Papers
2
Total Citations
21
H-Index
1
About
Liqi Yan’s research lies at the intersection of embodied AI, multimodal perception, and 3D scene understanding, with a focus on enabling intelligent agents to navigate and interpret complex indoor environments. In their highly cited work on multimodal aggregation, Yan introduced the Memory Vision-Voice Indoor Navigation (MVV-IN) model, which integrates visual and auditory cues through a meta-learning framework to enhance robot navigation. This pioneering approach, garnering 20 citations, demonstrated how agents can dynamically fuse sensory streams for more robust, human-like interaction. More recently, Yan has advanced planar reconstruction techniques for indoor scenes, developing methods to infer 3D planar parameters—normals and offsets—from sparse image views and relative camera poses. Though newly published with 1 citation, this work holds transformative potential for digital heritage preservation, architectural design, and autonomous navigation. Yan’s contributions bridge the gap between low-level geometric reasoning and high-level semantic understanding, offering scalable solutions for real-world spatial intelligence. Their research continues to shape how machines perceive, remember, and act within built environments, making them a rising voice in computer vision and robotics.
Research Focus
Key Achievements
Top Papers
- 1
- 2