FlexHDR: Modelling Alignment and Exposure Uncertainties for Flexible HDR Imaging

TL;DR

FlexHDR introduces a novel HDR imaging method that models alignment and exposure uncertainties, using an HDR-aware attention mechanism and multi-stage fusion to produce higher quality images with fewer artifacts.

Contribution

The paper presents a new HDR technique that jointly models uncertainties and employs a flexible, permutation-invariant fusion strategy for improved image quality.

Findings

Up to 1.1dB PSNR improvement over state-of-the-art methods.

Better detail, colors, and fewer artifacts in HDR images.

Robust merging of multiple LDR images regardless of order.

Abstract

High dynamic range (HDR) imaging is of fundamental importance in modern digital photography pipelines and used to produce a high-quality photograph with well exposed regions despite varying illumination across the image. This is typically achieved by merging multiple low dynamic range (LDR) images taken at different exposures. However, over-exposed regions and misalignment errors due to poorly compensated motion result in artefacts such as ghosting. In this paper, we present a new HDR imaging technique that specifically models alignment and exposure uncertainties to produce high quality HDR results. We introduce a strategy that learns to jointly align and assess the alignment and exposure reliability using an HDR-aware, uncertainty-driven attention map that robustly merges the frames into a single high quality HDR image. Further, we introduce a progressive, multi-stage image fusion approach that can flexibly merge any number of LDR images in a permutation-invariant manner. Experimental results show our method can produce better quality HDR images with up to 1.1dB PSNR improvement to the state-of-the-art, and subjective improvements in terms of better detail, colours, and fewer artefacts.