Yikang Ding

Papers

Total Citations

H-Index

About

Yikang Ding is a researcher advancing the field of autonomous driving and robotics through innovations in 3D object detection. His primary research focuses on point cloud perception, specifically enhancing grid-based detection methods that enable vehicles to interpret their surroundings from LiDAR data. In his most-cited work, "Enhancing Grid-Based 3D Object Detection in Autonomous Driving With Improved Dimensionality Reduction" (2023), Ding addresses a critical challenge in the field: the trade-offs inherent in projecting 3D point clouds into Bird’s Eye View (BEV) representations. While BEV allows leveraging mature 2D detection techniques, it inevitably loses valuable spatial information through dimensionality reduction. Ding’s contribution lies in developing improved dimensionality reduction strategies that preserve more geometric detail, thereby boosting detection accuracy without sacrificing computational efficiency. Although early in his career—with his top paper currently accumulating 4 citations—his work targets a fundamental bottleneck in real-world autonomous systems. By refining how point cloud data is compressed and interpreted, Ding is helping bridge the gap between theoretical detection frameworks and the robust, real-time performance demanded by self-driving vehicles, making him a promising voice in perception research.