Papers
2
Total Citations
87
H-Index
2
About
Fayao Liu is a leading researcher in computer vision and robotics, specializing in visual localization, embodied AI, and scene understanding. Her work addresses fundamental challenges in enabling machines to perceive and navigate dynamic environments with human-like robustness. Liu’s early landmark contribution, "Sequence searching with deep-learnt depth for condition- and viewpoint-invariant route-based place recognition" (2015, 57 citations), pioneered a deep learning approach to simultaneous extreme appearance and viewpoint changes—a problem that had stymied prior methods like FAB-MAP. This work laid critical groundwork for vision-based robot and vehicle localization under real-world conditions. More recently, Liu has advanced embodied AI through "SegEQA: Video Segmentation Based Visual Attention for Embodied Question Answering" (2019, 30 citations), introducing a novel video segmentation attention mechanism that enables agents to answer user questions by intelligently exploring and reasoning about their surroundings. This research has direct applications in autonomous driving and home robotics. Across her career, Liu’s work consistently bridges perception and action, earning recognition for tackling previously unsolved problems in place recognition and interactive AI. Her contributions continue to influence the development of more capable, context-aware autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2