CoGeoAD: Hierarchical Color-Geometric Fusion with Multi-View Attention for Zero-Shot 3D Anomaly Detection

2026-06-23

AI Summary

This work addresses the challenge of effectively integrating 2D texture and 3D geometry in zero-shot 3D anomaly detection by proposing a CLIP-based, multi-stage color-geometry hierarchical fusion framework. The method constructs pixel-aligned multi-view image pairs and leverages a data-driven multi-view attention mechanism to adaptively aggregate cross-view 3D information, enabling joint modeling of both textural and structural anomalies. Evaluated on the MVTec3D-AD and Eyecandies benchmarks, the approach achieves state-of-the-art performance, significantly enhancing the detection capability for multimodal anomalies in complex industrial scenarios.

Abstract

Zero-shot 3D anomaly detection is essential for industrial quality inspection, where labeled anomaly samples are scarce. Meanwhile, existing methods lack an effective mechanism to fuse complementary 2D color images with 3D geometric structures, limiting their ability to detect both surface and structural defects in a unified framework. To address these issues, we propose CoGeoAD, a unified CLIP-based framework that fuses color and geometric features by constructing pixel-aligned paired multi-view images. The framework introduces a Data-Driven Multi-View Attention (MVA) mechanism to adaptively aggregate 3D features and a Multi-Stage Color-Geometric Fusion (MS-CGF) module to hierarchically integrate multi-level features from both modalities. Extensive experiments on the MVTec3D-AD and Eyecandies benchmarks demonstrate that CoGeoAD achieves state-of-the-art performance, effectively capturing both structural and textural anomalies in complex industrial scenarios. our source code is available at https://github.com/kingdomShu/CoGeoAD.

Problem

• zero-shot 3D anomaly detection
• color-geometric fusion
• multi-view attention
• industrial quality inspection
• 3D defect detection

Innovation

• Zero-shot 3D anomaly detection
• Color-geometric fusion
• Multi-view attention
• CLIP-based framework
• Multi-stage feature integration