Chan Kit Yan
Papers
2
Total Citations
69
H-Index
2
About
Chan Kit Yan is a leading researcher in 3D scene understanding and autonomous robotics, with a focus on building high-level environmental representations from sequential visual data. Her major contributions lie in advancing 3D scene graph prediction and reasoning, a critical technology for long-term robotic autonomy. In her highly cited 2025 work, "Hyperrectangle Embedding for Debiased 3D Scene Graph Prediction From RGB Sequences" (41 citations), she introduced a novel hyperrectangle embedding method that addresses bias in scene graph generation from RGB streams, significantly improving prediction accuracy. Her follow-up paper, "History-Enhanced 3D Scene Graph Reasoning From RGB-D Sequences" (28 citations), tackles the semantic gap in building rich representations from RGB-D data by incorporating temporal history, enabling more coherent and context-aware scene understanding. Together, these works have garnered nearly 70 citations, reflecting their immediate impact on the field. Chan’s research bridges computer vision and robotics, offering practical solutions for robots to perceive and reason about dynamic environments over time. Her innovative approaches to debiasing and history-enhanced reasoning mark her as a rising star in embodied AI and spatial intelligence.
Research Focus
Key Achievements
Top Papers
- 1
- 2History-Enhanced 3D Scene Graph Reasoning From RGB-D Sequences28 citations · 2025