ArtHDR-Net: Perceptually Realistic and Accurate HDR Content Creation

TL;DR

ArtHDR-Net introduces a neural network architecture that enhances HDR content creation by focusing on perceptual quality, outperforming existing methods in visual perception metrics while maintaining structural accuracy.

Contribution

The paper presents ArtHDR-Net, a novel CNN-based architecture that emphasizes perceptual quality in HDR reconstruction, addressing a gap in preserving artistic intent.

Findings

Achieves state-of-the-art HDR-VDP-2 scores.

Maintains competitive PSNR and SSIM performance.

Focuses on perceptual quality over pixel-wise accuracy.

Abstract

High Dynamic Range (HDR) content creation has become an important topic for modern media and entertainment sectors, gaming and Augmented/Virtual Reality industries. Many methods have been proposed to recreate the HDR counterparts of input Low Dynamic Range (LDR) images/videos given a single exposure or multi-exposure LDRs. The state-of-the-art methods focus primarily on the preservation of the reconstruction's structural similarity and the pixel-wise accuracy. However, these conventional approaches do not emphasize preserving the artistic intent of the images in terms of human visual perception, which is an essential element in media, entertainment and gaming. In this paper, we attempt to study and fill this gap. We propose an architecture called ArtHDR-Net based on a Convolutional Neural Network that uses multi-exposed LDR features as input. Experimental results show that ArtHDR-Net can achieve state-of-the-art performance in terms of the HDR-VDP-2 score (i.e., mean opinion score index) while reaching competitive performance in terms of PSNR and SSIM.