用统一图网络和知识蒸馏，实现少样本精准事件定位
Few-Shot Precise Event Spotting via Unified Multi-Entity Graph and Distillation
- 构建人体骨骼与运动物体关键点的统一图结构
- 少样本下准确率显著优于基线模型
- 适合缺乏标注数据的体育分析场景
精确事件定位（PES）旨在识别体育视频中精确时间点的细粒度事件，是体育分析的关键。该任务因事件快速连续、运动模糊及视觉差异微小而极具挑战。现有方法多依赖大规模标注数据进行端到端训练，且仅使用像素或姿态输入，在少样本条件下表现不佳。本文提出统一多实体图网络（UMEG-Net），将人体骨骼与特定运动物体关键点融合为统一图结构，并设计基于先进GCN与多尺度时移的高效时空提取模块。为进一步提升性能，采用多模态知识蒸馏，将关键点图中的知识迁移到视觉表征。本方法在少量标注数据下仍具鲁棒性，在少样本设置中显著优于基线模型，为少样本精确事件定位提供了可扩展、高效的解决方案。代码已公开于https://github.com/LZYAndy/UMEG-Net。
原文摘要 · Abstract (English)
Precise event spotting (PES) aims to recognize fine-grained events at exact moments and has become a key component of sports analytics. This task is particularly challenging due to rapid succession, motion blur, and subtle visual differences. Consequently, most existing methods rely on domain-specific, end-to-end training with large labeled datasets and often struggle in few-shot conditions due to their dependence on pixel- or pose-based inputs alone. However, obtaining large labeled datasets is practically hard. We propose a Unified Multi-Entity Graph Network (UMEG-Net) for few-shot PES. UMEG-Net integrates human skeletons and sport-specific object keypoints into a unified graph and features an efficient spatio-temporal extraction module based on advanced GCN and multi-scale temporal shift. To further enhance performance, we employ multimodal distillation to transfer knowledge from keypoint-based graphs to visual representations. Our approach achieves robust performance with limited labeled data and significantly outperforms baseline models in few-shot settings, providing a scalable and effective solution for few-shot PES. Code is publicly available at https://github.com/LZYAndy/UMEG-Net.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。