Papers
2
Total Citations
7
H-Index
2
About
Yi Mu’s research bridges the gap between intelligent robotics and computer vision, with a particular focus on enabling machines to perceive and interact with complex physical environments. His early work laid foundational methods in automated forestry, where he pioneered the use of BP neural networks for tree trunk recognition—a critical step toward robotising harvesting in China. This early contribution, cited 5 times, demonstrated how neural networks could extract colour markers and training samples to guide robotic manipulation in unstructured outdoor settings. More recently, Mu has advanced the frontier of articulated object manipulation with his groundbreaking work on ArtGS, a framework that extends 3D Gaussian Splatting (3DGS) to create interactive visual-physical models. This 2025 paper, already garnering 2 citations, tackles the long-standing challenge of enabling robots to reason about kinematic constraints during manipulation. By integrating visual and physical modelling, ArtGS allows robots to not only see but also physically understand and manipulate objects with complex joints. Mu’s trajectory from forestry automation to cutting-edge 3DGS-based interaction showcases a career dedicated to making robots more perceptive and dexterous—a vital step toward truly autonomous physical intelligence.
Research Focus
Key Achievements
Top Papers
- 1The trunk of the image recognition based on BP neural network5 citations · 2014
- 2