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

Recently, long-tailed image classification harvests lots of researchattention, since the data distribution is long-tailed in many real-worldsituations. Piles of algorithms are devised to address the data imbalanceproblem by biasing the training process towards less frequent classes. However,they usually evaluate the performance on a balanced testing set or multipleindependent testing sets having distinct distributions with the training data.Considering the testing data may have arbitrary distributions, existingevaluation strategies are unable to reflect the actual classificationperformance objectively. We set up novel evaluation benchmarks based on aseries of testing sets with evolving distributions. A corpus of metrics aredesigned for measuring the accuracy, robustness, and bounds of algorithms forlearning with long-tailed distribution. Based on our benchmarks, we re-evaluatethe performance of existing methods on CIFAR10 and CIFAR100 datasets, which isvaluable for guiding the selection of data rebalancing techniques. We alsorevisit existing methods and categorize them into four types including databalancing, feature balancing, loss balancing, and prediction balancing,according the focused procedure during the training pipeline.

Revisiting Long-tailed Image Classification: Survey and Benchmarks with New Evaluation Metrics