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.

通过引入一种非二元鉴别器，该鉴别器是通过 VQ-VAE 自编码器获得的量化局部图像表示进行调整的，本文在 CLIC2020、DIV2K 和 Kodak 数据集上的评估表明，我们的鉴别器比最先进的 HiFiC 模型更有效地联合优化失真（例如，PSNR）和统计保真度（例如，FID）。在 CLIC2020 测试集上，我们获得与 HiFiC 相同的 FID，但比 HiFiC 少了 30-40％的比特。