Training-Free Rate-Distortion-Perception Traversal With Diffusion

TL;DR

This paper introduces a training-free method using pre-trained diffusion models to navigate the entire rate-distortion-perception tradeoff in lossy compression, combining theoretical optimality with empirical flexibility.

Contribution

It presents a novel, training-free framework that leverages diffusion models and a reverse channel coding module to adaptively traverse the RDP surface without retraining.

Findings

The diffusion decoder is proven optimal for the distortion-perception tradeoff under AWGN.

The framework achieves the optimal RDP function in the Gaussian case.

Empirical results show effective navigation of the RDP tradeoff across datasets.

Abstract

The rate-distortion-perception (RDP) tradeoff characterizes the fundamental limits of lossy compression by jointly considering bitrate, reconstruction fidelity, and perceptual quality. While recent neural compression methods have improved perceptual performance, they typically operate at fixed points on the RDP surface, requiring retraining to target different tradeoffs. In this work, we propose a training-free framework that leverages pre-trained diffusion models to traverse the entire RDP surface. Our approach integrates a reverse channel coding (RCC) module with a novel score-scaled probability flow ODE decoder. We theoretically prove that the proposed diffusion decoder is optimal for the distortion-perception tradeoff under AWGN observations and that the overall framework with the RCC module achieves the optimal RDP function in the Gaussian case. Empirical results across multiple datasets demonstrate the framework's flexibility and effectiveness in navigating the ternary RDP tradeoff using pre-trained diffusion models. Our results establish a practical and theoretically grounded approach to adaptive, perception-aware compression.