R. Venkatesh Babu

About

R. Venkatesh Babu is a leading researcher in computer vision and deep learning, with a focus on feature extraction, human motion modeling, and adversarial robustness. His early work on GPU-accelerated implementations of the Scale Invariant Feature Transform (SIFT) — including a real-time version with 37 citations — significantly advanced the speed and practicality of keypoint-based methods for object detection, tracking, and large-scale image retrieval. These contributions remain foundational for real-time vision systems. More recently, Babu has tackled the complex challenge of long-term human motion synthesis, introducing Cross-Conditioned Recurrent Networks (21 citations) to model inter-person motion interactions for applications in animation, human-robot interaction, and surveillance. His work on adversarial robustness, such as Feature Level Stochastic Smoothing (4 citations), addresses critical vulnerabilities in deep neural networks for safety-critical domains like autonomous navigation. Through these diverse contributions — spanning efficient feature extraction, generative sequence modeling, and robust deep learning — Babu has demonstrated a sustained impact on both foundational computer vision and emerging applications in robotics and AI safety.

Research Focus

Key Achievements

Top Papers

1. 1
   A real-time implementation of SIFT using GPU

   37 citations · 2014
2. 2
   Cross-Conditioned Recurrent Networks for Long-Term Synthesis of Inter-Person Human Motion Interactions

   21 citations · 2020
3. 3
   Speeding up SIFT using GPU

   11 citations · 2013
4. 4
   Boosting Adversarial Robustness using Feature Level Stochastic Smoothing

   4 citations · 2021

Key Collaborators

Contact & Links