Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

MIRAGE: Model-agnostic Industrial Realistic Anomaly Generation and Evaluation for Visual Anomaly Detection

Jinwei Hu, Francesco Borsatti, Arianna Stropeni, Davide Dalle Pezze, Manuel Barusco, Gian Antonio Susto

Abstract

Industrial visual anomaly detection (VAD) methods are typically trained on normal samples only, yet performance improves substantially when even limited anomalous data is available. Existing anomaly generation approaches either require real anomalous examples, demand expensive hardware, or produce synthetic defects that lack realism. We present MIRAGE (Model-agnostic Industrial Realistic Anomaly Generation and Evaluation), a fully automated pipeline for realistic anomalous image generation and pixel-level mask creation that requires no training and no anomalous images. Our pipeline accesses any generative model as a black box via API calls, uses a VLM for automatic defect prompt generation, and includes a CLIP-based quality filter to retain only well-aligned generated images. For mask generation at scale, we introduce a lightweight, training-free dual-branch semantic change detection module combining text-conditioned Grounding DINO features with fine-grained YOLOv26-Seg structural features. We benchmark four generation methods using Gemini 2.5 Flash Image (Nano Banana) as the generative backbone, evaluating performance on MVTec AD and VisA across two distinct tasks: (i) downstream anomaly segmentation and (ii) visual quality of the generated images, assessed via standard metrics (IS, IC-LPIPS) and a human perceptual study involving 31 participants and 1,550 pairwise votes. The results demonstrate that MIRAGE offers a scalable, accessible foundation for anomaly-aware industrial inspection that requires no real defect data. As a final contribution, we publicly release a large-scale dataset comprising 500 image-mask pairs per category for every MVTec AD and VisA class, over 13,000 pairs in total, alongside all generation prompts and pipeline code.

MIRAGE: Model-agnostic Industrial Realistic Anomaly Generation and Evaluation for Visual Anomaly Detection

Abstract

Industrial visual anomaly detection (VAD) methods are typically trained on normal samples only, yet performance improves substantially when even limited anomalous data is available. Existing anomaly generation approaches either require real anomalous examples, demand expensive hardware, or produce synthetic defects that lack realism. We present MIRAGE (Model-agnostic Industrial Realistic Anomaly Generation and Evaluation), a fully automated pipeline for realistic anomalous image generation and pixel-level mask creation that requires no training and no anomalous images. Our pipeline accesses any generative model as a black box via API calls, uses a VLM for automatic defect prompt generation, and includes a CLIP-based quality filter to retain only well-aligned generated images. For mask generation at scale, we introduce a lightweight, training-free dual-branch semantic change detection module combining text-conditioned Grounding DINO features with fine-grained YOLOv26-Seg structural features. We benchmark four generation methods using Gemini 2.5 Flash Image (Nano Banana) as the generative backbone, evaluating performance on MVTec AD and VisA across two distinct tasks: (i) downstream anomaly segmentation and (ii) visual quality of the generated images, assessed via standard metrics (IS, IC-LPIPS) and a human perceptual study involving 31 participants and 1,550 pairwise votes. The results demonstrate that MIRAGE offers a scalable, accessible foundation for anomaly-aware industrial inspection that requires no real defect data. As a final contribution, we publicly release a large-scale dataset comprising 500 image-mask pairs per category for every MVTec AD and VisA class, over 13,000 pairs in total, alongside all generation prompts and pipeline code.
Paper Structure (30 sections, 9 equations, 3 figures, 6 tables)

This paper contains 30 sections, 9 equations, 3 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Industrial Visual Anomaly Detection.
  4. Zero-Shot Anomaly Generation.
  5. Anomaly Generation for VAD.
  6. Change Detection via Feature Differencing.
  7. Methodology
  8. Anomaly Prompt Generation
  9. Anomalous Image Generation
  10. CLIP-Based Quality Filtering
  11. Dual-Branch Semantic Change Detection
  12. Semantic Branch: Grounding DINO Feature Diff.
  13. Structural Branch: YOLOv26-L-Seg Feature Diff.
  14. Mask Fusion and Calibration.
  15. Experimental Setup
  16. ...and 15 more sections

Figures (3)

  • Figure 1: Overview of MIRAGE.Top: A normal image and an anomaly text prompt are fed to Gemini 2.5 Flash Image, which generates a photorealistic anomalous image. Bottom (Mask Creation Pipeline): The normal image, anomalous image, and text prompt are processed through two branches. The semantic branch (left) computes text-conditioned feature differences using Grounding DINO, yielding a coarse semantic anomaly map. The structural branch (right) computes unconditional feature differences using YOLOv26-L-Seg, yielding a fine-grained structural change map. The element-wise product of both maps produces the final anomaly mask.
  • Figure 2: Examples of synthetic anomalies generated. From left to right: (1) original input image, (2) GLASS, (3) RealNet, (4) AnomalyAny, and (5) MIRAGE. The rows represent three object classes: bottle (top row), hazelnuts (mid row) and electronic sensor modules (bottom row). The relative generated anomaly map or mask is overlaid on the bottom right of each image.
  • Figure 3: Different types of synthetic anomalous images and corresponding heatmaps generated by MIRAGE.