English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Recent advances in visual anomaly detection research have seen AUROC andAUPRO scores on public benchmark datasets such as MVTec and VisA convergetowards perfect recall, giving the impression that these benchmarks arenear-solved. However, high AUROC and AUPRO scores do not always reflectqualitative performance, which limits the validity of these metrics inreal-world applications. We argue that the artificial ceiling imposed by thelack of an adequate evaluation metric restrains progression of the field, andit is crucial that we revisit the evaluation metrics used to rate ouralgorithms. In response, we introduce Per-IMage Overlap (PIMO), a novel metricthat addresses the shortcomings of AUROC and AUPRO. PIMO retains therecall-based nature of the existing metrics but introduces two distinctions:the assignment of curves (and respective area under the curve) is per-image,and its X-axis relies solely on normal images. Measuring recall per imagesimplifies instance score indexing and is more robust to noisy annotations. Aswe show, it also accelerates computation and enables the usage of statisticaltests to compare models. By imposing low tolerance for false positives onnormal images, PIMO provides an enhanced model validation procedure andhighlights performance variations across datasets. Our experiments demonstratethat PIMO offers practical advantages and nuanced performance insights thatredefine anomaly detection benchmarks -- notably challenging the perceptionthat MVTec AD and VisA datasets have been solved by contemporary models.Available on GitHub: https://github.com/jpcbertoldo/aupimo.

AUPIMO: Redefining Visual Anomaly Detection Benchmarks with High Speed and Low Tolerance