Shaocong Dong

Papers

Total Citations

H-Index

About

Shaocong Dong is a researcher specializing in autonomous driving perception and multi-sensor fusion for 3D object detection. His work sits at the intersection of computer vision, robotics, and deep learning, with a particular focus on developing robust and accurate perception systems capable of navigating complex real-world environments. Dong's most notable contribution is FusionRCNN, a pioneering framework that addresses a critical limitation in existing 3D object detection pipelines. While state-of-the-art two-stage detectors traditionally rely solely on LiDAR point clouds for proposal refinement, Dong recognized the untapped potential of combining LiDAR's precise depth information with the rich semantic detail captured by cameras. FusionRCNN elegantly bridges this gap, enabling more accurate and reliable 3D object detection through complementary multi-modal fusion. The work has garnered 55 citations, reflecting strong uptake within the autonomous driving and robotics research communities. His research speaks directly to one of the field's most pressing challenges — building perception systems that are both precise and resilient under real-world conditions. For students and researchers exploring sensor fusion, autonomous vehicles, or 3D scene understanding, Dong's contributions offer a compelling foundation that continues to influence how the community approaches multi-sensor perception architectures.