Quantifying Statistical Significance in Diffusion-Based Anomaly Localization via Selective Inference
This addresses reliability concerns in high-stakes decision-making tasks like medical diagnosis and industrial inspection, though it is incremental as it builds on existing diffusion-based methods.
The paper tackles the problem of unreliable anomaly localization in images by proposing a statistical framework based on selective inference to quantify the significance of detected anomalous regions, resulting in effective control of false positive detection rates as demonstrated in medical and industrial applications.
Anomaly localization in images (identifying regions that deviate from expected patterns) is vital in applications such as medical diagnosis and industrial inspection. A recent trend is the use of image generation models in anomaly localization, where these models generate normal-looking counterparts of anomalous images, thereby allowing flexible and adaptive anomaly localization. However, these methods inherit the uncertainty and bias implicitly embedded in the employed generative model, raising concerns about the reliability. To address this, we propose a statistical framework based on selective inference to quantify the significance of detected anomalous regions. Our method provides $p$-values to assess the false positive detection rates, providing a principled measure of reliability. As a proof of concept, we consider anomaly localization using a diffusion model and its applications to medical diagnoses and industrial inspections. The results indicate that the proposed method effectively controls the risk of false positive detection, supporting its use in high-stakes decision-making tasks.