Zhenpei Yang

Papers

Total Citations

H-Index

About

Zhenpei Yang is a computer vision researcher whose work focuses on 3D scene understanding, RGB-D perception, and pose estimation for robotics and graphics applications. His most influential contribution, "Extreme Relative Pose Estimation for RGB-D Scans via Scene Completion," addresses a critical challenge in the field: estimating the relative rigid pose between two RGB-D scans when they have minimal overlap. By leveraging scene completion techniques, Yang’s approach enables robust pose estimation even under extreme viewpoint changes—a task that traditional methods struggle with due to their reliance on substantial overlapping regions. This work has garnered 49 citations since 2019, reflecting its impact on advancing 3D reconstruction and robotic navigation. Yang’s research bridges the gap between computer vision and real-world deployment, offering practical solutions for environments where sensor data is sparse or fragmented. His contributions are particularly valuable for applications in autonomous systems, augmented reality, and large-scale 3D mapping, where accurate alignment of partial scans is essential. Through his innovative methodology, Yang has helped push the boundaries of what is possible in relative pose estimation, making him a notable figure in the 3D vision community.