Lossy image compression aims to represent images in as few bits as possible while maintaining fidelity to the original. Theoretical results indicate that optimizing distortion metrics such as PSNR or MS-SSIM necessarily leads to a discrepancy in the statistics of original images from those of reconstructions, in particular at low bitrates, often manifested by the blurring of the compressed images. Previous work has leveraged adversarial discriminators to improve statistical fidelity. Yet these binary discriminators adopted from generative modeling tasks may not be ideal for image compression. In this paper, we introduce a non-binary discriminator that is conditioned on quantized local image representations obtained via VQ-VAE autoencoders. Our evaluations on the CLIC2020, DIV2K and Kodak datasets show that our discriminator is more effective for jointly optimizing distortion (e.g., PSNR) and statistical fidelity (e.g., FID) than the state-of-the-art HiFiC model. On the CLIC2020 test set, we obtain the same FID as HiFiC with 30-40% fewer bits.

翻译：有损图像压缩旨在以尽可能少的比特数表示图像，同时保持与原图的保真度。理论结果表明，优化PSNR或MS-SSIM等失真度量必然导致重建图像与原始图像的统计特性存在差异，尤其是在低比特率下，这种差异常表现为压缩图像的模糊。先前的研究利用对抗性判别器来提升统计保真度，但这些源自生成建模任务的二元判别器对图像压缩而言可能并非理想选择。本文提出一种非二元判别器，该判别器基于通过VQ-VAE自编码器获得的量化局部图像表示进行条件化。我们在CLIC2020、DIV2K和Kodak数据集上的评估表明，与最先进的HiFiC模型相比，我们的判别器在联合优化失真（如PSNR）和统计保真度（如FID）方面更为有效。在CLIC2020测试集上，我们以比HiFiC少30-40%的比特数达到了相同的FID值。