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

Deep learning-based image compression has made great progresses recently.However, many leading schemes use serial context-adaptive entropy model toimprove the rate-distortion (R-D) performance, which is very slow. In addition,the complexities of the encoding and decoding networks are quite high and notsuitable for many practical applications. In this paper, we introduce fourtechniques to balance the trade-off between the complexity and performance. Weare the first to introduce deformable convolutional module in compressionframework, which can remove more redundancies in the input image, therebyenhancing compression performance. Second, we design a checkerboard contextmodel with two separate distribution parameter estimation networks anddifferent probability models, which enables parallel decoding withoutsacrificing the performance compared to the sequential context-adaptive model.Third, we develop an improved three-step knowledge distillation and trainingscheme to achieve different trade-offs between the complexity and theperformance of the decoder network, which transfers both the final andintermediate results of the teacher network to the student network to help itstraining. Fourth, we introduce $L_{1}$ regularization to make the numericalvalues of the latent representation more sparse. Then we only encode non-zerochannels in the encoding and decoding process, which can greatly reduce theencoding and decoding time. Experiments show that compared to thestate-of-the-art learned image coding scheme, our method can be about 20 timesfaster in encoding and 70-90 times faster in decoding, and our R-D performanceis also $2.3 \%$ higher. Our method outperforms the traditional approach inH.266/VVC-intra (4:4:4) and some leading learned schemes in terms of PSNR andMS-SSIM metrics when testing on Kodak and Tecnick-40 datasets.

Fast and High-Performance Learned Image Compression With Improved Checkerboard Context Model, Deformable Residual Module, and Knowledge Distillation