RBSRICNN: Raw Burst Super-Resolution through Iterative Convolutional Neural Network

TL;DR

This paper introduces RBSRICNN, an iterative neural network that enhances low-quality burst images from mobile devices by modeling the physical degradation process, outperforming existing methods in super-resolution tasks.

Contribution

It presents a novel burst super-resolution framework that integrates physical modeling, classical regularization, and deep learning with iterative refinement, unlike prior black-box approaches.

Findings

Effective in recovering high-resolution details from burst images.

Robust generalization to real low-resolution burst inputs.

Outperforms existing super-resolution methods in experiments.

Abstract

Modern digital cameras and smartphones mostly rely on image signal processing (ISP) pipelines to produce realistic colored RGB images. However, compared to DSLR cameras, low-quality images are usually obtained in many portable mobile devices with compact camera sensors due to their physical limitations. The low-quality images have multiple degradations i.e., sub-pixel shift due to camera motion, mosaick patterns due to camera color filter array, low-resolution due to smaller camera sensors, and the rest information are corrupted by the noise. Such degradations limit the performance of current Single Image Super-resolution (SISR) methods in recovering high-resolution (HR) image details from a single low-resolution (LR) image. In this work, we propose a Raw Burst Super-Resolution Iterative Convolutional Neural Network (RBSRICNN) that follows the burst photography pipeline as a whole by a forward (physical) model. The proposed Burst SR scheme solves the problem with classical image regularization, convex optimization, and deep learning techniques, compared to existing black-box data-driven methods. The proposed network produces the final output by an iterative refinement of the intermediate SR estimates. We demonstrate the effectiveness of our proposed approach in quantitative and qualitative experiments that generalize robustly to real LR burst inputs with onl synthetic burst data available for training.