AI for sports

As a rising domain in our daily activities and human culture, sports are gaining more and more attention, since there is a huge audience and economic value lurking behind it. Recently, deep learning methods have achieved remarkable performance compared with the best traditional methods while being significantly faster at inference time, making it possible to analyze the event in real time. With sufficient data, AI-based solutions can even be used to forecast the future performances of the current player or team. AI for sports special session pursues in-depth and comprehensive discussions among researchers from different fields.

This special session aims to follow up on the latest developments, research findings and trends on the following topics, including but not limited to:

• Group activity recognition
• Interaction recognition
• Action recognition
• Pose estimation
• Action quality evaluation
• Object detection
• Object tracking
• Highlight extraction for sports video
• Vision & Language
• Temporal action localization
• Sports analysis
• Transformer based methods

For authors:

1) Details at Call for Special Sessions
2) Paper templates and paper submission system at Instructions for Authors

Organizers:

• • Lifang Wu, Beijing University of Technology (lfwu@bjut.edu.cn)
  • Limin Wang, Nanjing University (lmwang@nju.edu.cn)
  • Xiangbo Shu, Nanjing University of Technology (shuxb@njust.edu.cn)