研究人员推出ProCon，一种用于图像异常检测的新型无需训练的框架。ProCon将内存检索转化为重构过程，将测试块投影到正常内存向量上来识别异常。该方法避免了解码器训练、骨干网络微调或伪异常监督的需要。ProCon在MVTec-AD、VisA和Real-IAD等多个基准测试中表现强劲，实现了高图像和像素级精度。 AI
影响 这一新框架有望简化和提高各种工业应用中异常检测系统的效率。
排序理由 该集群描述了一篇详细介绍新型异常检测框架的新研究论文。
在 Hugging Face Daily Papers 阅读 →
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
Memory-based anomaly detection is attractive because it localizes defects from normal images without training a decoder or synthesizing pseudo anomalies. However, most memory methods still use the memory bank as a nearest-neighbor lookup table: a test patch is treated as normal i…
arXiv:2607.04894v1 Announce Type: new Abstract: Memory-based anomaly detection is attractive because it localizes defects from normal images without training a decoder or synthesizing pseudo anomalies. However, most memory methods still use the memory bank as a nearest-neighbor l…
Memory-based anomaly detection is attractive because it localizes defects from normal images without training a decoder or synthesizing pseudo anomalies. However, most memory methods still use the memory bank as a nearest-neighbor lookup table: a test patch is treated as normal i…