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

Salient object detection on RGB-D images is an active topic in computervision. Although the existing methods have achieved appreciable performance,there are still some challenges. The locality of convolutional neural networkrequires that the model has a sufficiently deep global receptive field, whichalways leads to the loss of local details. To address the challenge, we proposea novel Group Transformer Network (GroupTransNet) for RGB-D salient objectdetection. This method is good at learning the long-range dependencies of crosslayer features to promote more perfect feature expression. At the beginning,the features of the slightly higher classes of the middle three levels and thelatter three levels are soft grouped to absorb the advantages of the high-levelfeatures. The input features are repeatedly purified and enhanced by theattention mechanism to purify the cross modal features of color modal and depthmodal. The features of the intermediate process are first fused by the featuresof different layers, and then processed by several transformers in multiplegroups, which not only makes the size of the features of each scale unified andinterrelated, but also achieves the effect of sharing the weight of thefeatures within the group. The output features in different groups complete theclustering staggered by two owing to the level difference, and combine with thelow-level features. Extensive experiments demonstrate that GroupTransNetoutperforms the comparison models and achieves the new state-of-the-artperformance.

GroupTransNet: Group Transformer Network for RGB-D Salient Object Detection