DiffStereo: High-Frequency Aware Diffusion Model for Stereo Image Restoration

TL;DR

DiffStereo introduces a high-frequency aware diffusion model for stereo image restoration, effectively preserving textures and details while reducing computational costs, outperforming existing methods in multiple restoration tasks.

Contribution

This is the first application of diffusion models to stereo image restoration, leveraging high-frequency representations and a novel integration scheme for improved performance.

Findings

Achieves higher reconstruction accuracy in stereo super-resolution.

Provides better perceptual quality in stereo deblurring.

Excels in low-light stereo image enhancement.

Abstract

Diffusion models (DMs) have achieved promising performance in image restoration but haven't been explored for stereo images. The application of DM in stereo image restoration is confronted with a series of challenges. The need to reconstruct two images exacerbates DM's computational cost. Additionally, existing latent DMs usually focus on semantic information and remove high-frequency details as redundancy during latent compression, which is precisely what matters for image restoration. To address the above problems, we propose a high-frequency aware diffusion model, DiffStereo for stereo image restoration as the first attempt at DM in this domain. Specifically, DiffStereo first learns latent high-frequency representations (LHFR) of HQ images. DM is then trained in the learned space to estimate LHFR for stereo images, which are fused into a transformer-based stereo image restoration network providing beneficial high-frequency information of corresponding HQ images. The resolution of LHFR is kept the same as input images, which preserves the inherent texture from distortion. And the compression in channels alleviates the computational burden of DM. Furthermore, we devise a position encoding scheme when integrating the LHFR into the restoration network, enabling distinctive guidance in different depths of the restoration network. Comprehensive experiments verify that by combining generative DM and transformer, DiffStereo achieves both higher reconstruction accuracy and better perceptual quality on stereo super-resolution, deblurring, and low-light enhancement compared with state-of-the-art methods.