Deep Chain HDRI: Reconstructing a High Dynamic Range Image from a Single Low Dynamic Range Image

TL;DR

This paper introduces a deep neural network that reconstructs high dynamic range images from a single low dynamic range image by inferring multiple exposures and merging them, improving HDR quality metrics significantly.

Contribution

The novel chaining neural network model effectively maps LDR to HDR images, enhancing HDR reconstruction from a single image with superior quality metrics.

Findings

HDR-VDP2 Q score improved by 6 points over conventional methods

Peak signal-to-noise ratio for tone-mapped images increased by 10 dB

Model successfully infers multiple exposures from a single LDR image

Abstract

In this paper, we propose a novel deep neural network model that reconstructs a high dynamic range (HDR) image from a single low dynamic range (LDR) image. The proposed model is based on a convolutional neural network composed of dilated convolutional layers, and infers LDR images with various exposures and illumination from a single LDR image of the same scene. Then, the final HDR image can be formed by merging these inference results. It is relatively easy for the proposed method to find the mapping between the LDR and an HDR with a different bit depth because of the chaining structure inferring the relationship between the LDR images with brighter (or darker) exposures from a given LDR image. The method not only extends the range, but also has the advantage of restoring the light information of the actual physical world. For the HDR images obtained by the proposed method, the HDR-VDP2 Q score, which is the most popular evaluation metric for HDR images, was 56.36 for a display with a 1920$\times$1200 resolution, which is an improvement of 6 compared with the scores of conventional algorithms. In addition, when comparing the peak signal-to-noise ratio values for tone mapped HDR images generated by the proposed and conventional algorithms, the average value obtained by the proposed algorithm is 30.86 dB, which is 10 dB higher than those obtained by the conventional algorithms.