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

Generative adversarial networks (GANs) have promoted remarkable advances insingle-image super-resolution (SR) by recovering photo-realistic images.However, high memory consumption of GAN-based SR (usually generators) causesperformance degradation and more energy consumption, hindering the deploymentof GAN-based SR into resource-constricted mobile devices. In this paper, wepropose a novel compression framework \textbf{M}ulti-scale \textbf{F}eature\textbf{A}ggregation Net based \textbf{GAN} (MFAGAN) for reducing the memoryaccess cost of the generator. First, to overcome the memory explosion of denseconnections, we utilize a memory-efficient multi-scale feature aggregation netas the generator. Second, for faster and more stable training, our methodintroduces the PatchGAN discriminator. Third, to balance the studentdiscriminator and the compressed generator, we distill both the generator andthe discriminator. Finally, we perform a hardware-aware neural architecturesearch (NAS) to find a specialized SubGenerator for the target mobile phone.Benefiting from these improvements, the proposed MFAGAN achieves up to\textbf{8.3}$\times$ memory saving and \textbf{42.9}$\times$ computationreduction, with only minor visual quality degradation, compared with ESRGAN.Empirical studies also show $\sim$\textbf{70} milliseconds latency on QualcommSnapdragon 865 chipset.

MFAGAN: A Compression Framework for Memory-Efficient On-Device Super-Resolution GAN