English(EN) CoRE: Weakly Supervised Coarse-to-Fine Risk Evidence Learning in Driving Videos

新的CoRE框架从粗粒度视频监督中学习细粒度风险证据

作者 PulseAugur 编辑部 · [1 个来源] · 2026-08-27 04:00

研究人员开发了CoRE，一种新颖的弱监督学习框架，可以从粗粒度的视频级预测中提取细粒度的风险证据。该方法训练一个视频级预测器，然后使用结构化干预来衡量候选的时间区域或实体如何影响预测。由此产生的目标被蒸馏到一个学生模型中，该模型可以直接预测时间和实体支持，而无需昂贵的细粒度注释。CoRE在驾驶视频风险评估、交通异常定位和一般异常检测基准测试中都显示出了有效性。 AI

影响该框架可以减少视频分析任务中对昂贵细粒度注释的需求。

排序理由该集群包含一篇详细介绍新学习框架的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CV 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 30 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇详细介绍新学习框架的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, other; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.CV TIER_1 English(EN) · Kaiser Hamid, Can Cui, Nade Liang · 2026-08-27 04:00

CoRE：驾驶视频中弱监督的粗粒度到细粒度风险证据学习

arXiv:2608.25344v1 Announce Type: new Abstract: Perceived risk in driving evolves over time and may be supported by specific scene entities, yet supervision is typically limited to coarse video-level judgments. Learning \emph{when} supporting evidence emerges and \emph{which enti…

本文如何被排名

报道来源 [1]

CoRE：驾驶视频中弱监督的粗粒度到细粒度风险证据学习

相关实体

相关话题