Google’s Novel Lossy Compression Method Targets Perfect Realism with Only a Single Diffusion Model

In the new paper Lossy Compression with Gaussian Diffusion, a Google Research team presents DiffC, a novel and simple lossy compression method that relies only on an unconditionally trained diffusion generative model and achieves state-of-the-art image compression results despite lacking an encoder transform.

ai artificial intelligence compression data compression deep-neural-networks diffusion diffusion model google machine learning machine learning & data science ml research technology