BoxComm: Benchmarking Category-Aware Commentary Generation and Narration Rhythm in Boxing

• URL: http://arxiv.org/abs/2604.04419v1
• Date: Mon, 06 Apr 2026 04:56:49 GMT
• Title: BoxComm: Benchmarking Category-Aware Commentary Generation and Narration Rhythm in Boxing
• Abstract summary: We present BoxComm, a large-scale dataset comprising 445 World Boxing Championship match videos with over 52K commentary sentences from professional broadcasts.<n>We propose a structured commentary taxonomy that categorizes each sentence into play-by-play, tactical, or contextual.<n>We introduce two novel and complementary evaluations tailored to sports commentary generation.
• Score: 9.244368483134101
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent multimodal large language models (MLLMs) have shown strong capabilities in general video understanding, driving growing interest in automatic sports commentary generation. However, existing benchmarks for this task focus exclusively on team sports such as soccer and basketball, leaving combat sports entirely unexplored. Notably, combat sports present distinct challenges: critical actions unfold within milliseconds with visually subtle yet semantically decisive differences, and professional commentary contains a substantially higher proportion of tactical analysis compared to team sports. In this paper, we present BoxComm, a large-scale dataset comprising 445 World Boxing Championship match videos with over 52K commentary sentences from professional broadcasts. We propose a structured commentary taxonomy that categorizes each sentence into play-by-play, tactical, or contextual, providing the first category-level annotation for sports commentary benchmarks. Building on this taxonomy, we introduce two novel and complementary evaluations tailored to sports commentary generation: (1) category-conditioned generation, which evaluates whether models can produce accurate commentary of a specified type given video context; and (2) commentary rhythm assessment, which measures whether freely generated commentary exhibits appropriate temporal pacing and type distribution over continuous video segments, capturing a dimension of commentary competence that prior benchmarks have not addressed. Experiments on multiple state-of-the-art MLLMs reveal that current models struggle on both evaluations. We further propose EIC-Gen, an improved baseline incorporating detected punch events to supply structured action cues, yielding consistent gains and highlighting the importance of perceiving fleeting and subtle events for combat sports commentary.

Related papers

• TF-CADE: Foreground-Concentrated Text-Video Alignment for Zero-Shot Temporal Action Detection [54.38725298163703]
  We propose a Text-Foreground Concentrated Alignment for zero-shot temporal action DEtector (TF-CADE)<n>ACA extracts action concentrate scores to aggregate temporally informative video segments into a foreground- weighted video embedding.<n> CCR strategy refines per-snippet confidence scores by lever- aging foreground-aware similarity, effectively suppressing irrelevant action classes during inference.
  arXiv  Detail & Related papers  (2026-08-18T06:43:45Z)
• SoccerLens: Grounded Soccer Video Understanding Beyond Accuracy [52.09081885158698]
  Vision-language models (VLMs) have recently shown strong potential in soccer video understanding.<n>Existing evaluation protocols focus on classification accuracy and do not assess visual grounding.<n>We introduce SoccerLens, a benchmark for grounded soccer video understanding.
  arXiv  Detail & Related papers  (2026-05-10T15:18:08Z)
• TennisExpert: Towards Expert-Level Analytical Sports Video Understanding [16.625250626542208]
  Tennis is one of the most widely followed sports, generating extensive broadcast footage with strong potential for professional analysis, automated coaching, and real-time commentary.<n>However, automatic tennis understanding remains underexplored due to lack of large-scale benchmarks with fine-grained annotations and expert-level commentary.<n>To address these challenges, we introduce TennisVL, a large-scale tennis benchmark comprising over 200 professional matches (471.9 hours) and 40,000+ rally-level clips.<n>We propose TennisExpert, a multimodal tennis understanding framework that integrates a video semantic with a memory-augmented model built on Qwen3-VL-8B
  arXiv  Detail & Related papers  (2026-03-11T15:12:10Z)
• Real-Time Generation of Game Video Commentary with Multimodal LLMs: Pause-Aware Decoding Approaches [69.57389826203699]
  We investigate whether in-context prompting alone can support real-time commentary generation that is both semantically relevant and well-timed.<n>We propose two prompting-based decoding strategies: 1) a fixed-interval approach, and 2) a novel dynamic interval-based decoding approach.<n> Experiments on Japanese and English datasets of racing and fighting games show that the dynamic interval-based decoding can generate commentary more closely aligned with human utterance timing and content using prompting alone.
  arXiv  Detail & Related papers  (2026-03-03T06:39:04Z)
• Generalizing Sports Feedback Generation by Watching Competitions and Reading Books: A Rock Climbing Case Study [27.82932642584153]
  Video-LLMs struggle on the challenging task of sports feedback generation.<n>Traditional text generation evaluation metrics fail to capture the unique aspects of sports feedback quality.<n>We propose two evaluation metrics: (1) specificity and (2) actionability.
  arXiv  Detail & Related papers  (2026-02-09T18:41:43Z)
• BoxingVI: A Multi-Modal Benchmark for Boxing Action Recognition and Localization [1.623267727687624]
  We present a comprehensive, well-annotated video dataset tailored for punch detection and classification in boxing.<n>The dataset comprises 6,915 high-quality punch clips categorized into six distinct punch types.<n>This contribution aims to accelerate progress in movement analysis, automated coaching, and performance assessment within boxing and related domains.
  arXiv  Detail & Related papers  (2025-11-20T16:37:07Z)
• Commentary Generation for Soccer Highlights [0.0]
  We extend MatchVoice to commentary generation for soccer highlights using the GOAL dataset.<n>We conduct extensive experiments to reproduce the original MatchTime results and evaluate our setup.<n>Our findings suggest the need for integrating techniques from broader video-language domains to further enhance performance.
  arXiv  Detail & Related papers  (2025-08-11T01:48:37Z)
• MatchTime: Towards Automatic Soccer Game Commentary Generation [52.431010585268865]
  We consider constructing an automatic soccer game commentary model to improve the audiences' viewing experience. First, observing the prevalent video-text misalignment in existing datasets, we manually annotate timestamps for 49 matches. Second, we propose a multi-modal temporal alignment pipeline to automatically correct and filter the existing dataset at scale. Third, based on our curated dataset, we train an automatic commentary generation model, named MatchVoice.
  arXiv  Detail & Related papers  (2024-06-26T17:57:25Z)
• Sports Video: Fine-Grained Action Detection and Classification of Table Tennis Strokes from Videos for MediaEval 2021 [0.0]
  This task tackles fine-grained action detection and classification from videos. The focus is on recordings of table tennis games. This work aims at creating tools for sports coaches and players in order to analyze sports performance.
  arXiv  Detail & Related papers  (2021-12-16T10:17:59Z)
• Sentence Punctuation for Collaborative Commentary Generation in Esports Live-Streaming [2.6330185040436347]
  This paper presents two strategies for sentence punctuation for text sequences of game commentary, that is, punctuating sentences by two or three text sequence(s) originally punctuated by Youtube to obtain a complete sentence of commentary. We conducted comparative experiments utilizing and fine-tuning a state-of-the-art pre-trained generative language model among two strategies and the baseline to generate collaborative commentary.
  arXiv  Detail & Related papers  (2021-10-24T11:26:29Z)
• Cross-Sentence Temporal and Semantic Relations in Video Activity Localisation [79.50868197788773]
  We develop a more accurate weakly-supervised solution by introducing Cross-Sentence Relations Mining. We explore two cross-sentence relational constraints: (1) trimmed ordering and (2) semantic consistency among sentences in a paragraph description of video activities. Experiments on two publicly available activity localisation datasets show the advantages of our approach over the state-of-the-art weakly supervised methods.
  arXiv  Detail & Related papers  (2021-07-23T20:04:01Z)
• Temporally-Aware Feature Pooling for Action Spotting in Soccer Broadcasts [86.56462654572813]
  We focus our analysis on action spotting in soccer broadcast, which consists in temporally localizing the main actions in a soccer game. We propose a novel feature pooling method based on NetVLAD, dubbed NetVLAD++, that embeds temporally-aware knowledge. We train and evaluate our methodology on the recent large-scale dataset SoccerNet-v2, reaching 53.4% Average-mAP for action spotting.
  arXiv  Detail & Related papers  (2021-04-14T11:09:03Z)
• Hybrid Dynamic-static Context-aware Attention Network for Action Assessment in Long Videos [96.45804577283563]
  We present a novel hybrid dynAmic-static Context-aware attenTION NETwork (ACTION-NET) for action assessment in long videos. We learn the video dynamic information but also focus on the static postures of the detected athletes in specific frames. We combine the features of the two streams to regress the final video score, supervised by ground-truth scores given by experts.
  arXiv  Detail & Related papers  (2020-08-13T15:51:42Z)

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.