Lucas-Kanade Reloaded: End-to-End Super-Resolution from Raw Image Bursts

TL;DR

This paper introduces an end-to-end super-resolution method from raw image bursts, combining classical inverse problem techniques with learned parameters to improve alignment, noise handling, and image prior modeling, achieving state-of-the-art results.

Contribution

It presents a hybrid algorithm that learns parameters end-to-end for super-resolution from raw bursts, addressing alignment, noise, and prior modeling in a unified framework.

Findings

Sets new state-of-the-art on multiple benchmarks.

Produces high-quality super-resolved images from raw smartphone bursts.

Demonstrates effectiveness on real-world raw camera data.

Abstract

This presentation addresses the problem of reconstructing a high-resolution image from multiple lower-resolution snapshots captured from slightly different viewpoints in space and time. Key challenges for solving this problem include (i) aligning the input pictures with sub-pixel accuracy, (ii) handling raw (noisy) images for maximal faithfulness to native camera data, and (iii) designing/learning an image prior (regularizer) well suited to the task. We address these three challenges with a hybrid algorithm building on the insight from Wronski et al. that aliasing is an ally in this setting, with parameters that can be learned end to end, while retaining the interpretability of classical approaches to inverse problems. The effectiveness of our approach is demonstrated on synthetic and real image bursts, setting a new state of the art on several benchmarks and delivering excellent qualitative results on real raw bursts captured by smartphones and prosumer cameras.