Yukang Huo

Papers

Total Citations

H-Index

About

Yukang Huo is a rising researcher at the forefront of computer vision and embodied intelligence, whose work bridges the gap between 3D scene understanding and robotic manipulation. His primary research areas include neural rendering, articulated object perception, and robust pose estimation. Huo’s major contributions are twofold: he has advanced the field of Neural Radiance Fields (NeRF) by providing a comprehensive review that synthesizes progress in 3D scene understanding and new view synthesis, and he has pioneered novel methods for handling the complex kinematics of articulated objects. Notably, his 2025 work on "Generalizable Articulated Object Perception with Superpoints" introduces a superpoint-based approach that dramatically improves part segmentation in 3D point clouds, a critical step for precise robotic manipulation. Additionally, his integrated differentiable rendering technique for category-level articulation pose estimation tackles the persistent challenges of kinematic constraints and self-occlusion. With his most-cited papers accumulating citations within just a year of publication, Huo’s impact is already evident. His research is directly shaping how robots interact with their environments, making him a key figure to watch in the evolution of embodied AI and 3D vision.