Huidan Han
Papers
1
Total Citations
12
H-Index
1
About
Huidan Han is a rising researcher in computer vision, with a primary focus on monocular 3D object detection—a challenging area that aims to infer 3D spatial information from a single 2D image. Her most notable contribution, "MonoSAID: Monocular 3D Object Detection based on Scene-Level Adaptive Instance Depth Estimation" (2023), introduces a novel framework that adaptively estimates depth at the instance level, significantly improving detection accuracy in complex driving scenes. This work has already garnered 12 citations, reflecting its timely impact on autonomous driving and robotics. Han’s research addresses a critical bottleneck in monocular perception: the ambiguity of depth from a single viewpoint. By proposing scene-level adaptive mechanisms, she enhances the robustness of depth estimation across varying environments, a key step toward safer and more reliable autonomous systems. Her work is particularly relevant for students and researchers interested in bridging the gap between 2D vision and 3D understanding, offering practical solutions that push the boundaries of what monocular systems can achieve.
Research Focus
Key Achievements
Top Papers
- 1