[论文解读] A Unified End-to-End Framework for Efficient Deep Image Compression

Jiaheng Liu, Guo Lu+2 more|arXiv (Cornell University)|Feb 9, 2020

Advanced Data Compression Techniques参考文献 37被引用 64

一句话总结

本文提出了 EDIC，一种统一的端到端图像压缩框架，使用通道注意力模块、高斯混合熵建模，以及解码端增强，以实现与现有自回归方法相比处于前沿的性能，同时解码速度明显更快，并扩展到视频压缩。

ABSTRACT

Image compression is a widely used technique to reduce the spatial redundancy in images. Recently, learning based image compression has achieved significant progress by using the powerful representation ability from neural networks. However, the current state-of-the-art learning based image compression methods suffer from the huge computational cost, which limits their capacity for practical applications. In this paper, we propose a unified framework called Efficient Deep Image Compression (EDIC) based on three new technologies, including a channel attention module, a Gaussian mixture model and a decoder-side enhancement module. Specifically, we design an auto-encoder style network for learning based image compression. To improve the coding efficiency, we exploit the channel relationship between latent representations by using the channel attention module. Besides, the Gaussian mixture model is introduced for the entropy model and improves the accuracy for bitrate estimation. Furthermore, we introduce the decoder-side enhancement module to further improve image compression performance. Our EDIC method can also be readily incorporated with the Deep Video Compression (DVC) framework to further improve the video compression performance. Simultaneously, our EDIC method boosts the coding performance significantly while bringing slightly increased computational cost. More importantly, experimental results demonstrate that the proposed approach outperforms the current state-of-the-art image compression methods and is up to more than 150 times faster in terms of decoding speed when compared with Minnen's method. The proposed framework also successfully improves the performance of the recent deep video compression system DVC. Our code will be released at https://github.com/liujiaheng/compression.

研究动机与目标

• 通过端到端学习在保持图像质量的同时降低压缩比。
• 提升熵建模的准确性，超越单高斯先验。
• 降低解码复杂度，以实现实际部署。
• 使图像框架能够无缝扩展到视频压缩。
• 演示在与现有视频编解码器（如 DVC）集成时的兼容性与收益。

提出的方法

• 用于图像压缩的自编码器风格网络，包含编码器、解码器、超编码器和超解码器模块。
• 通道注意力模块，用于捕获潜在特征中的通道关系。
• 将高斯混合模型（GMM）作为给定 z 的 y 的熵模型，提升码率估计。
• 解码端增强模块，用于减少伪影并提升重建质量。
• 通过速率失真目标 L = λD + R 进行端到端优化，其中 R 近似等于 H(ŷ) + H(ẑ)。
• 通过复用 EDIC 组件来处理残差和运动信息，将图像框架扩展到视频压缩。

实验结果

研究问题

• RQ1 EDIC 是否能够在解码速度显著更快的情况下，达到与最先进方法等效或更好的速率失真性能？
• RQ2 高斯混合熵模型是否相对于单一高斯提供有意义的比特率节省？
• RQ3 解码端增强模块在成本不可承受的前提下是否提升重建质量？
• RQ4 EDIC 是否能够有效集成到视频压缩框架中以提升性能？

主要发现

方法	解码时间	BDBR
Ballé’s [10]	0.013s	29.87%
Minnen’s [2]	2.426s	53.14%
EDIC(Ours)	0.016s	53.35%

• EDIC 在与 Minnen 和 Lee 等最先进方法的比较中实现了具有竞争力的图像压缩性能，并超越了传统编解码器（JPEG、JPEG2000、BPG）。
• EDIC 的解码速度远快于自回归先验（在 768×512 图像上大约快 150× 于 Minnen 的方法）。
• 高斯混合熵建模相对于单一高斯可节省比特，视觉示例显示边缘区域的比特分配改善。
• 解码端增强通过在解码端学习高频细节来减少伪影，提升重建质量。
• 与视频压缩（DVC 基线）集成时，EDIC 对残差信息和运动信息编码的改进会传递，带来比某些基线更好的 RD 性能。
• 消融研究表明每个模块（通道注意力、GMM、解码端增强）相对于基线单一高斯模型都提供显著提升。