Sport Understanding

Sport understanding, a burgeoning field within artificial intelligence, aims to enable computers to comprehend and reason about complex sports scenarios from various data modalities, including text, video, and audio. Current research focuses on developing and evaluating large language models specifically trained on sports data, employing architectures like RWKV, and improving video analysis techniques through event-based cameras and advanced multi-object tracking algorithms such as Deep HM-SORT and Deep-EIoU. These advancements are driving progress in applications ranging from automated highlight generation and personalized fan experiences to enhanced coaching strategies and medical diagnostics in sports.

Papers