Papers
9
Total Citations
138
H-Index
6
About
Jiamao Li is a leading researcher in robotic perception and scene understanding, whose work bridges the gap between computer vision and autonomous systems. His primary research areas include 3D semantic mapping, visual odometry, object pose estimation, and human-robot interaction. Li’s major contributions center on developing robust, real-time methods for robots to perceive and interact with dynamic environments. His most cited work, “RGB-D Semantic Segmentation and Label-Oriented Voxelgrid Fusion for Accurate 3D Semantic Mapping” (67 citations), introduces a novel fusion technique that enables task-driven robots to build detailed 3D semantic maps from RGB-D scans. He also advanced mobile robot localization with “Robust Stereo Visual Odometry Using Improved RANSAC-Based Methods” (18 citations), offering faster, more accurate motion estimation. Li’s impact is further demonstrated through his work on dynamic obstacle rejection for 3D map updating (14 citations) and category-level 6D object pose estimation (SD-Pose, 8 citations). His recent innovations include a pixel-wise voting network for single RGB image 6D object grasping and a topology-aware transformer for 3D hand pose estimation, showcasing his commitment to enabling intuitive human-robot collaboration.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Dynamic Obstacles Rejection for 3D Map Simultaneous Updating14 citations · 2018
- 4
- 5
- 6
- 7
- 8
- 9Self-supervised Scale Recovery for Decoupled Visual-inertial Odometry3 citations · 2023