Dual-Camera Super-Resolution with Aligned Attention Modules

TL;DR

This paper introduces a dual-camera super-resolution method using aligned attention modules, leveraging reference images and domain adaptation to improve image quality in smartphone applications.

Contribution

It proposes a novel aligned attention-based approach for dual-camera super-resolution and a self-supervised domain adaptation strategy for real-world images.

Findings

Significant improvement over state-of-the-art in quantitative metrics

Effective domain adaptation for real-world images

High-quality results demonstrated on a new dataset and benchmarks

Abstract

We present a novel approach to reference-based super-resolution (RefSR) with the focus on dual-camera super-resolution (DCSR), which utilizes reference images for high-quality and high-fidelity results. Our proposed method generalizes the standard patch-based feature matching with spatial alignment operations. We further explore the dual-camera super-resolution that is one promising application of RefSR, and build a dataset that consists of 146 image pairs from the main and telephoto cameras in a smartphone. To bridge the domain gaps between real-world images and the training images, we propose a self-supervised domain adaptation strategy for real-world images. Extensive experiments on our dataset and a public benchmark demonstrate clear improvement achieved by our method over state of the art in both quantitative evaluation and visual comparisons.