Sourabh Vora

Papers

Total Citations

294

H-Index

About

Sourabh Vora is a researcher specializing in 3D perception, autonomous driving, and multi-modal sensor fusion for robotics systems. He is perhaps best known as a lead contributor to **PointPillars** (2019), a landmark paper that revolutionized how point clouds are encoded for object detection tasks. By introducing a fast and efficient pillar-based encoding scheme, PointPillars achieved real-time inference speeds without sacrificing accuracy, making it one of the most widely adopted architectures in the autonomous driving community — a testament reflected in its remarkable 241 citations. His follow-up work, **PointPainting** (2020), tackled a long-standing challenge in sensor fusion: despite the complementary strengths of cameras and lidar, fusion methods had struggled to outperform lidar-only approaches. PointPainting offered an elegant sequential fusion strategy that "paints" lidar point clouds with semantic image features, meaningfully closing this performance gap and garnering 53 citations. Together, these contributions have had a tangible influence on how the field approaches perception pipelines for self-driving vehicles, establishing Vora as a significant voice in 3D object detection and sensor fusion research.