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

In this paper, we propose a simple yet effective transformer framework forself-supervised learning called DenseDINO to learn dense visualrepresentations. To exploit the spatial information that the dense predictiontasks require but neglected by the existing self-supervised transformers, weintroduce point-level supervision across views in a novel token-based way.Specifically, DenseDINO introduces some extra input tokens called referencetokens to match the point-level features with the position prior. With thereference token, the model could maintain spatial consistency and deal withmulti-object complex scene images, thus generalizing better on dense predictiontasks. Compared with the vanilla DINO, our approach obtains competitiveperformance when evaluated on classification in ImageNet and achieves a largemargin (+7.2% mIoU) improvement in semantic segmentation on PascalVOC under thelinear probing protocol for segmentation.

DenseDINO: Boosting Dense Self-Supervised Learning with Token-Based Point-Level Consistency