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

Traditional semantic segmentation tasks require a large number of labels andare difficult to identify unlearned categories. Few-shot semantic segmentation(FSS) aims to use limited labeled support images to identify the segmentationof new classes of objects, which is very practical in the real world. Previousresearches were primarily based on prototypes or correlations. Due to colors,textures, and styles are similar in the same image, we argue that the queryimage can be regarded as its own support image. In this paper, we proposed theTarget-aware Bi-Transformer Network (TBTNet) to equivalent treat of supportimages and query image. A vigorous Target-aware Transformer Layer (TTL) also bedesigned to distill correlations and force the model to focus on foregroundinformation. It treats the hypercorrelation as a feature, resulting asignificant reduction in the number of feature channels. Benefit from thischaracteristic, our model is the lightest up to now with only 0.4M learnableparameters. Futhermore, TBTNet converges in only 10% to 25% of the trainingepochs compared to traditional methods. The excellent performance on standardFSS benchmarks of PASCAL-5i and COCO-20i proves the efficiency of our method.Extensive ablation studies were also carried out to evaluate the effectivenessof Bi-Transformer architecture and TTL.

Target-aware Bi-Transformer for Few-shot Segmentation