Wanlong Li

Huawei Technologies (China), Huawei Technologies (Sweden)

Papers

Total Citations

133

H-Index

About

Wanlong Li is a leading researcher in 3D scene understanding for robotics and autonomous driving, with a focus on LiDAR-based perception. His work addresses critical challenges in how robots and autonomous vehicles interpret sparse, real-world point cloud data. Li’s major contributions include developing rotation-invariant neural networks for place recognition, exemplified by his highly cited paper “RINet” (67 citations), which enables robots to reliably identify previously visited locations regardless of viewing angle. He has also advanced semantic scene completion, proposing methods like the “Up-to-Down Network” (20 citations) that fuse multi-scale context to jointly estimate volumetric occupancy and semantic labels. A standout achievement is his work on semantic segmentation-assisted scene completion, which leverages semantic features to overcome LiDAR sparsity—a key hurdle in outdoor environments. With over 130 total citations across his top papers, Li’s research directly impacts the robustness and efficiency of autonomous systems, from navigation to obstacle avoidance. His innovative approaches to fusing geometry and semantics continue to shape the next generation of intelligent robotics.