Alert button

Lossy Compression with Gaussian Diffusion

Add code
Alert button
Jun 17, 2022
Figure 1 for Lossy Compression with Gaussian Diffusion
Figure 2 for Lossy Compression with Gaussian Diffusion
Figure 3 for Lossy Compression with Gaussian Diffusion
Figure 4 for Lossy Compression with Gaussian Diffusion

Share this with someone who'll enjoy it:

We describe a novel lossy compression approach called DiffC which is based on unconditional diffusion generative models. Unlike modern compression schemes which rely on transform coding and quantization to restrict the transmitted information, DiffC relies on the efficient communication of pixels corrupted by Gaussian noise. We implement a proof of concept and find that it works surprisingly well despite the lack of an encoder transform, outperforming the state-of-the-art generative compression method HiFiC on ImageNet 64x64. DiffC only uses a single model to encode and denoise corrupted pixels at arbitrary bitrates. The approach further provides support for progressive coding, that is, decoding from partial bit streams. We perform a rate-distortion analysis to gain a deeper understanding of its performance, providing analytical results for multivariate Gaussian data as well as initial results for general distributions. Furthermore, we show that a flow-based reconstruction achieves a 3 dB gain over ancestral sampling at high bitrates.

Share this with someone who'll enjoy it: