Enregistré dans:
Détails bibliographiques
Auteurs principaux: Bao, Zhongyuan, Zhang, Lejun
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2509.15602
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866913037698465792
author Bao, Zhongyuan
Zhang, Lejun
author_facet Bao, Zhongyuan
Zhang, Lejun
contents Multimodal large language models (MLLMs) excel at general video understanding but struggle with fast, high-frequency sports like tennis, where rally clips are short yet information-dense. To systematically evaluate MLLMs in this challenging domain, we present TennisTV, the first and most comprehensive benchmark for tennis video understanding. TennisTV models each rally as a temporal-ordered sequence of consecutive stroke events, using automated pipelines for filtering and question generation. It covers 8 tasks from the stroke level to the rally level and includes 2527 human-verified questions. Evaluating 17 representative MLLMs, we provide the first systematic assessment of tennis video understanding. Results yield two key insights: (i) frame-sampling density should be tailored and balanced across tasks, and (ii) improving temporal grounding is essential for stronger reasoning.
format Preprint
id arxiv_https___arxiv_org_abs_2509_15602
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle TennisTV: Do Multimodal Large Language Models Understand Tennis Rallies?
Bao, Zhongyuan
Zhang, Lejun
Computer Vision and Pattern Recognition
Multimodal large language models (MLLMs) excel at general video understanding but struggle with fast, high-frequency sports like tennis, where rally clips are short yet information-dense. To systematically evaluate MLLMs in this challenging domain, we present TennisTV, the first and most comprehensive benchmark for tennis video understanding. TennisTV models each rally as a temporal-ordered sequence of consecutive stroke events, using automated pipelines for filtering and question generation. It covers 8 tasks from the stroke level to the rally level and includes 2527 human-verified questions. Evaluating 17 representative MLLMs, we provide the first systematic assessment of tennis video understanding. Results yield two key insights: (i) frame-sampling density should be tailored and balanced across tasks, and (ii) improving temporal grounding is essential for stronger reasoning.
title TennisTV: Do Multimodal Large Language Models Understand Tennis Rallies?
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.15602