Yuanzheng Cai

Papers

Total Citations

H-Index

About

Yuanzheng Cai is a researcher whose work lies at the intersection of computer vision, 3D scene understanding, and multi-modal learning. His most influential contribution, "Adversarial unsupervised domain adaptation for 3D semantic segmentation with multi-modal learning" (2021), has garnered 53 citations, reflecting its impact on advancing robust perception systems. In this work, Cai pioneered a novel approach that integrates adversarial training with unsupervised domain adaptation, enabling 3D semantic segmentation models to generalize across different sensor modalities and environments without requiring labeled target data. This breakthrough is critical for real-world applications like autonomous driving and robotics, where labeled data is scarce and domain shifts are common. By fusing LiDAR and camera data through a multi-modal learning framework, Cai demonstrated how to bridge the gap between synthetic and real-world scenes, significantly improving segmentation accuracy in unfamiliar settings. His research not only pushes the boundaries of 3D vision but also provides practical solutions for deploying AI in dynamic, unlabeled environments. For students and researchers, Cai’s work offers a compelling example of how adversarial methods and multi-modal fusion can solve fundamental challenges in domain adaptation, making him a notable figure in the evolving field of 3D scene understanding.