Papers
4
Total Citations
80
H-Index
4
About
Kun Zhang is a researcher whose work spans computer vision, robotics, and autonomous systems, with particular focus on 3D shape recognition, LiDAR-based mapping, and dynamic robot manipulation. His early contribution, "Multi-View CNN Feature Aggregation with ELM Auto-Encoder for 3D Shape Recognition" (2018), established his expertise in deep learning approaches to 3D visual understanding, accumulating 55 citations and demonstrating the value of combining multi-view convolutional features with efficient encoding architectures. Building on this foundation, Zhang has made significant strides in autonomous robotics infrastructure, developing DORF, a dynamic object removal framework that addresses the persistent challenge of ghost artifacts in urban LiDAR mapping — a problem directly impacting robot navigation and localization reliability. His more recent work pushes boundaries further: TNDF-Fusion tackles large-scale neural distance field mapping under real-world memory constraints, while TossNet introduces proprioceptive sensing solutions for high-speed, nonlinear robot throwing tasks. Collectively, Zhang's research reflects a coherent vision of making robotic perception and manipulation more robust, scalable, and deployable in complex real-world environments, positioning him as a rising contributor to the robotics and autonomous systems community.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4