Max Robotics — Global Service Robot Directory + US Certification

About

Xianghui Xie is a computer vision researcher whose work centers on 3D human-object interaction tracking, scene reconstruction, and human motion understanding. His most significant contribution, **BEHAVE: Dataset and Method for Tracking Human Object Interactions** (2022), has accumulated over 146 citations and stands as a landmark resource in the field, providing a richly annotated benchmark that enables the community to model how humans interact with objects in natural, unconstrained environments — a capability critical to gaming, virtual and mixed reality, human behavior analysis, and human-robot collaboration. Building on this foundation, Xie has pushed the boundaries of monocular reconstruction with work such as *Visibility Aware Human-Object Interaction Tracking from Single RGB Camera* (2023), which addresses the challenge of maintaining consistent 3D spatial relationships across video frames using only a single RGB input. His more recent work, *PhySIC* (2025), further advances the field by tackling depth ambiguity, occlusion, and physically implausible contact in single-image human-scene reconstruction. Taken together, Xie's research meaningfully advances the realism and physical plausibility of 3D human understanding, making his contributions highly relevant to researchers working at the intersection of computer vision, robotics, and graphics.