Yue Liao

Papers

Total Citations

H-Index

About

Yue Liao is a leading researcher in computer vision and multimodal AI, with a focus on human-centric scene understanding and fine-grained vision-language reasoning. Her most cited work introduces the concept of human-centric relation segmentation, a task that bridges object detection, semantic segmentation, and relational reasoning to enable machines to interpret complex, detail-rich commands—such as distinguishing objects in a person’s left versus right hand. This foundational contribution, published in 2021 with 10 citations, addresses a critical gap in robotic and interactive AI systems, where precise spatial and relational understanding is essential. Liao’s research advances the integration of vision and language, pushing beyond coarse recognition to fine-grained, context-aware interpretation. Her work is particularly impactful for applications in human-robot interaction, assistive technologies, and embodied AI. By tackling problems that require nuanced understanding of human-centric scenes, Liao has established herself as a key figure in developing more intuitive and capable AI systems that can operate effectively in real-world, human environments.