Hung‐Ju Liao
Papers
1
Total Citations
33
H-Index
1
About
Hung-Ju Liao is a leading researcher in computer vision and 3D scene understanding, with a focus on articulated object modeling and neural rendering. His most influential work, **CLA-NeRF** (Category-Level Articulated Neural Radiance Field, 2022, 33 citations), introduces a groundbreaking framework that learns to represent and manipulate articulated objects—such as cabinets or drawers—without requiring CAD models or depth data. By training solely on RGB images with known camera poses, CLA-NeRF achieves simultaneous view synthesis, part segmentation, and articulated pose estimation at the object category level, a significant leap toward generalizable 3D perception. This work bridges neural radiance fields with robotic manipulation and augmented reality, enabling machines to understand object structure and motion from limited visual input. Liao’s contributions are pivotal for advancing interactive 3D vision, reducing reliance on expensive 3D annotations, and paving the way for scalable, category-level understanding of dynamic environments. His research continues to inspire new directions in self-supervised learning for articulated objects, making him a rising voice in the field.
Research Focus
Key Achievements
Top Papers
- 1CLA-NeRF: Category-Level Articulated Neural Radiance Field33 citations · 2022