Nath Kundu*, Rakesh Mugalodi*, Varun Jampani, Rahul Venkatesh, R Venkatesh Babu
ECCV (oral), 2020

Adversarial Machine Learning

Semantic Adversarial Robustness with Differentiable Ray-Tracing
Rahul Venkatesh, Eric Wong, J Zico Kolter
DiffCVGP Workshop at NeurIPS, 2020

Object Pose Estimation

Object pose estimation from monocular image using multi-view keypoint correspondence
Rahul Venkatesh*, Jogendra Nath Kundu*, Aditya Ganeshan*, R Venkatesh Babu
Geometry Meets Deep Learning workshop at ECCV, 2018

iSPA-Net: Iterative Semantic Pose Alignment Network
Rahul Venkatesh*, Jogendra Nath Kundu*, Aditya Ganeshan*, Aditya Prakash, R Venkatesh Babu
ACMMM, 2018

Human Pose Estimation

Kinematic-Structure-Preserved Representation for Unsupervised 3D Human Pose Estimation
Rahul Venkatesh*, Jogendra Nath Kundu*, Siddharth Seth*, Rakesh Mugalodi, R Venkatesh Babu
AAAI (oral), 2020

Cross-Dataset Adaptation via Probabilistic Amodal 3D Human Pose Completion
Rahul Venkatesh*, Jogendra Nath Kundu*, Jay Patravali*, R Venkatesh Babu
WACV, 2020

Unsupervised Cross-Modal Alignment for Multi-Person 3D Pose Estimation
Jogendra Nath Kundu* Revanur Ambareesh*, Govind Vitthal Waghmare, Rahul Venkatesh, R Venkatesh Babu
ECCV, 2020

Cross-Conditioned Recurrent Networks for Long-Term Synthesis of Inter-Person Human Motion Interactions
Jogendra Nath Kundu*, Himanshu Buckchash*, Priyanka Mandikal, Anirudh Jamkhandi, Rahul Venkatesh, R Venkatesh Babu
WACV, 2020

Unsupervised, Source-Free Domain Adaptation

Class-Incremental Domain Adaptation
Rahul Venkatesh*, Jogendra Nath Kundu*, Naveen Venkat, R Venkatesh Babu
ECCV, 2020

Universal Source-Free Domain Adaptation
Jogendra Nath Kundu*, Naveen Venkat*, Rahul Venkatesh, R Venkatesh Babu
CVPR, 2020

Towards Inheritable Models for Open-Set Domain Adaptation
Jogendra Nath Kundu*, Naveen Venkat*, Revanur Ambareesh, Rahul Venkatesh, R Venkatesh Babu
CVPR (oral), 2020

Data-Efficient Object Tracking

Siamese network for underwater multiple object tracking
Rahul Venkatesh, Revanur Ambareesh, Shobha G
ICMLC (oral), Best Presentation Award, 2020

Traffic-flow Optimization

An adaptive model for traffic flow optimisation in dynamic environments
Rahul Venkatesh, Rajashree Shettar, KN Subramanya
International Journal of Computational Science and Engineering, 2019

* => (Equal Contribution)

Source: Jon Barron -> template