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

One-to-one (o2o) label assignment plays a key role for transformer basedend-to-end detection, and it has been recently introduced in fullyconvolutional detectors for end-to-end dense detection. However, o2o candegrade the feature learning efficiency due to the limited number of positivesamples. Though extra positive samples are introduced to mitigate this issue inrecent DETRs, the computation of self- and cross- attentions in the decoderlimits its practical application to dense and fully convolutional detectors. Inthis work, we propose a simple yet effective one-to-few (o2f) label assignmentstrategy for end-to-end dense detection. Apart from defining one positive andmany negative anchors for each object, we define several soft anchors, whichserve as positive and negative samples simultaneously. The positive andnegative weights of these soft anchors are dynamically adjusted during trainingso that they can contribute more to ``representation learning'' in the earlytraining stage, and contribute more to ``duplicated prediction removal'' in thelater stage. The detector trained in this way can not only learn a strongfeature representation but also perform end-to-end dense detection. Experimentson COCO and CrowdHuman datasets demonstrate the effectiveness of the o2fscheme. Code is available at https://github.com/strongwolf/o2f.

One-to-Few Label Assignment for End-to-End Dense Detection