A Video-Based Framework for Automatic 3D Localization of Multiple Basketball Players: A Combinatorial Optimization Approach

Sports complexity must be investigated at competitions; therefore, non-invasive methods are essential. In this context, computer vision, image processing, and machine learning techniques can be useful in designing a non-invasive system for data acquisition that identifies players' positions in official basketball matches. Here, we propose and evaluate a novel video-based framework to perform automatic 3D localization of multiple basketball players. The introduced framework comprises two parts. The first stage is player detection, which aims to identify players' heads at the camera image level. This stage is based on background segmentation and on classification performed by an artificial neural network. The second stage is related to 3D reconstruction of the player positions from the images provided by the different cameras used in the acquisition. This task is tackled by formulating a constrained combinatorial optimization problem that minimizes the re-projection error while maximizing the number of detections in the formulated 3D localization problem.