FineBadminton: A Multi-Level Dataset for Fine-Grained Badminton Video Understanding

TL;DR

This paper introduces FineBadminton, a large-scale, multi-level dataset for detailed badminton video analysis, along with benchmarks and baseline methods to advance multimodal large language models in sports understanding.

Contribution

The paper presents a novel dataset with rich annotations, a benchmark for evaluation, and optimized baseline approaches for fine-grained badminton video understanding.

Findings

Current MLLMs struggle with deep sports video analysis.

Proposed methods improve performance on the FBBench benchmark.

FineBadminton facilitates research in fine-grained sports video understanding.

Abstract

Fine-grained analysis of complex and high-speed sports like badminton presents a significant challenge for Multimodal Large Language Models (MLLMs), despite their notable advancements in general video understanding. This difficulty arises primarily from the scarcity of datasets with sufficiently rich and domain-specific annotations. To bridge this gap, we introduce FineBadminton, a novel and large-scale dataset featuring a unique multi-level semantic annotation hierarchy (Foundational Actions, Tactical Semantics, and Decision Evaluation) for comprehensive badminton understanding. The construction of FineBadminton is powered by an innovative annotation pipeline that synergistically combines MLLM-generated proposals with human refinement. We also present FBBench, a challenging benchmark derived from FineBadminton, to rigorously evaluate MLLMs on nuanced spatio-temporal reasoning and tactical comprehension. Together, FineBadminton and FBBench provide a crucial ecosystem to catalyze research in fine-grained video understanding and advance the development of MLLMs in sports intelligence. Furthermore, we propose an optimized baseline approach incorporating Hit-Centric Keyframe Selection to focus on pivotal moments and Coordinate-Guided Condensation to distill salient visual information. The results on FBBench reveal that while current MLLMs still face significant challenges in deep sports video analysis, our proposed strategies nonetheless achieve substantial performance gains. The project homepage is available at https://finebadminton.github.io/FineBadminton/.