MEGA Hub
New

PL-NBA: A Possession-level Universal Basketball Video Dataset Supporting Multiple Visual Understanding Tasks

Published on Aug 20cs.CV

Authors

Do you know Yunhao Zhao?You can claim authorship or link another user.Do you know Haoying Sun?You can claim authorship or link another user.Do you know Jiarui Li?You can claim authorship or link another user.Do you know Zhuming Wang?You can claim authorship or link another user.Do you know Ya Jing?You can claim authorship or link another user.Do you know Xiangbo Shu?You can claim authorship or link another user.Do you know Lifang Wu?You can claim authorship or link another user.Do you know Changwen Chen?You can claim authorship or link another user.

Abstract

Visual understanding in sports has emerged as a hot topic in computer vision in recent years. Most existing basketball video datasets adopt single action or activity as sample, which can neither preserve the temporal continuity of game events nor support complex tasks such as action anticipation. To address this issue, this paper constructs the first possession-level basketball video dataset (PL-NBA), in which each sample is composed of a complete NBA offensive possession. Collected from 60 NBA games, PL-NBA contains 11,000 valid offensive possession clips and 31,567 annotated events with player names, captions, event types and timestamps. Each video clip includes multiple events and preserves the continuity of events, which is helpful for analysis of tactic. Experiment is conducted on multiple visual understanding tasks, including event recognition, video captioning, temporal action localization and action anticipation. Experimental results show that existing methods achieve limited performance on above four tasks, demonstrating that PL-NBA is a challenging benchmark for sports video understanding.

Community

00