Yuyin Sun

Papers

Total Citations

H-Index

About

Yuyin Sun is a pioneering researcher in robot perception and human-robot interaction, whose work centers on enabling robots to identify and understand objects through natural language descriptions. Her key research areas include object recognition, multimodal scene understanding, and lifelong learning for robotics. Sun's most influential contribution, "Attribute based object identification" (2013, 53 citations), broke new ground by bridging computer vision and robotics—showing how robots could identify objects using language-based attributes rather than just visual features, a critical step toward more intuitive human-robot collaboration. She further advanced this with "Learning to identify new objects" (2014, 8 citations), which tackled the challenge of robots recognizing novel objects described in everyday language. In her more recent work, "Multimodal Neural Radiance Field" (2023, 16 citations), Sun pushes the frontier of scene reconstruction by integrating multiple sensory modalities into NeRF representations for richer robot perception. Her "NEOL: Toward Never-Ending Object Learning for robots" (2016, 7 citations) introduced a paradigm-shifting framework for continuous, lifelong object learning—enabling robots to expand their knowledge incrementally without forgetting previously learned concepts. Sun's research directly addresses the fundamental challenge of making robots truly useful in dynamic, human-centered environments, where they must understand not just what objects look like, but what they are called and how people refer to them.