Deep HDR Hallucination for Inverse Tone Mapping

TL;DR

This paper introduces a GAN-based approach for inverse tone mapping that effectively hallucinate missing HDR information from LDR images, outperforming existing methods in dynamic range expansion and plausibility.

Contribution

It presents a novel GAN-based method for HDR hallucination in inverse tone mapping, incorporating a density-based normalization and HDR data augmentation techniques.

Findings

Quantitatively competitive with state-of-the-art methods

Provides good dynamic range expansion in well-exposed areas

Generates plausible hallucinations for saturated and under-exposed regions

Abstract

Inverse Tone Mapping (ITM) methods attempt to reconstruct High Dynamic Range (HDR) information from Low Dynamic Range (LDR) image content. The dynamic range of well-exposed areas must be expanded and any missing information due to over/under-exposure must be recovered (hallucinated). The majority of methods focus on the former and are relatively successful, while most attempts on the latter are not of sufficient quality, even ones based on Convolutional Neural Networks (CNNs). A major factor for the reduced inpainting quality in some works is the choice of loss function. Work based on Generative Adversarial Networks (GANs) shows promising results for image synthesis and LDR inpainting, suggesting that GAN losses can improve inverse tone mapping results. This work presents a GAN-based method that hallucinates missing information from badly exposed areas in LDR images and compares its efficacy with alternative variations. The proposed method is quantitatively competitive with state-of-the-art inverse tone mapping methods, providing good dynamic range expansion for well-exposed areas and plausible hallucinations for saturated and under-exposed areas. A density-based normalisation method, targeted for HDR content, is also proposed, as well as an HDR data augmentation method targeted for HDR hallucination.