Vijay Vasudevan
Papers
2
Total Citations
9,848
H-Index
2
About
Vijay Vasudevan is a leading researcher at the intersection of large-scale machine learning systems and 3D computer vision. He is best known as a principal architect of TensorFlow, the open-source deep learning framework that revolutionized the field. His seminal paper, *"TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems"* (2016), has amassed over 9,700 citations, establishing the foundational paradigm for expressing and executing machine learning algorithms across diverse hardware—from mobile devices to massive server clusters. This work has been instrumental in democratizing AI research and production deployment worldwide. More recently, Vasudevan has made significant contributions to 3D perception for robotics. In his highly innovative paper *"To the Point: Efficient 3D Object Detection in the Range Image with Graph Convolution Kernels"* (2021), he proposed a novel approach that learns 3D representations directly from 2D range images using graph convolution kernels. This work enables efficient, real-time 3D object detection, a critical capability for autonomous navigation and robotic manipulation. By bridging systems engineering with cutting-edge perception, Vasudevan’s research continues to shape both the infrastructure and the algorithms powering modern AI.
Research Focus
Key Achievements
Top Papers
- 1TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems9,777 citations · 2016
- 2