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

The Transformer architecture has witnessed a rapid development in recentyears, outperforming the CNN architectures in many computer vision tasks, asexemplified by the Vision Transformers (ViT) for image classification. However,existing visual transformer models aim to extract semantic information forhigh-level tasks, such as classification and detection.These methods ignore theimportance of the spatial resolution of the input image, thus sacrificing thelocal correlation information of neighboring pixels. In this paper, we proposea Patch Pyramid Transformer(PPT) to effectively address the aboveissues.Specifically, we first design a Patch Transformer to transform the imageinto a sequence of patches, where transformer encoding is performed for eachpatch to extract local representations. In addition, we construct a PyramidTransformer to effectively extract the non-local information from the entireimage. After obtaining a set of multi-scale, multi-dimensional, and multi-anglefeatures of the original image, we design the image reconstruction network toensure that the features can be reconstructed into the original input. Tovalidate the effectiveness, we apply the proposed Patch Pyramid Transformer toimage fusion tasks. The experimental results demonstrate its superiorperformance, compared to the state-of-the-art fusion approaches, achieving thebest results on several evaluation indicators. Thanks to the underlyingrepresentational capacity of the PPT network, it can directly be applied todifferent image fusion tasks without redesigning or retraining the network.

PPT Fusion: Pyramid Patch Transformerfor a Case Study in Image Fusion