Zero-Shot Industrial Anomaly Segmentation with Image-Aware Prompt Generation

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Park, SoYoung, Lee, Hyewon, Choi, Mingyu, Han, Seunghoon, Lee, Jong-Ryul, Lim, Sungsu, Kim, Tae-Ho
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910914032173056
author Park, SoYoung
Lee, Hyewon
Choi, Mingyu
Han, Seunghoon
Lee, Jong-Ryul
Lim, Sungsu
Kim, Tae-Ho
author_facet Park, SoYoung
Lee, Hyewon
Choi, Mingyu
Han, Seunghoon
Lee, Jong-Ryul
Lim, Sungsu
Kim, Tae-Ho
contents Anomaly segmentation is essential for industrial quality, maintenance, and stability. Existing text-guided zero-shot anomaly segmentation models are effective but rely on fixed prompts, limiting adaptability in diverse industrial scenarios. This highlights the need for flexible, context-aware prompting strategies. We propose Image-Aware Prompt Anomaly Segmentation (IAP-AS), which enhances anomaly segmentation by generating dynamic, context-aware prompts using an image tagging model and a large language model (LLM). IAP-AS extracts object attributes from images to generate context-aware prompts, improving adaptability and generalization in dynamic and unstructured industrial environments. In our experiments, IAP-AS improves the F1-max metric by up to 10%, demonstrating superior adaptability and generalization. It provides a scalable solution for anomaly segmentation across industries
format Preprint
id arxiv_https___arxiv_org_abs_2504_13560
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Zero-Shot Industrial Anomaly Segmentation with Image-Aware Prompt Generation
Park, SoYoung
Lee, Hyewon
Choi, Mingyu
Han, Seunghoon
Lee, Jong-Ryul
Lim, Sungsu
Kim, Tae-Ho
Computer Vision and Pattern Recognition
Artificial Intelligence
Anomaly segmentation is essential for industrial quality, maintenance, and stability. Existing text-guided zero-shot anomaly segmentation models are effective but rely on fixed prompts, limiting adaptability in diverse industrial scenarios. This highlights the need for flexible, context-aware prompting strategies. We propose Image-Aware Prompt Anomaly Segmentation (IAP-AS), which enhances anomaly segmentation by generating dynamic, context-aware prompts using an image tagging model and a large language model (LLM). IAP-AS extracts object attributes from images to generate context-aware prompts, improving adaptability and generalization in dynamic and unstructured industrial environments. In our experiments, IAP-AS improves the F1-max metric by up to 10%, demonstrating superior adaptability and generalization. It provides a scalable solution for anomaly segmentation across industries
title Zero-Shot Industrial Anomaly Segmentation with Image-Aware Prompt Generation
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2504.13560