MAGIC通过三重机制生成高保真且多样化的缺陷图像，提升工业质检模型泛化能力。
MAGIC: Few-Shot Mask-Guided Anomaly Inpainting with Prompt Perturbation, Spatially Adaptive Guidance, and Context Awareness
- 用高斯提示扰动学习真实缺陷的平滑流形，避免少样本过拟合。
- 空间自适应引导在异常与背景区域施加不同强度指导，保持细节清晰。
- 上下文感知掩码对齐使缺陷位置更自然，适合下游检测任务部署。
少样本异常生成是工业质量控制的关键挑战。尽管扩散模型具有潜力，现有方法存在局限：全局提示引导会破坏正常区域，现有基于修复的方法常缺乏对下游模型至关重要的分布内多样性。我们提出MAGIC，一个微调的修复框架，能生成严格遵循掩码且高度多样化的高质量异常。MAGIC引入三个互补组件：(i) 高斯提示扰动，通过在真实异常的平滑流形上学习和采样，防止少样本场景下的过拟合；(ii) 空间自适应引导，在异常与背景区域应用不同强度的引导；(iii) 上下文感知掩码对齐，将掩码重新定位至目标物体中更合理的置放位置。在统一评估协议下，MAGIC在多个异常数据集上的下游任务中均优于现有最先进方法。
原文摘要 · Abstract (English)
Few-shot anomaly generation is a key challenge in industrial quality control. Although diffusion models are promising, existing methods struggle: global prompt-guided approaches corrupt normal regions, and existing inpainting-based methods often lack the in-distribution diversity essential for robust downstream models. We propose MAGIC, a fine-tuned inpainting framework that generates high-fidelity anomalies that strictly adhere to the mask while maximizing this diversity. MAGIC introduces three complementary components: (i) Gaussian prompt perturbation, which prevents model overfitting in the few-shot setting by learning and sampling from a smooth manifold of realistic anomalies, (ii) spatially adaptive guidance that applies distinct guidance strengths to the anomaly and background regions, and (iii) context-aware mask alignment to relocate masks for plausible placement within the host object. Under consistent identical evaluation protocol, MAGIC outperforms state-of-the-art methods on diverse anomaly datasets in downstream tasks.
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。