← Back to all news
arXiv Computer Vision

SoccerNet-FoulRet: Retrieving Semantically Similar Soccer Foul Videos

SoccerNet‑FoulRet is a new benchmark for retrieving semantically similar soccer foul videos, addressing inconsistencies in refereeing decisions by allowing comparison of a contentious foul with past precedents. The task requires retrieving relevant fouls regardless of camera angle, teams, or appearance, differing from prior video‑to‑video retrieval that focuses on visual similarity. The benchmark, built from the SoccerNet‑MVFoul dataset, evaluates zero‑shot video and vision‑language embeddings and a fine‑tuned baseline on 693 human‑verified queries, revealing that even the best zero‑shot model achieves less than 5% HitRate@10, highlighting the challenge of semantic foul retrieval.

Related stories

arXiv Computer Vision
Oct 28

Zero-Shot Multimodal Event Spotting in Soccer Broadcasts

The paper introduces a zero‑shot multimodal framework for spotting soccer events in full‑length broadcasts without task‑specific training. It segments matches into overlapping clips, represents them with a compact visual storyboard and aligned commentary, and processes these jointly with a general‑purpose multimodal model. On three annotated matches, the best setup achieves 72.0% F1@5s, surpassing SoccerNet‑trained baselines, and the storyboard reduces inference cost by 72%. "whyItMatters":"The approach enables efficient, high‑accuracy event detection and highlight generation in soccer broadcasts without relying on large annotated datasets or soccer‑specific training, as demonstrated by its superior performance and reduced inference cost."

By Sadia Kamal, Mahmudul Hasan, Tony Braskich
multimodal
More like this →
arXiv Computer Vision
Sep 24

MultiVENT-Raw: A Benchmark for Retrieval and Reasoning over Raw Videos

MultiVENT‑Raw is a new multilingual benchmark comprising nearly 120,000 raw videos—continuous footage from cell phones, hand‑held cameras, or CCTV—totaling over 5,300 hours. The dataset includes 130 events and 222 event‑centric queries, along with human‑annotated relevance judgments and extracted key facts for relevant videos. It supports two tasks: retrieving videos relevant to a query event and generating a coherent report summarizing event‑related videos for a target user, with baseline models showing these tasks remain challenging.

By Reno Kriz, David Etter, Alexander Martin, Cameron Carpenter, Debashish Chakraborty, Hannah Recknor, Reihaneh Iranmanesh, Matthew Maciejewski, Kenton Murray, Eugene Yang, Benjamin Van Durme, Aaron Steven White, Andrew Yates, William Walden
multimodalbenchmarks
More like this →
arXiv AI
Sep 18

Can Vision-Language Models Judge Olympic Diving? From Reasoning to Scores in Zero-Shot Action Quality Assessment

The paper investigates whether open‑source Vision‑Language Models (VLMs) can perform zero‑shot action quality assessment (AQA) on Olympic diving videos. Using the AQA‑7 benchmark, the authors propose a regression framework that combines VLM‑generated semantic reasoning, phase‑level sub‑scores, TF‑IDF vectorization, dimensionality reduction, and ensemble learning to predict final competition scores. While individual VLMs achieve moderate Spearman correlations (<0.32), the ensemble approach boosts performance to 0.67, demonstrating that VLM‑derived textual reasoning features are more informative than raw numerical sub‑scores for AQA. whyItMatters":"The study shows that VLMs can serve as explainable, semi‑automated tools for evaluating sports performance, potentially aiding expert judging in complex, subjective Olympic events."

By Henry O. Velesaca, David Freire-Obregon, Luigi Miranda, Abel Reyes-Angulo
llmsmultimodalbenchmarks
More like this →
arXiv AI
Sep 17

ShotFinder: Imagination-Driven Open-Domain Video Shot Retrieval via Web Search

ShotFinder introduces a new benchmark for open‑domain video shot retrieval, formalizing editing requirements as keyframe‑oriented shot descriptions and adding five controllable constraints—temporal order, color, visual style, audio, and resolution. The benchmark comprises 1,210 high‑quality YouTube samples across 20 themes, generated with large models and verified by humans. A three‑stage retrieval pipeline—query expansion via video imagination, candidate video retrieval, and description‑guided shot localization—shows a notable performance gap to humans, especially for color and visual style constraints.

By Tao Yu, Haopeng Jin, Hao Wang, Shenghua Chai, Yujia Yang, Junhao Gong, Jiaming Guo, Minghui Zhang, Xinlong Chen, Zhenghao Zhang, Yuxuan Zhou, Yufei Xiong, Shanbin Zhang, Jiabing Yang, YiFan Zhang, Hongzhu Yi, Xinming Wang, Cheng Zhong, Xiao Ma, Zhang Zhang, Yan Huang, Liang Wang
llmsmultimodalbenchmarks
More like this →