SoccerNet-Caption
Models write timestamped commentary for soccer broadcasts. The set has almost 37k commentaries over 715.9 hours of video. Streaming reports also score when the model chooses to speak.
It is a SoccerNet challenge task. Microsoft's Mage-VL card uses it to score response timing for proactive streaming.
- Made by
- University of Liege and KAUST (SoccerNet)
- Released
- April 2023
- Kind
- Video captioning
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Open-ended
- Tasks
- Public, plus a private test set
In the authors’ words
Soccer is more than just a game - it is a passion that transcends borders and unites people worldwide. From the roar of the crowds to the excitement of the commentators, every moment of a soccer match is a thrill. Yet, with so many games happening simultaneously, fans cannot watch them all live. Notifications for main actions can help, but lack the engagement of live commentary, leaving fans feeling disconnected. To fulfill this need, we propose in this paper a novel task of dense video captioning focusing on the generation of textual commentaries anchored with single timestamps. To support this task, we additionally present a challenging dataset consisting of almost 37k timestamped commentaries across 715.9 hours of soccer broadcast videos. Additionally, we propose a first benchmark and baseline for this task, highlighting the difficulty of temporally anchoring commentaries yet showing the capacity to generate meaningful commentaries. By providing broadcasters with a tool to summarize the content of their video with the same level of engagement as a live game, our method could help satisfy the needs of the numerous fans who follow their team but cannot necessarily watch the live game. We believe our method has the potential to enhance the accessibility and understanding of soccer content for a wider audience, bringing the excitement of the game to more people.
OPPO - VideoText captioning system tops the leaderboard with 26.7
Best score per entry from the benchmark's own leaderboard, read on November 12th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders.
- 1 OPPO - VideoText captioning system 2023 challenge 26.7
- 2 HZC captioning system 2023 challenge 21.3
- 3 Baseline 2 captioning system 2023 challenge 21.3
- 4 justplay captioning system 2023 challenge 21.2
- 5 aisoccer captioning system 2023 challenge 21.0
- 6 Baseline 1 captioning system 2023 challenge 15.2
Source: Official SoccerNet-Caption leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | OPPO - VideoText captioning system | 26.7 | 2023 challenge | Official leaderboard | |
| 2 | HZC captioning system | 21.3 | 2023 challenge | Official leaderboard | |
| 3 | Baseline 2 captioning system | 21.3 | 2023 challenge | Official leaderboard | |
| 4 | justplay captioning system | 21.2 | 2023 challenge | Official leaderboard | |
| 5 | aisoccer captioning system | 21.0 | 2023 challenge | Official leaderboard | |
| 6 | Baseline 1 captioning system | 15.2 | 2023 challenge | Official leaderboard |
No lab model card reports it yet
We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark.
More in Audio and video
- Video-MMMU In use 300 expert-level lecture videos with questions testing perception, comprehension and adaptation of knowledge. Accuracy.
Qwen 3.8 Max 88.7% - LVBench In use Hours-long video understanding with multiple-choice questions on reasoning, grounding and summarization.
Gemini 4 Argon 91.7% - MMVU In use 3,000 expert-annotated questions on videos across 27 disciplines. Multiple-choice and open-ended accuracy. Qwen 3.8 Max 82.4%
- MLVU In use Multi-task long video understanding (minutes to hours) with multiple-choice and generation tasks. Accuracy. Qwen 3.8 Max 90.8%
- FLEURS In use Models transcribe read speech in 102 languages built from FLoRes-101 sentences. Model cards usually report word error rate, so lower scores are better.
Phi-4-multimodal 4.00% - VoiceBench In use Spoken instructions with speaker, environment and content variation. Voice assistant responses scored by GPT judge or rules. Qwen3.5-Omni-Plus 93.1%