High-Fidelity Zero-Shot Texture Anomaly Localization Using Feature Correspondence Analysis
This work addresses the problem of identifying abnormal regions in homogeneous images for applications like quality control, though it appears incremental as it builds on existing zero-shot methods with a novel mapping technique.
The paper tackles zero-shot anomaly localization on textures by using a bijective mapping from the 1D Wasserstein Distance to pinpoint non-conformities in local contexts, achieving over a 40% reduction in error over previous state-of-the-art on the MVTec AD dataset.
We propose a novel method for Zero-Shot Anomaly Localization on textures. The task refers to identifying abnormal regions in an otherwise homogeneous image. To obtain a high-fidelity localization, we leverage a bijective mapping derived from the 1-dimensional Wasserstein Distance. As opposed to using holistic distances between distributions, the proposed approach allows pinpointing the non-conformity of a pixel in a local context with increased precision. By aggregating the contribution of the pixel to the errors of all nearby patches we obtain a reliable anomaly score estimate. We validate our solution on several datasets and obtain more than a 40% reduction in error over the previous state of the art on the MVTec AD dataset in a zero-shot setting. Also see https://reality.tf.fau.de/pub/ardelean2024highfidelity.html.