Kernel-aware Burst Blind Super-Resolution

TL;DR

This paper introduces a kernel-aware burst super-resolution method that estimates degradation kernels and aligns raw images to improve high-resolution reconstruction from low-quality burst images, outperforming existing approaches.

Contribution

The paper proposes a novel kernel-guided strategy with a pyramid kernel-aware deformable alignment module for burst super-resolution, addressing unknown degradations in real-world images.

Findings

Achieves state-of-the-art performance on synthetic datasets.

Demonstrates effectiveness on real-world burst images.

Outperforms existing methods in HR image quality.

Abstract

Burst super-resolution (SR) technique provides a possibility of restoring rich details from low-quality images. However, since real world low-resolution (LR) images in practical applications have multiple complicated and unknown degradations, existing non-blind (e.g., bicubic) designed networks usually suffer severe performance drop in recovering high-resolution (HR) images. In this paper, we address the problem of reconstructing HR images from raw burst sequences acquired from a modern handheld device. The central idea is a kernel-guided strategy which can solve the burst SR problem with two steps: kernel estimation and HR image restoration. The former estimates burst kernels from raw inputs, while the latter predicts the super-resolved image based on the estimated kernels. Furthermore, we introduce a pyramid kernel-aware deformable alignment module which can effectively align the raw images with consideration of the blurry priors. Extensive experiments on synthetic and real-world datasets demonstrate that the proposed method can perform favorable state-of-the-art performance in the burst SR problem. Our codes are available at \url{https://github.com/shermanlian/KBNet}.