Yuexin Ma
Papers
8
Total Citations
60
H-Index
5
About
Yuexin Ma is pioneering the frontier of human-centric 4D scene understanding, seamlessly blending wearable sensing, LiDAR, and multi-modal AI to capture and interpret dynamic human-environment interactions at scale. Her landmark work, *HSC4D* (2022, 24 citations), introduced a wearable IMU and LiDAR system for creating large-scale, interactive digital worlds—a breakthrough for virtual reality and embodied AI. Ma has also redefined biometric security through gait recognition in free environments using single LiDAR (2022, 10 citations; 2024, 6 citations), enabling non-intrusive identification in smart homes and healthcare. Her recent contributions extend to robotic dexterity with *DexGrasp Anything* (2025, 5 citations), a physics-aware framework for universal grasping, and *SemGeoMo* (2025, 4 citations), which generates contextually aware human motions for virtual and physical robots. Ma’s work on *WildRefer* (2024, 5 citations) advances 3D object localization in dynamic scenes using natural language, while *HUNTER* (2024, 4 citations) tackles unsupervised human detection via synthetic-to-real transfer. With over 60 citations across her most influential papers, Ma is shaping the future of human-centered AI, robotics, and immersive media.
Research Focus
Key Achievements
Top Papers
- 1
- 2Gait Recognition in Large-scale Free Environment via Single LiDAR10 citations · 2022
- 3Gait Recognition in Large-scale Free Environment via Single LiDAR6 citations · 2024
- 4
- 5
- 6
- 7
- 8