Siyuan Xiang
Papers
5
Total Citations
80
H-Index
4
About
Siyuan Xiang is a researcher at the intersection of robotics, computer vision, and construction automation, whose work pushes the boundaries of how machines perceive and interact with the built environment. His primary research areas include augmented reality for collaborative robotics, spatial reasoning in artificial intelligence, and multi-sensory perception for robotic manipulation. Xiang’s most impactful contribution is his pioneering work on mobile projective augmented reality for construction co-robots, which has garnered 48 citations and lays the groundwork for intuitive human-robot collaboration on dynamic job sites. He also introduced the SPARE3D dataset (18 citations), a benchmark designed to test deep networks’ capacity for spatial reasoning from three-view line drawings—a fundamental challenge in mimicking human visual intelligence. Beyond these, Xiang has explored through-wall object recognition and pose estimation, enabling robots to perceive occluded environments, and developed a multi-sensory pouring network that integrates audio and visual cues for granular media manipulation, inspired by human dexterity. His work, presented at venues like the International Symposium on Automation and Robotics in Construction (ISARC), demonstrates a consistent drive to equip robots with robust, human-like perception and reasoning capabilities for real-world applications.
Research Focus
Key Achievements
Top Papers
- 1Mobile projective augmented reality for collaborative robots in construction48 citations · 2021
- 2SPARE3D: A Dataset for SPAtial REasoning on Three-View Line Drawings18 citations · 2020
- 3Towards Mobile Projective AR for Construction Co-Robots6 citations · 2019
- 4Through-Wall Object Recognition and Pose Estimation4 citations · 2019
- 5