Xunjin Wu

Papers

Total Citations

H-Index

About

Xunjin Wu is a researcher whose work lies at the intersection of computer vision and robotics, with a particular focus on viewpoint estimation and domain adaptation. His most-cited paper, "Domain Adaptation for Viewpoint Estimation with Image Generation" (2021, 6 citations), addresses a critical bottleneck in robotic manipulation and grasping: the scarcity of accurately annotated training data for viewpoint estimation. Wu's key contribution is a novel framework that leverages image generation to bridge the domain gap between synthetic and real-world data, enabling more robust and generalizable viewpoint estimation models. This work is foundational for purposive perception and fine pose estimation—essential precursors for precise robotic interaction. By tackling data scarcity through generative techniques, Wu has advanced the practical deployment of vision systems in robotics. His research demonstrates a keen understanding of the challenges in real-world applications, offering scalable solutions that reduce reliance on costly manual annotations. For students and researchers in robotics and computer vision, Wu's work provides a compelling example of how domain adaptation can unlock new capabilities in autonomous systems.