Collaborative Discrepancy Optimization for Reliable Image Anomaly Localization
This addresses the issue of overgeneralization in anomaly localization for industrial inspection applications, representing an incremental improvement with specific gains.
The study tackled the problem of unreliable predictions in unsupervised image anomaly localization due to overgeneralization by proposing collaborative discrepancy optimization (CDO), which optimized normal and abnormal feature distributions using synthetic anomalies, achieving great performance with real-time efficiency on MVTec2D and MVTec3D datasets.
Most unsupervised image anomaly localization methods suffer from overgeneralization because of the high generalization abilities of convolutional neural networks, leading to unreliable predictions. To mitigate the overgeneralization, this study proposes to collaboratively optimize normal and abnormal feature distributions with the assistance of synthetic anomalies, namely collaborative discrepancy optimization (CDO). CDO introduces a margin optimization module and an overlap optimization module to optimize the two key factors determining the localization performance, i.e., the margin and the overlap between the discrepancy distributions (DDs) of normal and abnormal samples. With CDO, a large margin and a small overlap between normal and abnormal DDs are obtained, and the prediction reliability is boosted. Experiments on MVTec2D and MVTec3D show that CDO effectively mitigates the overgeneralization and achieves great anomaly localization performance with real-time computation efficiency. A real-world automotive plastic parts inspection application further demonstrates the capability of the proposed CDO. Code is available on https://github.com/caoyunkang/CDO.