O-VAD: Industrial Video Anomaly Detection through Object-Centric Tracking and Reasoning

  • 类型:arxiv
  • 标识:2607.18142
  • 链接:https://arxiv.org/abs/2607.18142
  • 主分类:multimodal
  • 形态:method
  • 被引:0
  • 被引来源:Semantic Scholar
  • S2被引:0
  • 影响力被引:0
  • TLDR:This work introduces a training-free agentic framework for anomaly detection free of domain-specific knowledge, designed to track spatial-temporal dynamics and underlying transformations of detected objects over time, and then reason over the object-wise temporal state trajectories to identify abnormal objects in grounded frames.
  • 待LLM分类:否
  • 标题中文:O-VAD: 基于以对象为中心的跟踪与推理的工业视频异常检测
  • TLDR中文:该工作提出一个面向异常检测的免训练 agentic 框架,无需领域特定知识,旨在追踪被检测对象随时间变化的空间-时间动态与底层变换,然后基于逐对象的时间状态轨迹进行推理,在 grounding 帧中识别异常对象。
  • 来源文件
  • /inbox/tom/_candidates/2026-07-28-rag-retrieval-reranking-candidates.json
  • /inbox/tom/_candidates/2026-07-28-agent-rag-longcontext-candidates.json
  • [S2 enrich]