A Patch-Based Algorithm For Diverse and High Fidelity Single Image Generation

The performance of variational auto-encoders (VAE) for image com- pression has steadily grown in recent years, thus becoming compet- itive with advanced visual data compression technologies. These neural networks transform the source image into a latent space with a channel-wise representation. in most works, the latents are scalar quantized before being entropy coded. On the other hand, vector quantizers generally achieve denser packings of high-dimensional data regardless of the source distribution. Hence, low-complexity variants of these quantizers are implemented in the compression standards JPEG 2000 and Versatile Video Coding. in this paper we demonstrate coding gains by using trellis-coded quantization (TCQ) over scalar quantization. For the optimization of the networks with regard to TCQ, we employ a specific noisy representation of the fea- tures during the training stage. For variable-rate VAEs, we obtained 7.7% average BD-rate savings on the Kodak images by using TCQ over scalar quantization. When different networks per target bitrate are optimized, we report a relative coding gain of 2.4% due to TCQ.