Idempotence and Perceptual Image Compression
This addresses perceptual image compression for applications requiring high-quality reconstructions, offering a novel theoretical equivalence that simplifies implementation.
The paper tackles perceptual image compression by introducing a new paradigm that uses unconditional generative models with idempotence constraints, which is theoretically equivalent to conditional generative codecs and eliminates the need for training new models. Empirically, it outperforms state-of-the-art methods like HiFiC and ILLM in terms of Fréchet Inception Distance (FID).
Idempotence is the stability of image codec to re-compression. At the first glance, it is unrelated to perceptual image compression. However, we find that theoretically: 1) Conditional generative model-based perceptual codec satisfies idempotence; 2) Unconditional generative model with idempotence constraint is equivalent to conditional generative codec. Based on this newfound equivalence, we propose a new paradigm of perceptual image codec by inverting unconditional generative model with idempotence constraints. Our codec is theoretically equivalent to conditional generative codec, and it does not require training new models. Instead, it only requires a pre-trained mean-square-error codec and unconditional generative model. Empirically, we show that our proposed approach outperforms state-of-the-art methods such as HiFiC and ILLM, in terms of Fréchet Inception Distance (FID). The source code is provided in https://github.com/tongdaxu/Idempotence-and-Perceptual-Image-Compression.