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

By integrating certain optimization solvers with deep neural networks, deepunfolding network (DUN) with good interpretability and high performance hasattracted growing attention in compressive sensing (CS). However, existing DUNsoften improve the visual quality at the price of a large number of parametersand have the problem of feature information loss during iteration. In thispaper, we propose an Optimization-inspired Cross-attention Transformer (OCT)module as an iterative process, leading to a lightweight OCT-based UnfoldingFramework (OCTUF) for image CS. Specifically, we design a novel Dual CrossAttention (Dual-CA) sub-module, which consists of an Inertia-Supplied CrossAttention (ISCA) block and a Projection-Guided Cross Attention (PGCA) block.ISCA block introduces multi-channel inertia forces and increases the memoryeffect by a cross attention mechanism between adjacent iterations. And, PGCAblock achieves an enhanced information interaction, which introduces theinertia force into the gradient descent step through a cross attention block.Extensive CS experiments manifest that our OCTUF achieves superior performancecompared to state-of-the-art methods while training lower complexity. Codes areavailable at https://github.com/songjiechong/OCTUF.

Optimization-Inspired Cross-Attention Transformer for Compressive Sensing