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

Existing out-of-distribution (OOD) methods have shown great success onbalanced datasets but become ineffective in long-tailed recognition (LTR)scenarios where 1) OOD samples are often wrongly classified into head classesand/or 2) tail-class samples are treated as OOD samples. To address theseissues, current studies fit a prior distribution of auxiliary/pseudo OOD datato the long-tailed in-distribution (ID) data. However, it is difficult toobtain such an accurate prior distribution given the unknowingness of real OODsamples and heavy class imbalance in LTR. A straightforward solution to avoidthe requirement of this prior is to learn an outlier class to encapsulate theOOD samples. The main challenge is then to tackle the aforementioned confusionbetween OOD samples and head/tail-class samples when learning the outlierclass. To this end, we introduce a novel calibrated outlier class learning(COCL) approach, in which 1) a debiased large margin learning method isintroduced in the outlier class learning to distinguish OOD samples from bothhead and tail classes in the representation space and 2) an outlier-class-awarelogit calibration method is defined to enhance the long-tailed classificationconfidence. Extensive empirical results on three popular benchmarks CIFAR10-LT,CIFAR100-LT, and ImageNet-LT demonstrate that COCL substantially outperformsstate-of-the-art OOD detection methods in LTR while being able to improve theclassification accuracy on ID data. Code is available athttps://github.com/mala-lab/COCL.

Out-of-Distribution Detection in Long-Tailed Recognition with Calibrated Outlier Class Learning