解决工业缺陷检测中模型遗忘问题，实现高效稳定增量学习。
ONER: Online Experience Replay for Incremental Anomaly Detection
- 动态生成图像条件提示，防止旧知识被覆盖。
- 引入语义原型，减少跨任务特征混淆，提升检测精度。
- 参数少、训练快，适合实际产线部署。
增量异常检测旨在序列化识别工业生产线中的缺陷，但面临灾难性遗忘问题，主要源于参数更新时的知识覆盖以及任务间特征冲突。本文提出ONER（Online Experience Replay）端到端框架，通过融合两类经验协同解决上述问题：(1) 分解提示，从可复用模块动态生成图像条件提示，保留先验知识，避免知识覆盖；(2) 语义原型，在像素与图像层级强制潜在特征空间的可分性，缓解跨任务特征冲突。大量实验表明，ONER性能领先，相比先前方法在MVTec AD数据集上实现+4.4%像素级AUROC和+28.3%像素级AUPR提升。尤为突出的是，仅需0.019M参数和每任务5个训练轮次，验证了其在真实工业场景中的高效性与稳定性。
原文摘要 · Abstract (English)
Incremental anomaly detection aims to sequentially identify defects in industrial product lines but suffers from catastrophic forgetting, primarily due to knowledge overwriting during parameter updates and feature conflicts between tasks. In this work, We propose ONER (ONline Experience Replay), an end-to-end framework that addresses these issues by synergistically integrating two types of experience: (1) decomposed prompts, which dynamically generate image-conditioned prompts from reusable modules to retain prior knowledge thus prevent knowledge overwriting, and (2) semantic prototypes, which enforce separability in latent feature spaces at pixel and image levels to mitigate cross-task feature conflicts. Extensive experiments demonstrate the superiority of ONER, achieving state-of-the-art performance with +4.4% Pixel AUROC and +28.3% Pixel AUPR improvements on the MVTec AD dataset over prior methods. Remarkably, ONER achieves this with only 0.019M parameters and 5 training epochs per task, confirming its efficiency and stability for real-world industrial deployment.
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。