The field of computer vision for sports analytics is rapidly evolving, with a focus on developing innovative methods for analyzing and understanding sports-related data. Recent research has explored the use of deep learning frameworks for converting chessboard images to Forsyth-Edwards Notation, enabling the provision of real-time analysis and suggestions for players. Additionally, there has been a surge in research on group activity recognition, with a particular emphasis on comparing the effectiveness of different modalities, such as video and tracking data, for recognizing group activities. Other notable areas of research include the development of multimodal datasets for play-by-play action spotting in soccer, and the creation of benchmarks for boxing action recognition and localization. These advances have the potential to revolutionize the field of sports analytics, enabling more accurate and detailed analysis of player and team performance. Noteworthy papers in this area include: CVChess, which presents a deep learning framework for converting chessboard images to Forsyth-Edwards Notation. Pixels or Positions, which introduces a multimodal dataset for group activity recognition and demonstrates the effectiveness of tracking-based approaches. FOOTPASS, which provides a benchmark for play-by-play action spotting in soccer. BoxingVI, which offers a comprehensive dataset for boxing action recognition and localization.