Sport Discipline

Sport discipline research focuses on leveraging computer vision and machine learning to analyze athletic performance and enhance training methodologies. Current research emphasizes developing robust deep learning models, including architectures like Hybrid Task Cascade and Set Transformers, to address challenges such as occlusion in instance segmentation and contextual classification of actions like sprints. These advancements enable automated, large-scale analysis of athletic movements, offering valuable insights for coaches, athletes, and researchers, improving performance and understanding of various sports.