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

Convolutional neural networks (CNNs) demand huge DRAM bandwidth forcomputational imaging tasks, and block-based processing has recently beenapplied to greatly reduce the bandwidth. However, the induced additionalcomputation for feature recomputing or the large SRAM for feature reusing willdegrade the performance or even forbid the usage of state-of-the-art models. Inthis paper, we address these issues by considering the overheads and hardwareconstraints in advance when constructing CNNs. We investigate a novel modelfamily---ERNet---which includes temporary layer expansion as another means forincreasing model capacity. We analyze three ERNet variants in terms of hardwarerequirement and introduce a hardware-aware model optimization procedure.Evaluations on Full HD and 4K UHD applications will be given to show theeffectiveness in terms of image quality, pixel throughput, and SRAM usage. Theresults also show that, for block-based inference, ERNet can outperform thestate-of-the-art FFDNet and EDSR-baseline models for image denoising andsuper-resolution respectively.

ERNet Family: Hardware-Oriented CNN Models for Computational Imaging Using Block-Based Inference