Context-adaptive Entropy Model for End-to-end Optimized Image Compression

By Jooyoung Lee, Seunghyun Cho, Seunghwa Beack. 2018

449 citations recorded by the source corpus.

We propose a context-adaptive entropy model for use in end-to-end optimized image compression. Our model exploits two types of contexts, bit-consuming contexts and bit-free contexts, distinguished based upon whether additional bit allocation is required. Based on these contexts, we allow the model to more accurately estimate the distribution of each latent representation with a more generalized form of the approximation models, which accordingly leads to an enhanced compression performance. Based on the experimental results, the proposed method outperforms the traditional image codecs, such as BPG and JPEG2000, as well as other previous artificial-neural-network (ANN) based approaches, in terms of the peak signal-to-noise ratio (PSNR) and multi-scale structural similarity (MS-SSIM) index.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

High-Fidelity Generative Image Compression (2020, methodology): “density of the auto-encoder latents for more efficient arithmetic coding, using hierarchical priors, auto-regression with various context shapes, or a combination thereof [6, 31, 28, 39, 32, 26, 33].”
DeepJSCC-f: Deep Joint Source-Channel Coding of Images With Feedback (2019, result): “As opposed to most current image compression standards, such as JPEG, JPEG2000, BPG, that depend on transform coding followed by quantization and entropy coding, researcher have studied autoencoder-based DNNs with impressive results [31]–[37].”
COIN: COmpression with Implicit Neural representations (2021, methodology): “Learned image compression methods are commonly based on hierarchical variational autoencoders [2, 24, 20] with a learned prior and latent variables being discretized for the purpose of entropy coding.”
Toward Semantic Communications: Deep Learning-Based Image Semantic Coding (2022, background): “Some learning-based codecs are based on convolutional neural networks (CNNs) [17], [31], [42], [43], while others on recurrent neural networks (RNNs) [15], [16].”
Variable Rate Deep Image Compression With a Conditional Autoencoder (2019, background): “In Figure 9, we compare our variable-rate networks optimized for MSE, MS-SSIM and combined MSE+MS-SSIM losses, respectively.”
Checkerboard Context Model for Efficient Learned Image Compression (2021, methodology): “The context model [26, 18], inspired by the concept of context from traditional codecs, is used to predict the probability of unknown codes based on latents that have already been decoded (as shown in Figure 1(a-b)).”
Learned Image Compression With Discretized Gaussian Mixture Likelihoods and Attention Modules (2020, methodology): “RD curves of [21] are from their released source code 1 .”
Lossy Compression for Lossless Prediction (2021, methodology): “Later approaches improved upon the original model by adding a hierarchical layers and autoregressive components (Ballé et al., 2018; Minnen et al., 2018; Lee et al., 2019; Johnston et al., 2018).”