Event ActivityNet：用于未修剪动作理解的大规模模拟事件基准
Event ActivityNet: A Large-Scale Simulated-Event Benchmark for Untrimmed Action Understanding
浏览论文内容
中文总结 AI 辅助
本文提出大规模模拟事件基准Event ActivityNet，解决长时序事件动作理解数据集不足的问题，建立基线并验证其预训练对原生事件微调的性能提升效果。
中文摘要 AI 辅助
长时序基于事件的动作理解研究仍未得到充分探索，原因在于现有数据集大多由短的、经过修剪的片段组成，而收集带有密集时间标注的原生事件流成本高昂。我们推出Event ActivityNet，这是一个源自人工标注的未修剪ActivityNet视频的大规模模拟事件基准。它包含3263个视频、200个动作类别，总时长106.94小时，配有匹配的5-bin和9-bin事件体素表示、时间动作标注以及带时间戳的字幕。该基准支持标注片段动作识别、辅助事件-语言对齐以及因果在线时间动作定位。我们按解码帧顺序直接从非插值源视频生成事件体素，保留每个视频合理的标称或平均帧率元数据用于近似时间映射，并使用动作中心重建LPIPS作为保留可重建内容的软诊断指标。我们为自适应事件 framing、提示-字幕对齐以及仅事件、仅RGB和RGB-事件定位建立了基线。在渐进嵌套尺度训练协议下，识别Top-1准确率从52.25提升至66.42，在线时间定位平均mAP从21.7提升至29.0。此外，分阶段Event ActivityNet预训练后再进行原生事件微调，在多种监督预算下始终优于仅目标训练和从头联合训练。Event ActivityNet为长时序事件建模提供了可扩展的基准，不过面向部署的结论仍需原生相机评估。
英文摘要
Long-horizon event-based action understanding remains underexplored because existing datasets largely comprise short, trimmed clips, while collecting native event streams with dense temporal annotations is costly. We introduce Event ActivityNet, a large-scale simulated-event benchmark derived from human-annotated, untrimmed ActivityNet videos. It comprises 3,263 videos, 200 action classes, and 106.94 hours, with matched 5-bin and 9-bin event-voxel representations, temporal action annotations, and timestamped captions. The benchmark supports annotated-segment action recognition, auxiliary event-language alignment, and causal online temporal action localization. We generate event voxels directly from non-interpolated source videos in decoded frame order, retain per-video rational nominal or average frame-rate metadata for approximate time mapping, and use action-center reconstruction LPIPS as a soft diagnostic of retained reconstructable content. We establish baselines for adaptive event framing, prompt-caption alignment, and event-only, RGB-only, and RGB-event localization. Under a progressive nested-scale training protocol, recognition Top-1 accuracy increases from 52.25 to 66.42, while online temporal localization average mAP improves from 21.7 to 29.0. Moreover, staged Event ActivityNet pretraining followed by native-event fine-tuning consistently outperforms target-only and joint-from-scratch training across multiple supervision budgets. Event ActivityNet provides a scalable benchmark for long-horizon event modeling, although native-camera evaluation remains essential for deployment-oriented conclusions.
