PO-ELIC: Perception-Oriented Efficient Learned Image Coding

In the past years, learned image compression (LIC) has achieved remarkable performance. The recent LIC methods outperform VVC in both PSNR and MS-SSIM. However, the low bit-rate reconstructions of LIC suffer from artifacts such as blurring, color drifting and texture missing. Moreover, those varied...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2022-05
Main Authors: He, Dailan, Yang, Ziming, Yu, Hongjiu, Xu, Tongda, Luo, Jixiang, Chen, Yuan, Gao, Chenjian, Shi, Xinjie, Qin, Hongwei, Wang, Yan
Format: Article
Language:English
Subjects:
Blurring
Image coding
Image compression
Image quality
Perception
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In the past years, learned image compression (LIC) has achieved remarkable performance. The recent LIC methods outperform VVC in both PSNR and MS-SSIM. However, the low bit-rate reconstructions of LIC suffer from artifacts such as blurring, color drifting and texture missing. Moreover, those varied artifacts make image quality metrics correlate badly with human perceptual quality. In this paper, we propose PO-ELIC, i.e., Perception-Oriented Efficient Learned Image Coding. To be specific, we adapt ELIC, one of the state-of-the-art LIC models, with adversarial training techniques. We apply a mixture of losses including hinge-form adversarial loss, Charbonnier loss, and style loss, to finetune the model towards better perceptual quality. Experimental results demonstrate that our method achieves comparable perceptual quality with HiFiC with much lower bitrate.
ISSN:2331-8422