Shao-Kang Huang

Papers

Total Citations

H-Index

About

Shao-Kang Huang is a researcher at the forefront of computer vision and human-robot interaction, with a particular focus on 3D object pose estimation and human action recognition. His work addresses fundamental challenges in enabling machines to perceive and interact with their environment. Huang has made significant contributions to iterative pose refinement, developing novel frameworks that combine projection loss functions with discriminative refinement techniques to dramatically improve the accuracy of 3D object pose estimation from RGBD data—a critical capability for robotics and augmented reality applications. His 2020 and 2021 papers on this topic have each garnered 5 citations, reflecting their growing influence in the field. More recently, Huang has expanded into human motion analysis, introducing an innovative architecture that synergizes Long Short-Term Memory networks with depthwise separable convolutional neural networks for skeleton-based action recognition (2025, 6 citations). This work demonstrates his ability to bridge temporal modeling and efficient spatial feature extraction. Huang’s research continues to push the boundaries of how computers understand both static objects and dynamic human actions in three-dimensional space.