A Survey on Industrial Anomaly Synthesis

Yanshu Wang; Xichen Xu; Jinbao Wang; Chengbin Ma; Xiaoning Lei; Guoyang Xie; Guannan Jiang; Zhichao Lu

cs.CVFeb 23, 2025

A Survey on Industrial Anomaly Synthesis

Authors: Yanshu Wang, Xichen Xu, Jinbao Wang, Chengbin Ma, Xiaoning Lei, Guoyang Xie, Guannan Jiang, Zhichao Lu

Organizations: Shanghai Jiao Tong University · Department of Intelligent Manufacturing, CATL · City University of Hong Kong

Abstract

This paper presents a comprehensive review of industrial anomaly synthesis (IAS). Existing surveys on industrial anomalies mainly focus on anomaly detection, while IAS is typically treated as an auxiliary component rather than as an independent topic. However, owing to its increasing importance in data augmentation, downstream model training, and controllable industrial inspection, IAS has become a research direction of growing interest. To address the lack of a dedicated review, we survey a broad range of representative methods and organize them into four paradigms: hand-crafted synthesis, distribution hypothesis-based synthesis, generative model (GM)-based synthesis, and vision-language model (VLM)-based synthesis. We further establish a dedicated taxonomy for IAS, which supports more systematic comparison across methods and offers a clearer view of the field's development. Beyond methodological categorization, we summarize the datasets, benchmarks, and evaluation metrics commonly adopted in IAS, and review recent advances in multimodal anomaly synthesis that remain underexplored in prior surveys. We also provide deployment-oriented comparisons and practical guidance by analyzing input requirements, output forms, controllability, cost, downstream tasks, and practical limitations across IAS subcategories. Overall, this survey provides a structured understanding of existing IAS methods, evaluation settings, practical trade-offs, current limitations, and promising future directions, and is intended to serve as a reference for subsequent research in this area. More resources are available at https://github.com/M-3LAB/awesome-anomaly-synthesis.

Figures & tables

Figure 1: Publication and citation trends of industrial anomaly detection and industrial anomaly synthesis from 2019 to 2026* (2026* denotes partial-year statistics up to March 30, 2026) . Industrial anomaly detection shows steady growth, while IAS has attracted increasing attention.

Ref.	Year	Perspective	IAS Cat.	Multimodal	Description
( Chen et al., 2021 )	2021	Detection	0	✗	Visual inspection review with emphasis on augmentation and practical pipelines.
( Xia et al., 2022 )	2022	Detection	4	✗	GAN-centric review covering GAN-based techniques related to AD and synthesis.
( Liu et al., 2024 )	2024	Detection	2	✗	Broad anomaly detection survey; IAS is not a primary focus.
( Cao et al., 2024 )	2024	Detection	4	✗	Discusses anomaly-related topics and mentions multimodal cues; IAS is not systematically reviewed.
( Mao et al., 2025 )	2025	Detection	1	✗	Industrial anomaly detection survey covering paradigms, benchmarks, and metrics; includes synthesis-related discussion.
( Li et al., 2025 )	2025	Detection	1	✗	Industrial visual anomaly detection survey covering learning strategies, generative modeling, and multimodal/VLM-based detection methods.

Table 1: Comparison of previous surveys and our survey.

Figure 2: Publication-source distribution of the 64 representative IAS methods reviewed in this survey. Conference papers constitute 59.4%, primarily from major venues such as CVPR, AAAI, ECCV, and ICCV. Journal articles account for 31.2% and span established venues including IEEE TIP, IEEE TII, IEEE TIM, IEEE TASE, and Pattern Recognition. The remaining 9.4% come from other sources, mainly preprints.

Panel of Figure 3 — Figure 3: A taxonomy of industrial anomaly synthesis (IAS).

Criterion	Hand-crafted	Distribution hypothesis-based	GM-based	VLM-based
Core mechanism	Rule-guided synthesis by manually specifying texture, location, or intensity.	Distribution-guided synthesis by perturbing normal-sample distributions.	Task-specific synthesis with generative models trained or adapted on industrial data.	Multimodal semantic synthesis with large-scale pre-trained vision-language models.
Knowledge source	Human expertise, industrial priors, and image processing rules.	Normal-sample distributions or constrained feature representations.	Target-domain industrial data and task-specific anomaly synthesis priors learned during training.	Pre-trained vision-language knowledge and industrial data.
Large-scale pre-training	Not required.	Usually not required; frozen feature extractors may be used for representation.	Usually not required; synthesis ability mainly comes from target-domain training.	Required; generality and semantic knowledge come from large-scale pre-training.
Task adaptation	Manual adjustment of synthesis rules.	Estimation of normal patterns and construction of abnormal deviations.	Dedicated training, re-training, fine-tuning, or domain adaptation for specific categories, datasets, or anomaly types.	Prompting, instruction tuning, adapters, LoRA, textual inversion, or few-shot reference adaptation.
Modality	Unimodal, based on images.	Unimodal, based on images or visual representations.	Unimodal, based on images.	Multimodal, involving images, text, or other reference cues.
Generalization	Rules need redesign across categories or anomaly types.	Assumption-dependent; robustness decreases when normal patterns change.	Good performance within the trained domain, but limited under cross-category shifts.	Good potential for cross-category, few-shot, and semantically specified anomaly synthesis.

Table 2: Comparison of the four main paradigms for IAS.

Dataset	Year	Venue	Data type	Samples	Classes	IAS relevance				Site
			(Real/Syn./Mixed)			H	DH	GM	VLM
MVTec-AD ( Bergmann et al., 2019 )	2019	CVPR	Real	5,354	15	✓	✓	✓	✓	link
VisA ( Zou et al., 2022 )	2022	ECCV	Real	10,821	12	✓	✓	✓	✓	link
BTAD ( Mishra et al., 2021 )	2021	ISIE	Real	2,830	3	✓	✓	✓	✗	link
MPDD ( Jezek et al., 2021 )	2022	ICUMT	Real	1346	6	✓	✓	✓	✗	link
KSDD ( Tabernik et al., 2020 )	2020	JIM	Real	399	1	✓	✓	✓	✗	link

Table 3: Representative datasets commonly used in IAS-related studies. “Data type” indicates whether the benchmark is primarily based on real industrial imagery (Real), fully synthetic data (Syn.), or a mixture of real and synthetic components (Mixed).

Synthesis / Editing Metrics
Metric	Level	Formula	Remarks/usage
FID	$\downarrow$	$\begin{aligned} \\|\mu_{r}-\mu_{g}\\|_{2}^{2}+\mathrm{Tr}\!\left(\Sigma_{r}+\Sigma_{g}-2(\Sigma_{r}\Sigma_{g})^{1/2}\right)\end{aligned}$	Distributional realism between real and generated features (often computed on full images or cropped regions / patches).
Reference-based LPIPS	$\downarrow$	$d(x,y)$	Perceptual distance between a generated or edited image and a reference image; used for perceptual fidelity, editing consistency, or background preservation.
IC-LPIPS	$\uparrow$	$\frac{1}{K}\sum_{k=1}^{K}\frac{2}{n_{k}(n_{k}-1)}\sum_{i<j}^{n_{k}}d(x_{i}^{(k)},x_{j}^{(k)})$	Intra-cluster LPIPS used for anomaly diversity in recent IAS evaluations. Generated anomalies are grouped by real anomaly references or clusters, and higher intra-cluster pairwise LPIPS indicates greater diversity.
PSNR	$\uparrow$	$10\log_{10}\!\left(\mathrm{MAX}^{2}/\mathrm{MSE}\right)$	Pixel-level reconstruction fidelity; often used to assess background preservation in local editing.
SSIM	$\uparrow$	$\dfrac{(2\mu_{x}\mu_{y}+c_{1})(2\sigma_{xy}+c_{2})}{(\mu_{x}^{2}+\mu_{y}^{2}+c_{1})(\sigma_{x}^{2}+\sigma_{y}^{2}+c_{2})}$	Structure preservation under synthesis or editing.

Table 4: Key metrics commonly used in IAS evaluation.

Figure 4: Hand-crafted synthesis. (a) Overview of three representative hand-crafted pipelines, including self-contained synthesis, external-dependent synthesis, and inpainting-based synthesis. These methods synthesize anomaly content through rule-based image manipulation without training a dedicated generative model. (b) Representative qualitative results adapted from ( Xu et al., 2025b ; Xu et al., 2025c ) , including examples of methods such as CutPaste ( Li et al., 2021 ) and DRAEM ( Zavrtanik et al., 2021a ) .

Family	Resource burden			Practical ceiling		Deployment / limitation analysis
	Train	Infer	Memory	Control	Realism	Suitable scenario	Practical limitation
Self-contained	N/A	Low	Low	Medium	Low	Rapid lightweight augmentation	Limited diversity and synthetic artifacts
External-dependent	N/A	Low	Medium	Medium	Medium	Broad appearance diversification	Boundary inconsistency and semantic mismatch
Inpainting-based	N/A	Low	Low	Medium	Low	Reconstruction-oriented pipelines	Missing/corrupted appearance bias

Table 5: Practical profile of hand-crafted IAS

Figure 5: Distribution hypothesis-based synthesis. (a) Overview of the distribution hypothesis-based paradigm. It constructs abnormal samples or representations by introducing controlled deviations around the distribution of normal data, including prior-dependent strategies guided by explicit geometric assumptions and data-driven strategies based on direct perturbation, replacement, or sampling. (b) Representative qualitative illustrations adapted from ( Chen et al., 2024b ) .

Family	Resource burden			Practical ceiling		Deployment / limitation analysis
	Train	Infer	Memory	Control	Realism	Suitable scenario	Practical limitation
Prior-dependent	Medium	Low	Medium	Medium	Low	Detection under structured normal distributions	Unstable normal-geometry assumption
Data-driven	Medium	Low	Medium	Low	Low	Detection/localization with feature perturbation	Strong reliance on feature quality

Table 6: Practical profile of distribution hypothesis-based IAS

Figure 6: GM-based synthesis. (a) Overview of the GM-based paradigm. It employs deep GMs, such as GANs and diffusion models, for IAS, including full-image synthesis, full-image translation, and local anomaly synthesis with region-wise editing. (b) Representative qualitative results adapted from SDGAN ( Niu et al., 2020 ) , RealNet ( Zhang et al., 2024 ) , and SARD ( Wang et al., 2025 ) .

Family	Resource burden			Practical ceiling		Deployment / limitation analysis
	Train	Infer	Memory	Control	Realism	Suitable scenario	Practical limitation
Full-image synthesis	Medium	Medium	Medium	Low	High	Whole-image anomaly dataset expansion	Mode collapse and domain overfitting
Full-image translation	Medium	Medium	Medium	Low	High	Normal-to-abnormal source translation	Weak spatial controllability
Local anomaly synthesis	Medium	High	High	High	High	Spatially controlled segmentation augmentation	Boundary artifacts and mask dependence

Table 7: Practical profile of GM-based IAS

Figure 7: VLM-based synthesis. (a) Overview of the VLM-based paradigm. It leverages large-scale pre-trained vision–language or multimodal diffusion backbones together with text prompts, reference images, spatial conditions, and other auxiliary cues for IAS. Single-stage methods produce the final abnormal output through one unified dominant synthesis process, whereas multi-stage methods explicitly introduce multiple synthesis stages or coordinated sub-pipelines, such as mask synthesis, anomaly-part synthesis, or progressive refinement. (b) Representative qualitative results adapted from FAST ( Xu et al., 2025b ) , AnomalyDiffusion ( Hu et al., 2024b ) , and SeaS ( Dai et al., 2025 ) .

Family	Resource burden			Practical ceiling		Deployment / limitation analysis
	Train	Infer	Memory	Control	Realism	Suitable scenario	Practical limitation
Single-stage VLM-based	Medium	High	High	Medium	High	Multimodal-conditioned anomaly synthesis	Semantic misalignment and weak masks
Multi-stage VLM-based	High	High	High	High	High	Structured synthesis with aligned supervision	Error propagation and pipeline complexity

Table 8: Practical profile of VLM-based IAS

Paradigm	Subcategory	Real abnormal data	Spatial annotation	Text / multimodal prompt	Reference input	Synthesis level	Output form
Hand-crafted Synthesis	Self-contained	–	Optional	–	–	Pixel	Abnormal image + mask
	External-dependent	–	Optional	–	Required	Pixel	Abnormal image + mask
	Inpainting-based	–	Optional	–	–	Pixel	Abnormal image + mask
Distribution hypothesis-based Synthesis	Prior-dependent	–	–	–	–	Latent	Abnormal feature
	Data-driven	–	–	–	–	Latent	Abnormal feature
GM-based Synthesis	Full-image synthesis	Required	Optional	–	–	Latent + pixel	Abnormal image

Table 9: Input requirements and output forms of IAS subcategories.

Paradigm	Subcategory	Cost	Controllability	Downstream tasks	Key trade-off
Hand-crafted Synthesis	Self-contained	Low	Medium	C/D/S	Good source-background consistency Limited anomaly diversity
	External-dependent	Low	Medium	C/D/S	Richer anomaly appearance Higher risk of local inconsistency
	Inpainting-based	Low	Medium	C/D/S	Simple masking-based synthesis Anomalies may look missing or corrupted
Distribution hypothesis-based Synthesis	Prior-dependent	Medium	Medium	C/D	Clear latent-space structure Limited flexibility for complex abnormal distributions
	Data-driven	Medium	Low	C/D	Adaptive feature perturbation Strong reliance on feature quality
GM-based Synthesis	Full-image synthesis	Medium	Low	C/D	High generation freedom Weaker preservation of source-image structure

Table 10: Deployment-oriented comparison of IAS subcategories.

Synthetic Anomaly Generation

Synthetic anomaly generation creates controlled abnormal examples to train, test, and evaluate detectors when real anomalies are scarce.

See all 65 papers

AnomalyAgent: Agentic Industrial Anomaly Synthesis via Tool-Augmented Reinforcement Learning
Apr 9, 2026Jiaming Su, Tengchao Yang, Ruikang Zhang +3Synthetic Anomaly Generation Tool-Using Agents
Shanghai Jiao Tong University Shanghai, China · Tongji University Shanghai, China · Fudan University Shanghai, China
FLASH: A "Generate Once, Synthesize Many" Framework for Synthetic Anomaly Generation in Industrial Anomaly Detection
Sep 29, 2026Abhay Kumar Das, Rajesh Gangireddy, Ashwin Vaidya +1Synthetic Data Augmentation Industrial Anomaly Detection
Silicon University, Bhubaneswar, India · Intel
UniScale: Arbitrary-Scale Industrial Anomaly Generation
Aug 8, 2026Shilei Zeng, Linxin Guan, Xurui Li +2Industrial Anomaly Detection Synthetic Anomaly Generation
School of Electronic Information and Communications, Huazhong University of Science and Technology

Apr 9, 2026J/K move · Enter open · S save

AnomalyAgent: Agentic Industrial Anomaly Synthesis via Tool-Augmented Reinforcement Learning

Jiaming Su, Tengchao Yang, Ruikang Zhang, Zhengan Yan+2

Shanghai Jiao Tong University Shanghai, China · Tongji University Shanghai, China · Fudan University Shanghai, China

Synthetic Anomaly Generation Tool-Using Agents Industrial Anomaly Detection RL for Generative Models

Open paper AskPDF arXiv

Industrial anomaly generation is a crucial method for alleviating the data scarcity problem in anomaly detection tasks. Most existing anomaly synthesis methods rely on single-step generation mechanisms, lacking complex reasoning and iterative optimization capabilities, making it difficult to generate anomaly samples with high semantic realism. We propose AnomalyAgent, an anomaly synthesis agent with self-reflection, knowledge retrieval, and iterative refinement capabilities, aiming to generate realistic and diverse anomalies. Specifically, AnomalyAgent is equipped with five tools: Prompt Generation (PG), Image Generation (IG), Quality Evaluation (QE), Knowledge Retrieval (KR), and Mask Generation (MG), enabling closed-loop optimization. To improve decision-making and self-reflection, we construct structured trajectories from real anomaly images and design a two-stage training framework: supervised fine-tuning followed by reinforcement learning. This process is driven by a three-part reward mechanism: (1) task rewards to supervise the quality and location rationality of generated anomalies; (2) reflection rewards to train the model's ability to improve anomaly synthesis prompt; (3) behavioral rewards to ensure adherence to the trajectory. On the MVTec-AD dataset, AnomalyAgent achieves IS/IC-L of 2.10/0.33 for anomaly generation, 57.0% classification accuracy using ResNet34, and 99.3%/74.2% AP at the image/pixel level using a simple UNet, surpassing all zero-shot SOTA methods. The code is publicly available at https://github.com/sssjmmm/AnomalyAgent.

Abstract

Synthetic Anomaly Generation

Explore similar work

AnomalyAgent: Agentic Industrial Anomaly Synthesis via Tool-Augmented Reinforcement Learning

FLASH: A "Generate Once, Synthesize Many" Framework for Synthetic Anomaly Generation in Industrial Anomaly Detection

UniScale: Arbitrary-Scale Industrial Anomaly Generation