A Survey on Industrial Anomalies Synthesis

2025-02-23

AI Summary

Existing industrial anomaly synthesis research suffers from methodological fragmentation, absence of a systematic taxonomy, and insufficient exploration of cross-modal and vision-language model applications. This paper proposes the first fine-grained Industrial Anomaly Synthesis (IAS) taxonomy tailored to industrial scenarios, categorizing approaches into four paradigms: handcrafted methods, distributional assumptions, generative models (GANs, VAEs, diffusion models), and vision-language models (e.g., CLIP, Flamingo)—with explicit inclusion of cross-modal synthesis and large-model-driven directions. We provide a unified critical review of over 40 representative methods, establishing an extensible IAS knowledge framework. Furthermore, we open-source a comprehensive codebase and resource repository on GitHub, offering a structured guideline for algorithm selection, benchmark development, and industrial deployment. This work lays foundational infrastructure for rigorous, reproducible, and scalable research in industrial anomaly synthesis.

Technology Category

• Computer Vision: Computational Photography, Image & Video Synthesis
• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

This paper comprehensively reviews anomaly synthesis methodologies. Existing surveys focus on limited techniques, missing an overall field view and understanding method interconnections. In contrast, our study offers a unified review, covering about 40 representative methods across Hand-crafted, Distribution-hypothesis-based, Generative models (GM)-based, and Vision-language models (VLM)-based synthesis. We introduce the first industrial anomaly synthesis (IAS) taxonomy. Prior works lack formal classification or use simplistic taxonomies, hampering structured comparisons and trend identification. Our taxonomy provides a fine-grained framework reflecting methodological progress and practical implications, grounding future research. Furthermore, we explore cross-modality synthesis and large-scale VLM. Previous surveys overlooked multimodal data and VLM in anomaly synthesis, limiting insights into their advantages. Our survey analyzes their integration, benefits, challenges, and prospects, offering a roadmap to boost IAS with multimodal learning. More resources are available at https://github.com/M-3LAB/awesome-anomaly-synthesis.

Problem

• Comprehensively reviews anomaly synthesis methodologies
• Introduces first industrial anomaly synthesis taxonomy
• Explores cross-modality synthesis and large-scale VLM

Innovation

• Unified anomaly synthesis review
• First industrial anomaly taxonomy
• Explores cross-modality synthesis