2D_3D Feature Fusion via Cross-Modal Latent Synthesis and Attention Guided Restoration for Industrial Anomaly Detection

2025-10-20 · Usman Ali, Ali Zia, Abdul Rehman, Umer Ramzan, Zohaib Hassan, Talha Sattar, Jing Wang, Wei Xiang arxiv

Industrial anomaly detection (IAD) increasingly benefits from integrating 2D and 3D data, but robust cross-modal fusion remains challenging. We propose a novel unsupervised framework, Multi-Modal Attention-Driven Fusion Restoration (MAFR), which synthesises a unified latent space from RGB images and point clouds using a shared fusion encoder, followed by attention-guided, modality-specific decoders. Anomalies are localised by measuring reconstruction errors between input features and their restored counterparts. Evaluations on the MVTec 3D-AD and Eyecandies benchmarks demonstrate that MAFR achieves state-of-the-art results, with a mean I-AUROC of 0.972 and 0.901, respectively. The framework also exhibits strong performance in few-shot learning settings, and ablation studies confirm the critical roles of the fusion architecture and composite loss. MAFR offers a principled approach for fusing visual and geometric information, advancing the robustness and accuracy of industrial anomaly detection. Code is available at https://github.com/adabrh/MAFR

📄 PDF Abstract BibTeX arXiv:2510.21793

Code (0)

등록된 구현이 없습니다.

Tasks

Few-Shot Learning Anomaly Detection Point Clouds

Wavelet-Fusion Diffusion Model for Multimodal Brain MRI Synthesis with Modality and Metadata Conditioning

2026-05-30 · Muhammad Nabi Yasinzai, Remika Mito, Mangor Pedersen arxiv

Multimodal MRI provides complementary information for neuroimaging analysis, where different imaging modalities capture distinct anatomical, tissue, and pathological features that support the development and evaluation o…

Make-A-Volume: Leveraging Latent Diffusion Models for Cross-Modality 3D Brain MRI Synthesis

2023-07-19 · Lingting Zhu, Zeyue Xue, Zhenchao Jin, Xian Liu 외

Cross-modality medical image synthesis is a critical topic and has the potential to facilitate numerous applications in the medical imaging field. Despite recent successes in deep-learning-based generative models, most c…

Computational Efficiency Image Generation

CoLa-Diff: Conditional Latent Diffusion Model for Multi-Modal MRI Synthesis

2023-03-24 · Lan Jiang, Ye Mao, Xi Chen, Xiangfeng Wang 외

MRI synthesis promises to mitigate the challenge of missing MRI modality in clinical practice. Diffusion model has emerged as an effective technique for image synthesis by modelling complex and variable data distribution…

CoLA Image Generation

Multimodal synthesis of MRI and tabular data with diffusion in a joint latent space via cross-attention

2026-05-05 · Daniel Mensing, Jan Kapar, Jochen G. Hirsch, Matthias Günther 외 arxiv

We propose a multimodal latent diffusion model that jointly synthesizes volumetric magnetic resonance imaging (MRI) and tabular clinical data within a shared latent space via cross-attention. This approach enables cohere…

Representation Learning Image Generation

ShaLa: Multimodal Shared Latent Space Modelling

2025-08-24 · Jiali Cui, Yan-Ying Chen, Yanxia Zhang, Matthew Klenk arxiv

This paper presents a novel generative framework for learning shared latent representations across multimodal data. Many advanced multimodal methods focus on capturing all combinations of modality-specific details across…