Papers
3
Total Citations
24
H-Index
2
About
Junkang Chen is an emerging researcher specializing in skeleton-based action recognition, human-robot interaction (HRI), and sign language recognition, with a focus on developing advanced deep learning architectures for real-world applications. His work centers on designing innovative neural network frameworks that extract and leverage skeletal motion data to enable machines to interpret human movements with greater accuracy and efficiency. Chen's most recognized contribution, "TMS-Net" (2023), introduced a multi-feature, multi-stream, multi-level information sharing network for skeleton-based sign language recognition, earning 18 citations and demonstrating his ability to tackle complex, multi-dimensional recognition challenges. His research further extends into HRI through novel architectures such as the residual activation fish-shaped network, an imaginatively structured model comprising fish tail, body, and head components, designed to improve robustness in action classification. His work on real-time multi-feature sharing networks underscores a commitment to practical deployment in critical domains including emergency rescue, telemedicine, and industrial teleoperation. Though early in his citation trajectory, Chen's consistent output in 2023 across complementary topics signals a coherent and productive research agenda. Students and practitioners working at the intersection of computer vision, robotics, and human-computer interaction will find his architectural innovations particularly relevant and inspiring.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3