用对比学习提升缺陷检测，尤其在光照/对焦变化下更准。
Contrastive Knowledge Distillation for Anomaly Detection in Multi-Illumination/Focus Display Images
- 不依赖正负样本对，通过拉近/推远师生特征距离来训练。
- 在MMdAD数据集上AUROC和准确率均超越现有方法。
- 适合需要高精度检测的工业视觉质检场景。
本文针对多光照、多对焦显示图像中的自动缺陷检测问题。由于显示表面的微小缺陷在RGB图像中难以察觉，且仅用正常数据训练的模型表现受限，我们提出一种基于对比学习的知识蒸馏新方法。在以多分辨率知识蒸馏（MKD）为基础框架的前提下，引入多分辨率对比蒸馏（MCD），无需正负样本对，通过拉近或推远教师与学生网络特征间的距离实现学习。此外，设计融合模块将多通道信息转换并聚合至MCD的三通道输入层。所提方法在自建的多光照多对焦显示图像异常检测数据集（MMdAD）上，在AUROC和准确率两项指标上显著优于现有先进方法。
原文摘要 · Abstract (English)
In this paper, we tackle automatic anomaly detection in multi-illumination and multi-focus display images. The minute defects on the display surface are hard to spot out in RGB images and by a model trained with only normal data. To address this, we propose a novel contrastive learning scheme for knowledge distillation-based anomaly detection. In our framework, Multiresolution Knowledge Distillation (MKD) is adopted as a baseline, which operates by measuring feature similarities between the teacher and student networks. Based on MKD, we propose a novel contrastive learning method, namely Multiresolution Contrastive Distillation (MCD), which does not require positive/negative pairs with an anchor but operates by pulling/pushing the distance between the teacher and student features. Furthermore, we propose the blending module that transforms and aggregate multi-channel information to the three-channel input layer of MCD. Our proposed method significantly outperforms competitive state-of-the-art methods in both AUROC and accuracy metrics on the collected Multi-illumination and Multi-focus display image dataset for Anomaly Detection (MMdAD).
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。