Zixuan Tang
Papers
1
Total Citations
29
H-Index
1
About
Zixuan Tang is a rising researcher in computer vision and human-robot interaction, with a focus on advancing skeleton-based action recognition. Their most impactful work, the "Interactive Spatiotemporal Token Attention Network," tackles the challenging problem of general interactive action recognition—a critical capability for enabling seamless human-robot collaboration. Tang identified limitations in existing late fusion and co-attention methods, which either lack sufficient learning capacity or struggle to scale to multiple interacting entities. By introducing a novel token-based attention mechanism that jointly models spatial and temporal dynamics, Tang's approach achieves more robust and efficient recognition of complex interactions. This 2023 paper has already garnered 29 citations, reflecting its timely contribution to the field. Tang's research bridges the gap between theoretical attention models and practical robotic systems, offering solutions that are both computationally efficient and highly adaptive. Their work is particularly notable for its potential to enhance real-time human-robot interaction in dynamic environments, from collaborative manufacturing to assistive technologies. As an emerging voice in interactive AI, Tang continues to push the boundaries of how machines understand and respond to human actions.
Research Focus
Key Achievements
Top Papers
- 1