无需真实异常样本，零样本生成逼真局部缺陷。
DeltaDeno: Zero-Shot Anomaly Generation via Delta-Denoising Attribution
- 通过对比双分支扩散过程的去噪差异定位缺陷区域。
- 生成缺陷图像在下游检测任务中显著提升性能。
- 适合无标注异常数据的工业质检场景使用。
异常生成常被建模为少量异常样本的微调，这与异常稀缺性相悖且易过拟合类别先验。本文解决完全无真实异常样本或训练数据的场景，提出免训练的零样本异常生成方法DeltaDeno。该方法通过共享时间步长下最小提示对驱动的双扩散分支，对比去噪差异，累积每步去噪差值形成图像特定的定位图，生成掩码引导后续扩散步骤中的潜在空间补全，同时保留周围上下文以生成逼真局部缺陷。为提升稳定性和控制力，DeltaDeno进行词元级提示优化，强化异常词元并对预测区域施加空间注意力偏置。在多个公开数据集上的实验表明，该方法在生成质量、真实感及下游检测性能上均取得显著一致提升。代码将开源于 https://github.com/CROVO1026/DeltaDeno。
原文摘要 · Abstract (English)
Anomaly generation is often framed as few-shot fine-tuning with anomalous samples, which contradicts the scarcity that motivates generation and tends to overfit category priors. We tackle the setting where no real anomaly samples or training are available. We propose Delta-Denoising (\textbf{DeltaDeno}), a training-free zero-shot anomaly generation method that localizes and edits defects by contrasting two diffusion branches driven by a minimal prompt pair under a shared schedule. By accumulating per-step denoising deltas into an image-specific localization map, we obtain a mask to guide the latent inpainting during later diffusion steps and preserve the surrounding context while generating realistic local defects. To improve stability and control, DeltaDeno performs token-level prompt refinement that aligns shared content and strengthens anomaly tokens, and applies a spatial attention bias restricted to anomaly tokens in the predicted region. Experiments on public datasets show that DeltaDeno achieves great generation, realism and consistent gains in downstream detection performance. Code will be made publicly available at https://github.com/CROVO1026/DeltaDeno.
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。