用自适应特征扰动提升单类异常检测的泛化能力
Removing Geometric Bias in One-Class Anomaly Detection with Adaptive Feature Perturbation
- 在预训练特征空间中动态调整噪声分布,生成更真实的伪异常特征
- 在标准与无几何偏见数据集上均超越基线方法,准确率提升显著
- 适合需要高鲁棒性的工业缺陷检测等实际场景
单类异常检测旨在识别不属于预定义正常类别的对象。由于训练数据缺乏异常样本,现有方法通常基于正常图像进行数据增强以生成伪异常数据。然而,表现最佳的方法隐式利用了基准数据集中的几何偏见,限制了其在更一般场景下的适用性;而其他方法依赖基础加噪方案,难以有效捕捉正常数据的内在结构。此外,多数方法仍局限于图像域生成伪异常,仅从正常类端到端训练模型,忽视了信息的丰富表示。为克服上述局限,本文采用冻结但丰富的预训练特征空间,提出一种新颖的自适应线性特征扰动技术:根据每个样本自适应调整噪声分布,对特征向量施加衰减线性扰动,并通过对比学习目标引导分类过程。在标准及无几何偏见数据集上的实验表明,该方法优于可比基线。代码已公开。
原文摘要 · Abstract (English)
One-class anomaly detection aims to detect objects that do not belong to a predefined normal class. In practice training data lack those anomalous samples; hence state-of-the-art methods are trained to discriminate between normal and synthetically-generated pseudo-anomalous data. Most methods use data augmentation techniques on normal images to simulate anomalies. However the best-performing ones implicitly leverage a geometric bias present in the benchmarking datasets. This limits their usability in more general conditions. Others are relying on basic noising schemes that may be suboptimal in capturing the underlying structure of normal data. In addition most still favour the image domain to generate pseudo-anomalies training models end-to-end from only the normal class and overlooking richer representations of the information. To overcome these limitations we consider frozen yet rich feature spaces given by pretrained models and create pseudo-anomalous features with a novel adaptive linear feature perturbation technique. It adapts the noise distribution to each sample applies decaying linear perturbations to feature vectors and further guides the classification process using a contrastive learning objective. Experimental evaluation conducted on both standard and geometric bias-free datasets demonstrates the superiority of our approach with respect to comparable baselines. The codebase is accessible via our public repository.
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。