🤖 AI Summary
This paper addresses the lack of interpretability in black-box models for visual anomaly detection (VAD). It presents a systematic survey of explainable VAD methods across image and video modalities. For the first time, it establishes a unified multimodal interpretability framework, clarifying how gradient- and feature-based attribution, attention visualization, generative explanations, concept activation mapping, and counterfactual reasoning are adapted to VAD—and identifying cross-modal transfer constraints. The survey integrates major datasets, evaluation metrics, and methodological taxonomies into a structured knowledge graph. Additionally, it releases an authoritative open-source resource repository (GitHub Awesome-XAD), featuring a benchmarking platform for interpretability evaluation and a method selection guide. These contributions provide both theoretical foundations and practical tools to advance trustworthy, deployable VAD research in academia and industry.
📝 Abstract
Anomaly detection and localization of visual data, including images and videos, are of great significance in both machine learning academia and applied real-world scenarios. Despite the rapid development of visual anomaly detection techniques in recent years, the interpretations of these black-box models and reasonable explanations of why anomalies can be distinguished out are scarce. This paper provides the first survey concentrated on explainable visual anomaly detection methods. We first introduce the basic background of image-level and video-level anomaly detection. Then, as the main content of this survey, a comprehensive and exhaustive literature review of explainable anomaly detection methods for both images and videos is presented. Next, we analyze why some explainable anomaly detection methods can be applied to both images and videos and why others can be only applied to one modality. Additionally, we provide summaries of current 2D visual anomaly detection datasets and evaluation metrics. Finally, we discuss several promising future directions and open problems to explore the explainability of 2D visual anomaly detection. The related resource collection is given at https://github.com/wyzjack/Awesome-XAD.
