Yebin Liu

Papers

Total Citations

H-Index

About

Yebin Liu is a leading researcher in computer vision and human motion analysis, with a focus on understanding and predicting human behavior in complex environments. Their key research areas include human motion prediction, scene-aware interaction modeling, and novel view synthesis for immersive applications. Liu’s most impactful contribution is the development of GIMO (Gaze-Informed Human Motion Prediction in Context), a pioneering framework that integrates gaze data to enhance the accuracy of human motion forecasts in real-world settings. This work, which has garnered 59 citations, addresses a critical challenge for assistive robotics and AR/VR systems by combining scene context with human intention cues, enabling safer and more intuitive human-machine interactions. Liu has also explored cutting-edge techniques in surgical imaging, such as Vision Transformer-based multiplane images for single-view view synthesis, demonstrating versatility in applying computer vision to medical domains. Their research bridges the gap between perceptual understanding and predictive modeling, with implications for autonomous systems, virtual reality, and human-robot collaboration. Liu’s work continues to shape how machines anticipate and respond to human actions in contextually rich environments.