Papers
2
Total Citations
8
H-Index
2
About
Xiuquan Qiao is a leading researcher in computer vision and robotics, specializing in 3D perception, articulated object modeling, and category-level pose estimation. His major contributions lie in developing novel frameworks that enable machines to understand and interact with complex real-world objects from limited visual data. Notably, his work on "DTF-Net" (2023, 6 citations) introduced a deformable template field for simultaneous 6D pose estimation and 3D shape reconstruction from RGB-D images, tackling the challenge of open-world object variations without relying on category-specific templates. More recently, his "SM³" framework (2024, 2 citations) pioneered a self-supervised, multi-task approach to reconstruct articulated objects and estimate their movable joint structures using only multi-view 2D images, moving beyond traditional supervised methods constrained by annotated datasets. This work is pivotal for robotics, enabling autonomous systems to model and manipulate objects like cabinets or doors without prior category knowledge. Qiao’s research pushes the boundaries of generalizable 3D understanding, offering scalable solutions for robotic manipulation and augmented reality. His innovative use of self-supervision and deformable representations marks him as a rising star in the field, with high potential for future impact on embodied AI and autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2