Towards Comprehensive Basketball Understanding

• URL: http://arxiv.org/abs/2608.23435v1
• Date: Mon, 24 Aug 2026 16:09:21 GMT
• Title: Towards Comprehensive Basketball Understanding
• Abstract summary: Basketball understanding requires recognizing events, localizing actions, identifying players, and relating these to structured game knowledge.<n>We introduce BasketballBench, a multimodal benchmark comprising 7,980 questions across ten tasks in text, image, and video.<n>We propose BasketballSkills, an agent that composes eight basketball-specific perception and retrieval tools under four reusable skills that specify tool order, evidence bindings, and stopping conditions.
• Score: 46.9044032598676
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Understanding a basketball game requires recognizing events, localizing actions, identifying players, and relating these to structured game knowledge. Existing benchmarks primarily evaluate these abilities one at a time, leaving the interactions among these abilities under-explored. We introduce BasketballBench, a multimodal benchmark comprising 7,980 questions across ten tasks in text, image, and video. It is built from the 2025-2026 NBA season and includes official playby-play, rosters and profiles for 530 active players, and 2,501 possession-level broadcast clips. We further propose BasketballSkills, an agent that composes eight basketball-specific perception and retrieval tools under four reusable skills that specify tool order, evidence bindings, and stopping conditions. Experiments show that current MLLMs struggle particularly on questions requiring the integration of multiple capabilities, whereas BasketballSkills outperforms them, highlighting the effectiveness of explicitly composing domain-specific capabilities for comprehensive basketball understanding.

Related papers

• PL-NBA: A Possession-level Universal Basketball Video Dataset Supporting Multiple Visual Understanding Tasks [53.77958773620005]
  This paper constructs the first possession-level basketball video dataset (PL-NBA)<n>PL-NBA contains 11,000 valid offensive possession clips and 31,567 annotated events with player names, captions, event types and timestamps.<n>Experiment is conducted on multiple visual understanding tasks, including event recognition, video captioning, temporal action localization and action anticipation.
  arXiv  Detail & Related papers  (2026-08-20T05:29:05Z)
• BasketEvent: Understanding Who Did What and When in Basketball Videos [51.926771112716686]
  We introduce BasketEvent, a player- centric basketball event understanding dataset curated from real NBA broadcasts.<n>We propose PlayNet, a player-centric reasoning framework that maps basketball videos to player-level event predictions.<n>Our data, code, and models will be made publicly available.
  arXiv  Detail & Related papers  (2026-07-23T12:39:56Z)
• See It Before You Grab It: Deep Learning-based Action Anticipation in Basketball [0.0]
  This work introduces the task of action anticipation in basketball broadcast videos, focusing on predicting which team will gain possession of the ball following a shot attempt.<n>To benchmark this task, a new self-curated dataset comprising 100,000 basketball video clips, over 300 hours of footage, and more than 2,000 manually annotated rebound events is presented.<n> Comprehensive baseline results are reported using state-of-the-art action anticipation methods, representing the first application of deep learning techniques to basketball rebound prediction.
  arXiv  Detail & Related papers  (2025-12-17T12:39:31Z)
• BASKET: A Large-Scale Video Dataset for Fine-Grained Skill Estimation [23.088610274138848]
  BASKET contains 4,477 hours of video capturing 32,232 basketball players from all over the world.<n>Our dataset includes a massive number of skilled participants with unprecedented diversity in terms of gender, age, skill level, geographical location.
  arXiv  Detail & Related papers  (2025-03-26T17:59:02Z)
• SPORTU: A Comprehensive Sports Understanding Benchmark for Multimodal Large Language Models [15.062299319625701]
  SPORTU is a benchmark designed to assess Multimodal Large Language Models (MLLMs) across multi-level sports reasoning tasks.<n>SPORTU comprises two key components: SPORTU-text, featuring 900 multiple-choice questions with human-annotated explanations for rule comprehension and strategy understanding.<n>SPORTU-video consists of 1,701 slow-motion video clips across 7 different sports and 12,048 QA pairs, designed to assess multi-level reasoning.
  arXiv  Detail & Related papers  (2024-10-11T02:58:38Z)
• SkillMimic: Learning Basketball Interaction Skills from Demonstrations [85.23012579911378]
  We introduce SkillMimic, a unified data-driven framework that fundamentally changes how agents learn interaction skills.<n>Our key insight is that a unified HOI imitation reward can effectively capture the essence of diverse interaction patterns from HOI datasets.<n>For evaluation, we collect and introduce two basketball datasets containing approximately 35 minutes of diverse basketball skills.
  arXiv  Detail & Related papers  (2024-08-12T15:19:04Z)
• Knowledge Guided Entity-aware Video Captioning and A Basketball Benchmark [49.54265459763042]
  We construct a multimodal basketball game knowledge graph (KG_NBA_2022) to provide additional knowledge beyond videos. Then, a dataset that contains 9 types of fine-grained shooting events and 286 players' knowledge is constructed based on KG_NBA_2022. We develop a knowledge guided entity-aware video captioning network (KEANet) based on a candidate player list in encoder-decoder form for basketball live text broadcast.
  arXiv  Detail & Related papers  (2024-01-25T02:08:37Z)
• Understanding why shooters shoot -- An AI-powered engine for basketball performance profiling [70.54015529131325]
  Basketball is dictated by many variables, such as playstyle and game dynamics. It is crucial that the performance profiles can reflect the diverse playstyles. We present a tool that can visualize player performance profiles in a timely manner.
  arXiv  Detail & Related papers  (2023-03-17T01:13:18Z)
• A Survey on Video Action Recognition in Sports: Datasets, Methods and Applications [60.3327085463545]
  We present a survey on video action recognition for sports analytics. We introduce more than ten types of sports, including team sports, such as football, basketball, volleyball, hockey and individual sports, such as figure skating, gymnastics, table tennis, diving and badminton. We develop a toolbox using PaddlePaddle, which supports football, basketball, table tennis and figure skating action recognition.
  arXiv  Detail & Related papers  (2022-06-02T13:19:36Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.