Zecheng Yu
Papers
1
Total Citations
12
H-Index
1
About
Zecheng Yu is a rising researcher in computer vision and embodied AI, with a primary focus on understanding human-object interactions from egocentric perspectives. His work centers on the concept of object affordance—the action possibilities that objects offer based on human motor capabilities and physical properties. Yu’s most cited paper, "Fine-grained Affordance Annotation for Egocentric Hand-Object Interaction Videos" (2023, 12 citations), introduces a novel annotation framework that captures nuanced, temporally-aware affordance labels in first-person video. This contribution addresses a critical gap in the field, enabling more accurate action anticipation and robot imitation learning by grounding AI systems in real-world human behavior. By providing fine-grained affordance data, Yu’s work directly supports downstream tasks in robotics and activity recognition, where understanding not just what an object is, but what can be done with it, is essential. His research bridges perception and action, offering a foundational resource for developing intelligent agents that can learn from and interact with their environment as humans do. As an early-career researcher, Yu’s focused contributions to affordance understanding signal a promising trajectory in the intersection of computer vision, cognitive science, and robotics.
Research Focus
Key Achievements
Top Papers
- 1