Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Jiang, Sheng, Ning, Yuanmin, Huang, Bingxi, Chen, Peiyin, Chen, Zhaohui
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2510.00603
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915528003551232
author Jiang, Sheng
Ning, Yuanmin
Huang, Bingxi
Chen, Peiyin
Chen, Zhaohui
author_facet Jiang, Sheng
Ning, Yuanmin
Huang, Bingxi
Chen, Peiyin
Chen, Zhaohui
contents Automated structural defect annotation is essential for ensuring infrastructure safety while minimizing the high costs and inefficiencies of manual labeling. A novel agentic annotation framework, Agent-based Defect Pattern Tagger (ADPT), is introduced that integrates Large Vision-Language Models (LVLMs) with a semantic pattern matching module and an iterative self-questioning refinement mechanism. By leveraging optimized domain-specific prompting and a recursive verification process, ADPT transforms raw visual data into high-quality, semantically labeled defect datasets without any manual supervision. Experimental results demonstrate that ADPT achieves up to 98% accuracy in distinguishing defective from non-defective images, and 85%-98% annotation accuracy across four defect categories under class-balanced settings, with 80%-92% accuracy on class-imbalanced datasets. The framework offers a scalable and cost-effective solution for high-fidelity dataset construction, providing strong support for downstream tasks such as transfer learning and domain adaptation in structural damage assessment.
format Preprint
id arxiv_https___arxiv_org_abs_2510_00603
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LVLMs as inspectors: an agentic framework for category-level structural defect annotation
Jiang, Sheng
Ning, Yuanmin
Huang, Bingxi
Chen, Peiyin
Chen, Zhaohui
Computer Vision and Pattern Recognition
Automated structural defect annotation is essential for ensuring infrastructure safety while minimizing the high costs and inefficiencies of manual labeling. A novel agentic annotation framework, Agent-based Defect Pattern Tagger (ADPT), is introduced that integrates Large Vision-Language Models (LVLMs) with a semantic pattern matching module and an iterative self-questioning refinement mechanism. By leveraging optimized domain-specific prompting and a recursive verification process, ADPT transforms raw visual data into high-quality, semantically labeled defect datasets without any manual supervision. Experimental results demonstrate that ADPT achieves up to 98% accuracy in distinguishing defective from non-defective images, and 85%-98% annotation accuracy across four defect categories under class-balanced settings, with 80%-92% accuracy on class-imbalanced datasets. The framework offers a scalable and cost-effective solution for high-fidelity dataset construction, providing strong support for downstream tasks such as transfer learning and domain adaptation in structural damage assessment.
title LVLMs as inspectors: an agentic framework for category-level structural defect annotation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2510.00603