Single-Step Reconstruction-Free Anomaly Detection and Segmentation via Diffusion Models

2025-08-06

AI Summary

Existing diffusion-based anomaly detection methods rely on image reconstruction, suffering from three key bottlenecks: high computational cost, reconstruction artifacts, and difficulty in selecting optimal intermediate noise layers. This paper proposes RADAR—a reconstruction-free, single-step inference framework for anomaly detection and segmentation. RADAR leverages multi-scale attention feature discrepancies induced by the forward diffusion process to directly generate pixel-level anomaly maps, completely bypassing reverse sampling. This design eliminates dependence on anomaly priors and manual noise-layer tuning, significantly improving real-time performance and robustness. Evaluated on MVTec-AD and a 3D-printed materials dataset, RADAR achieves 7% and 13% improvements in F1-score, respectively, and consistently outperforms state-of-the-art diffusion-based and statistical learning methods across accuracy, precision, and recall—setting new SOTA results.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Machine Learning: Adversarial Learning & Robustness

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Generative models have demonstrated significant success in anomaly detection and segmentation over the past decade. Recently, diffusion models have emerged as a powerful alternative, outperforming previous approaches such as GANs and VAEs. In typical diffusion-based anomaly detection, a model is trained on normal data, and during inference, anomalous images are perturbed to a predefined intermediate step in the forward diffusion process. The corresponding normal image is then reconstructed through iterative reverse sampling. However, reconstruction-based approaches present three major challenges: (1) the reconstruction process is computationally expensive due to multiple sampling steps, making real-time applications impractical; (2) for complex or subtle patterns, the reconstructed image may correspond to a different normal pattern rather than the original input; and (3) Choosing an appropriate intermediate noise level is challenging because it is application-dependent and often assumes prior knowledge of anomalies, an assumption that does not hold in unsupervised settings. We introduce Reconstruction-free Anomaly Detection with Attention-based diffusion models in Real-time (RADAR), which overcomes the limitations of reconstruction-based anomaly detection. Unlike current SOTA methods that reconstruct the input image, RADAR directly produces anomaly maps from the diffusion model, improving both detection accuracy and computational efficiency. We evaluate RADAR on real-world 3D-printed material and the MVTec-AD dataset. Our approach surpasses state-of-the-art diffusion-based and statistical machine learning models across all key metrics, including accuracy, precision, recall, and F1 score. Specifically, RADAR improves F1 score by 7% on MVTec-AD and 13% on the 3D-printed material dataset compared to the next best model. Code available at: https://github.com/mehrdadmoradi124/RADAR

Problem

• Eliminates need for iterative reconstruction in anomaly detection
• Addresses computational inefficiency in real-time anomaly segmentation
• Overcomes challenge of selecting noise levels for anomaly identification

Innovation

• Uses diffusion models for anomaly detection
• Eliminates reconstruction step for efficiency
• Directly generates anomaly maps via attention