Papers
4
Total Citations
161
H-Index
3
About
Di Xie is a leading researcher at the intersection of computer vision, surgical robotics, and lifelong machine learning. His work addresses critical challenges in enabling autonomous systems to perceive, learn, and adapt in real-world environments. Xie’s most impactful contribution is the HeiChole benchmark (96 citations), which provides the first standardized validation of machine learning algorithms for surgical workflow and skill analysis—a foundational step toward context-aware cognitive surgical assistants that can improve patient safety and surgical training. He has also pioneered methods for source-free unsupervised domain adaptation (51 citations), allowing neural networks to adapt to new visual domains without access to original training data, a breakthrough for practical robot vision. Demonstrating leadership in the field, Xie co-organized the IROS 2019 Lifelong Robotic Vision Challenge, which attracted over 150 teams worldwide and established the OpenLORIS benchmark for continual object recognition. Through these contributions, Xie is shaping the future of robots that can learn continuously, adapt autonomously, and assist in high-stakes environments like the operating room.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4