Difei Gao
Papers
3
Total Citations
43
H-Index
3
About
Difei Gao is a researcher at the forefront of embodied AI and human-robot interaction, with a focus on affordance learning and egocentric vision. Her work bridges the gap between human demonstration and robotic execution, enabling intelligent systems to understand and replicate hand-object interactions from video. Gao’s most-cited paper, "Affordance Grounding from Demonstration Video to Target Image" (2023, 22 citations), introduces a novel framework that transfers affordance knowledge from expert demonstrations to novel target images—a critical step for AR assistants and service robots. She also developed the AssistQ dataset (2022, 15 citations), which pioneers affordance-centric, question-driven task completion for egocentric assistants, allowing systems to answer user queries about how to interact with objects. In her work on GazeVQA (2023, 6 citations), Gao advanced video question answering by integrating multiview eye-gaze data, enhancing task-oriented collaboration between humans and robots. With a growing citation impact, Gao’s research is shaping how machines learn from human behavior, making her a rising leader in interactive AI and computer vision.
Research Focus
Key Achievements
Top Papers
- 1Affordance Grounding from Demonstration Video to Target Image22 citations · 2023
- 2
- 3