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

This paper presents a cross channel context model for latents in deep imagecompression. Generally, deep image compression is based on an autoencoderframework, which transforms the original image to latents at the encoder andrecovers the reconstructed image from the quantized latents at the decoder. Thetransform is usually combined with an entropy model, which estimates theprobability distribution of the quantized latents for arithmetic coding.Currently, joint autoregressive and hierarchical prior entropy models arewidely adopted to capture both the global contexts from the hyper latents andthe local contexts from the quantized latent elements. For the local contexts,the widely adopted 2D mask convolution can only capture the spatial context.However, we observe that there are strong correlations between differentchannels in the latents. To utilize the cross channel correlations, we proposeto divide the latents into several groups according to channel index and codethe groups one by one, where previously coded groups are utilized to providecross channel context for the current group. The proposed cross channel contextmodel is combined with the joint autoregressive and hierarchical prior entropymodel. Experimental results show that, using PSNR as the distortion metric, thecombined model achieves BD-rate reductions of 6.30% and 6.31% over the baselineentropy model, and 2.50% and 2.20% over the latest video coding standardVersatile Video Coding (VVC) for the Kodak and CVPR CLIC2020 professionaldataset, respectively. In addition, when optimized for the MS-SSIM metric, ourapproach generates visually more pleasant reconstructed images.

A Cross Channel Context Model for Latents in Deep Image Compression