论文标题
具有生成对抗网络的Fideltity-Controllity-Extreme图像压缩
Fidelity-Controllable Extreme Image Compression with Generative Adversarial Networks
论文作者
论文摘要
我们提出了一种基于GAN的图像压缩方法，在低于0.1BPP的极低比特率下起作用。在极低的比特率下，大多数现有的学习图像压缩方法都遭受模糊的损失。尽管Gan可以帮助重建尖锐的图像，但仍有两个缺点。首先，Gan使培训不稳定。其次，重建通常包含令人不快的噪音或人工制品。为了解决这两个缺点，我们的方法采用了两阶段培训和网络插值。两阶段的训练可有效稳定培训。此外，网络插值在各个阶段都使用模型，并减少了不良的噪声和伪像，同时保持重要边缘。因此，我们可以在不重新训练模型的情况下控制感知质量和忠诚度之间的权衡。实验结果表明，我们的模型可以重建高质量的图像。此外，我们的用户研究证实，我们的重建比最先进的基于GAN的图像压缩模型更可取。代码将可用。
We propose a GAN-based image compression method working at extremely low bitrates below 0.1bpp. Most existing learned image compression methods suffer from blur at extremely low bitrates. Although GAN can help to reconstruct sharp images, there are two drawbacks. First, GAN makes training unstable. Second, the reconstructions often contain unpleasing noise or artifacts. To address both of the drawbacks, our method adopts two-stage training and network interpolation. The two-stage training is effective to stabilize the training. Moreover, the network interpolation utilizes the models in both stages and reduces undesirable noise and artifacts, while maintaining important edges. Hence, we can control the trade-off between perceptual quality and fidelity without re-training models. The experimental results show that our model can reconstruct high quality images. Furthermore, our user study confirms that our reconstructions are preferable to state-of-the-art GAN-based image compression model. The code will be available.
