仅用一个时间点标注，实现流式视频动作定位，突破传统依赖完整标注的局限。
OnPoint: Offline-to-Online Multi-Level Distillation for Point-Supervised Online Temporal Action Localization
- 从离线教师模型蒸馏多层级知识到在线学生模型
- 在五个数据集上超越强基线，平均性能提升显著
- 适合实时视频分析场景，尤其适用于标注成本高的应用
时间动作定位（TAL）通常依赖段落标注或对完整视频的离线访问，限制了可扩展性和在线应用。本文提出点监督在线动作定位（POTAL），仅需每实例一个时间点即可在流式视频中定位动作。为此，我们设计了OnPoint，一种离线到在线的多层级知识蒸馏框架，通过伪段落实例蒸馏、类别激活序列蒸馏和前瞻窗口级蒸馏，将点监督的离线教师模型知识迁移至在线学生模型。此外，通过引入原始点标签并结合动作置信度引导的注意力校准来优化锚框解码，提升了模型鲁棒性。在五个数据集上的实验表明，OnPoint持续优于多个强基线，为POTAL奠定了坚实基础。
原文摘要 · Abstract (English)
Temporal Action Localization (TAL) typically relies on segment annotations or offline access to full videos, limiting scalability and online use. We introduce Point-Supervised Online TAL (POTAL), which localizes actions in streaming videos using only one temporal point per instance. To solve POTAL, we propose OnPoint, an offline-to-online multi-level distillation framework that transfers knowledge from a point-supervised offline teacher to an online student via (i) pseudo-segment instance distillation, (ii) class-activation sequence distillation, and (iii) anticipatory window-level distillation. We further improve robustness by incorporating the original point labels into student training and by refining anchor decoding with actionness-guided attention calibration. Experiments on five datasets show OnPoint consistently outperforms strong baselines, establishing a solid foundation for POTAL.
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。