Deep Snapshot HDR Imaging Using Multi-Exposure Color Filter Array

TL;DR

This paper introduces a deep learning framework for HDR imaging from multi-exposure RAW data, utilizing luminance normalization to improve quality across luminance levels, outperforming existing snapshot HDR methods.

Contribution

It proposes a novel luminance normalization technique for HDR reconstruction from multi-exposure RAW data, enhancing visual quality and reducing artifacts.

Findings

Outperforms existing snapshot HDR methods in quality

Effectively handles errors in both bright and dark regions

Produces high-quality HDR images with fewer artifacts

Abstract

In this paper, we propose a deep snapshot high dynamic range (HDR) imaging framework that can effectively reconstruct an HDR image from the RAW data captured using a multi-exposure color filter array (ME-CFA), which consists of a mosaic pattern of RGB filters with different exposure levels. To effectively learn the HDR image reconstruction network, we introduce the idea of luminance normalization that simultaneously enables effective loss computation and input data normalization by considering relative local contrasts in the "normalized-by-luminance" HDR domain. This idea makes it possible to equally handle the errors in both bright and dark areas regardless of absolute luminance levels, which significantly improves the visual image quality in a tone-mapped domain. Experimental results using two public HDR image datasets demonstrate that our framework outperforms other snapshot methods and produces high-quality HDR images with fewer visual artifacts.