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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Compared with model architectures, the training process, which is alsocrucial to the success of detectors, has received relatively less attention inobject detection. In this work, we carefully revisit the standard trainingpractice of detectors, and find that the detection performance is often limitedby the imbalance during the training process, which generally consists in threelevels - sample level, feature level, and objective level. To mitigate theadverse effects caused thereby, we propose Libra R-CNN, a simple but effectiveframework towards balanced learning for object detection. It integrates threenovel components: IoU-balanced sampling, balanced feature pyramid, and balancedL1 loss, respectively for reducing the imbalance at sample, feature, andobjective level. Benefitted from the overall balanced design, Libra R-CNNsignificantly improves the detection performance. Without bells and whistles,it achieves 2.5 points and 2.0 points higher Average Precision (AP) than FPNFaster R-CNN and RetinaNet respectively on MSCOCO.

Libra R-CNN: Towards Balanced Learning for Object Detection