Alex Lang

Papers

Total Citations

294

H-Index

About

Alex Lang is a leading researcher in 3D perception for autonomous driving and robotics, with a focus on efficient and robust object detection from point clouds. His most influential contribution, **PointPillars** (2019, 241 citations), introduced a fast encoder that converts raw point cloud data into a format suitable for 2D convolutional neural networks, dramatically accelerating lidar-based detection without sacrificing accuracy. This work became a foundational baseline for real-time autonomous driving systems. Lang also advanced sensor fusion with **PointPainting** (2020, 53 citations), a sequential method that projects semantic segmentation from camera images onto lidar point clouds, enabling richer feature representations for 3D detection. His research addresses the critical challenge of fusing complementary sensor modalities—camera and lidar—to improve reliability in self-driving cars. Lang’s work is widely cited in both academia and industry, shaping modern perception stacks for autonomous vehicles. His contributions exemplify how practical, well-engineered solutions can push the boundaries of real-time 3D scene understanding.