Papers

2

Total Citations

21

H-Index

2

About

Teli Ma is a rising researcher at the forefront of embodied AI and autonomous perception, with key contributions in LiDAR semantic segmentation and robotic manipulation. Ma’s most influential work, “TFNet: Exploiting Temporal Cues for Fast and Accurate LiDAR Semantic Segmentation” (2024, 17 citations), introduces a novel framework that leverages temporal information to dramatically improve the speed and accuracy of 3D scene understanding—a critical capability for autonomous driving and robotics. This work addresses a fundamental bottleneck in real-time perception systems. Building on this, Ma’s recent research tackles the critical challenge of visual representation learning for robots, as seen in “Mitigating the Human-Robot Domain Discrepancy in Visual Pre-training for Robotic Manipulation” (2025, 4 citations). This work pioneers methods to bridge the gap between human-collected internet data and robot-specific domains, enabling more generalizable and sample-efficient policy learning. By addressing the scarcity of robot demonstration data through innovative pre-training strategies, Ma is helping to unlock more adaptable and capable embodied agents. With a clear trajectory from perception to manipulation, Teli Ma is shaping the future of how machines see and interact with the physical world.

Research Focus

Key Achievements

2
H-Index
2
Papers
21
Total Citations
11
Avg Citations/Paper
🏆 Most Cited Paper
TFNet: Exploiting Temporal Cues for Fast and Accurate LiDAR Semantic Segmentation
17 citations · 2024
📈 Most Prolific Year: 2024 (1 Papers)
🤝 Key Collaborators: 9
🏛 Institutions: University of Hong Kong, Hong Kong University of Science and Technology

Top Papers

  1. 1
  2. 2

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago