PO-ELIC: Perception-Oriented Efficient Learned Image Coding

TL;DR

PO-ELIC enhances learned image compression by integrating adversarial training and perceptual losses, achieving high perceptual quality at lower bitrates compared to existing methods.

Contribution

It introduces a perception-oriented LIC model that combines adversarial training with multiple loss functions for improved perceptual quality.

Findings

Achieves comparable perceptual quality to HiFiC at lower bitrates.

Outperforms previous LIC methods in perceptual quality metrics.

Demonstrates effective adaptation of ELIC with adversarial techniques.

Abstract

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.