Ying-Tian Liu
Papers
1
Total Citations
7
H-Index
1
About
Ying-Tian Liu is a rising researcher in computer vision, with a primary focus on self-supervised monocular depth estimation—a critical technology for autonomous driving and robotics. His most-cited work, "PPEA-Depth: Progressive Parameter-Efficient Adaptation for Self-Supervised Monocular Depth Estimation" (2024, 7 citations), tackles a fundamental limitation in the field: the static-scene assumption inherent in self-supervised methods, which degrades performance in dynamic environments. Liu introduces a novel progressive parameter-efficient adaptation framework that enables models to better handle moving objects and changing scenes without extensive retraining, significantly improving robustness and accuracy. This contribution addresses a key bottleneck in deploying self-supervised depth estimation in real-world, dynamic settings. With his work already gaining early traction, Liu is establishing himself as an innovator in efficient, adaptable vision systems. His research promises to advance the reliability of perception in autonomous systems, making him a notable figure to watch in the evolving landscape of self-supervised learning and 3D scene understanding.
Research Focus
Key Achievements
Top Papers
- 1