通过自适应抑制背景与修正混淆前景，提升少样本异常检测效果
Enhancing Few-Shot Out-of-Distribution Detection via the Refinement of Foreground and Background
- 自适应加权背景块的分类熵，避免统一压制
- 识别并修正与其它类别相似的前景区域，减少误判
- 可即插即用，显著提升现有方法在少样本异常检测的表现
基于CLIP的前景-背景（FG-BG）分解方法在提升少样本异常检测性能方面表现卓越。然而，现有方法仍存在若干局限：对分解得到的背景区域采用统一抑制策略，忽视了不同图像块对预测贡献的差异；对于前景区域，未充分考虑部分局部块可能在外观或语义上与其他类别相似，从而误导训练过程。为此，本文提出一种新的即插即用框架，包含三个核心组件：(1) 前景-背景分解模块，沿用先前方法将图像分为前景与背景区域；(2) 自适应背景抑制模块，根据图像块的分类熵进行自适应加权；(3) 混淆前景修正模块，识别并修正易混淆的前景块。大量实验表明，该框架显著提升了现有FG-BG分解方法的性能。代码已开源：https://github.com/lounwb/FoBoR。
原文摘要 · Abstract (English)
CLIP-based foreground-background (FG-BG) decomposition methods have demonstrated remarkable effectiveness in improving few-shot out-of-distribution (OOD) detection performance. However, existing approaches still suffer from several limitations. For background regions obtained from decomposition, existing methods adopt a uniform suppression strategy for all patches, overlooking the varying contributions of different patches to the prediction. For foreground regions, existing methods fail to adequately consider that some local patches may exhibit appearance or semantic similarity to other classes, which may mislead the training process. To address these issues, we propose a new plug-and-play framework. This framework consists of three core components: (1) a Foreground-Background Decomposition module, which follows previous FG-BG methods to separate an image into foreground and background regions; (2) an Adaptive Background Suppression module, which adaptively weights patch classification entropy; and (3) a Confusable Foreground Rectification module, which identifies and rectifies confusable foreground patches. Extensive experimental results demonstrate that the proposed plug-and-play framework significantly improves the performance of existing FG-BG decomposition methods. Code is available at: https://github.com/lounwb/FoBoR.
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。