Text-Guided Multimodal Unified Industrial Anomaly Detection

2026-04-24

AI Summary

This work addresses the challenges of ambiguous cross-modal alignment and insufficient geometric modeling in existing unsupervised industrial anomaly detection methods leveraging RGB-3D multimodal data. To overcome these limitations, we propose the first unified multimodal anomaly detection framework guided by textual semantics, which enables accurate single-model, cross-category anomaly detection through a geometry-aware cross-modal mapper and an object-conditioned textual feature adapter. By introducing textual semantics to guide multimodal alignment, our approach breaks the conventional constraint that single models are limited to single-category detection, thereby establishing a unified paradigm for unsupervised multimodal anomaly detection. Extensive experiments on the MVTec 3D-AD and Eyecandies datasets demonstrate state-of-the-art performance in both anomaly classification and localization under the unsupervised setting.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Industrial anomaly detection based on RGB-3D multimodal data has emerged as a mainstream paradigm for intelligent quality inspection. However, existing unsupervised methods suffer from two critical limitations: ambiguous cross-modal alignment caused by the lack of high-level semantic guidance and insufficient geometric modeling for RGB-to-3D feature mapping. To address these issues, we propose a unified multimodal industrial anomaly detection framework guided by text semantics. The framework consists of two core modules: a Geometry-Aware Cross-Modal Mapper to preserve geometric structure during modality conversion, and an Object-Conditioned Textual Feature Adaptor to align multimodal features with semantic priors. Furthermore, we establish a unified learning paradigm for multimodal industrial anomaly detection, which breaks the one-model-one-class constraint and enables accurate anomaly detection across diverse classes using a single model. Extensive experiments on the MVTec 3D-AD and Eyecandies datasets demonstrate that our method achieves state-of-the-art performance in classification and localization under unsupervised settings.

Problem

• industrial anomaly detection
• multimodal
• unsupervised learning
• cross-modal alignment
• geometric modeling

Innovation

• text-guided
• multimodal anomaly detection
• geometry-aware mapping
• unified learning paradigm
• unsupervised industrial inspection