High Perceptual Quality Wireless Image Delivery with Denoising Diffusion Models

TL;DR

This work introduces a novel scheme, where the conventional DeepJSCC encoder targets transmitting a lower resolution version of the image, which later can be refined thanks to the generative model available at the receiver, and utilizes the range-null space decomposition of the target image.

Abstract

We consider the image transmission problem over a noisy wireless channel via deep learning-based joint source-channel coding (DeepJSCC) along with a denoising diffusion probabilistic model (DDPM) at the receiver. Specifically, we are interested in the perception-distortion trade-off in the practical finite block length regime, in which separate source and channel coding can be highly suboptimal. We introduce a novel scheme, where the conventional DeepJSCC encoder targets transmitting a lower resolution version of the image, which later can be refined thanks to the generative model available at the receiver. In particular, we utilize the range-null space decomposition of the target image; DeepJSCC transmits the range-space of the image, while DDPM progressively refines its null space contents. Through extensive experiments, we demonstrate significant improvements in distortion and perceptual quality of reconstructed images compared to standard DeepJSCC and the state-of-the-art generative learning-based method.

Figures (5)

• Figure 1: Overview of the image transmission procedure using our method.
• Figure 2: Overview of the image restoration procedure by $R_{\boldsymbol{\psi}}$.
• Figure 3: The encoder (top) and decoder (bottom) architectures of the employed DeepJSCC scheme.
• Figure 4: and comparison of our method with the baselines for $\rho \in \{0.0013, 0.0052\}$ over different .
• Figure 5: Qualitative comparison of the reconstructed images from CelebA-HQ dataset for $\rho=0.0013$ and $\mathrm{SNR}_{\mathrm{test}}=3$ dB.