Learned Image Compression With Discretized Gaussian Mixture Likelihoods and Attention Modules

By Zhengxue Cheng, Heming Sun, Masaru Takeuchi, Jiro Katto. Computer Vision and Pattern Recognition, 2020

1,286 citations recorded by the source corpus.

Image compression is a fundamental research field and many well-known compression standards have been developed for many decades. Recently, learned compression methods exhibit a fast development trend with promising results. However, there is still a performance gap between learned compression algorithms and reigning compression standards, especially in terms of widely used PSNR metric. In this paper, we explore the remaining redundancy of recent learned compression algorithms. We have found accurate entropy models for rate estimation largely affect the optimization of network parameters and thus affect the rate-distortion performance. Therefore, in this paper, we propose to use discretized Gaussian Mixture Likelihoods to parameterize the distributions of latent codes, which can achieve a more accurate and flexible entropy model. Besides, we take advantage of recent attention modules and incorporate them into network architecture to enhance the performance. Experimental results demonstrate our proposed method achieves a state-of-the-art performance compared to existing learned compression methods on both Kodak and high-resolution datasets. To our knowledge our approach is the first work to achieve comparable performance with latest compression standard Versatile Video Coding (VVC) regarding PSNR. More importantly, our approach generates more visually pleasant results when optimized by MS-SSIM.

Subfields: Computer Vision and Pattern Recognition, Signal Processing

View Paper: https://www.semanticscholar.org/paper/5df999a1077524c9178b43c84f80492f8aaa5a7e

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

In Signal Processing

Citations in the corpus

Checkerboard Context Model for Efficient Learned Image Compression (2021, methodology): “tion on computational efficiency, most of the SOTA methods still rely on it [26, 10, 20].”
COIN: COmpression with Implicit Neural representations (2021, methodology): “We compare our model against three autoencoder based neural compression baselines which we refer to as BMS [2], MBT [24], and CST [7].”
Finite Scalar Quantization: VQ-VAE Made Simple (2023, background): “Neural compression Many works (Ball´e et al., 2016; Minnen et al., 2018; Lu et al., 2019; Mentzer et al., 2020; Cheng et al., 2020) rely on unbounded scalar quantization and constrain the entropy of the quantized representation to prevent spreading to all integers.”
CompressAI: a PyTorch library and evaluation platform for end-to-end compression research (2020, methodology): “Following the success of deep neural networks in out-performing most conventional approaches in computer vision applications, Artificial Neural Network (ANN) based codecs have recently demonstrated impressive results for compressing images [1–5].”
Deep Contextual Video Compression (2021, methodology): “We use cheng2020-anchor [20] for MSE target and use hyperprior [13] for MS-SSIM target.”
Toward Semantic Communications: Deep Learning-Based Image Semantic Coding (2022, methodology): “The work [44] proposed to use discretized Gaussian Mixture Likelihoods to parameterize the distributions of latent codes, which can achieve a more”
QARV: Quantization-Aware ResNet VAE for Lossy Image Compression (2023, background): “To improve compression efﬁciency, existing works have explored efﬁcient neural network blocks, such as residual networks [21], self-attention [22], and transformers [23], [24], [25], as well as developed more expressive entropy models, such as hierarchical [7], [26] and autoregressive models [8],…”
The Stable Signature: Rooting Watermarks in Latent Diffusion Models (2023, methodology): “Neural auto-encoders [4, 11, 20] follow the same trend , except with the one used by LDM (‘KL-f8’ for our LDM).”
ELIC: Efficient Learned Image Compression with Unevenly Grouped Space-Channel Contextual Adaptive Coding (2022, neutral): “Note that this is different from existing investigations that view GDN as activation function [14, 15, 17, 27].”