My research interests are at the intersection of computer vision, machine learning and optimization. My research goal is to develop autonomous systems to enable machines to understand and reason about human action and behavior allowing for richer interaction and seamless control. With this view, my current research focuses on detecting, tracking and reconstructing human pose from images and monocular video sequences.
Convolutional Pose Machines
Shih-En Wei, Varun Ramakrishna, Takeo Kanade, Yaser Sheikh
Preprint on arxiv, 2016.
Predicting Multiple Structured Visual Interpretations
Debadeepta Dey, Varun Ramakrishna, Martial Hebert, J.A. Bagnell.
In ICCV 2015.
PoseMachines: Articulated Pose Estimation via Inference Machines
Varun Ramakrishna, Daniel Munoz, Martial Hebert, J.A. Bagnell, Yaser Sheikh.
In ECCV 2014 (Oral presentation).
User-Specific Hand Modeling from Monocular Depth Sequences
Jonathan Taylor, Richard Stebbing, Varun Ramakrishna, Cem Keskin, Jamie Shotton, Shahram Izadi, Andrew Fitzgibbon, Aaron Hertzmann.
In CVPR 2014.
Tracking Human Pose by Tracking Symmetric Parts
Varun Ramakrishna, Yaser Sheikh, Takeo Kanade.
In CVPR 2013.
Reconstructing 3D Human Pose from 2D Image Landmarks
Varun Ramakrishna, Takeo Kanade, Yaser Sheikh.
In ECCV 2012.
Mode Marginals: Expressing Uncertainty in MRFs via Diverse M-Best Solutions
Varun Ramakrishna, Dhruv Batra.
In NIPS 2012 Workshop on Perturbations, Optimization, Statistics.
Multi-Pose Multi-Target Tracking for Human Activity Understanding
Hamid Izadinia, Varun Ramakrishna, Kris Kitani, Daniel Huber.
In WACV 2013.