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

Unsupervised object discovery and localization aims to detect or segmentobjects in an image without any supervision. Recent efforts have demonstrated anotable potential to identify salient foreground objects by utilizingself-supervised transformer features. However, their scopes only build uponpatch-level features within an image, neglecting region/image-level andcross-image relationships at a broader scale. Moreover, these methods cannotdifferentiate various semantics from multiple instances. To address theseproblems, we introduce Hierarchical mErging framework via contrAstive grouPing(HEAP). Specifically, a novel lightweight head with cross-attention mechanismis designed to adaptively group intra-image patches into semantically coherentregions based on correlation among self-supervised features. Further, to ensurethe distinguishability among various regions, we introduce a region-levelcontrastive clustering loss to pull closer similar regions across images. Also,an image-level contrastive loss is present to push foreground and backgroundrepresentations apart, with which foreground objects and background areaccordingly discovered. HEAP facilitates efficient hierarchical imagedecomposition, which contributes to more accurate object discovery while alsoenabling differentiation among objects of various classes. Extensiveexperimental results on semantic segmentation retrieval, unsupervised objectdiscovery, and saliency detection tasks demonstrate that HEAP achievesstate-of-the-art performance.

HEAP: Unsupervised Object Discovery and Localization with Contrastive Grouping