Skip to yearly menu bar Skip to main content

Workshop: Workshop on Distribution Shifts: Connecting Methods and Applications

Env-Aware Anomaly Detection: Ignore Style Changes, Stay True to Content!

Stefan Smeu · Elena Burceanu · Andrei L Nicolicioiu · Emanuela Haller


We introduce a formalization and benchmark for the unsupervised anomaly detection task in the distribution-shift scenario. Our work builds upon the iWildCam dataset, and, to the best of our knowledge, we are the first to propose such an approach for visual data. We empirically validate that environment-aware methods perform better in such cases when compared with the basic Empirical Risk Minimization (ERM). We next propose an extension for generating positive samples for contrastive methods that considers the environment labels when training, improving the ERM baseline score by 8.7%.

Chat is not available.