GlowGAN: Unsupervised Learning of HDR Images from LDR Images in the Wild

TL;DR

GlowGAN is an unsupervised generative model that learns to produce HDR images from in-the-wild LDR images by modeling the projection process, enabling realistic HDR synthesis and inverse tone mapping without paired training data.

Contribution

This work introduces the first unsupervised method for learning HDR image generation from LDR collections using GANs, incorporating a stochastic camera model for projection.

Findings

GlowGAN synthesizes photorealistic HDR images in challenging scenarios.

It enables unsupervised inverse tone mapping without HDR or paired images.

The method outperforms supervised models in reconstructing overexposed regions.

Abstract

Most in-the-wild images are stored in Low Dynamic Range (LDR) form, serving as a partial observation of the High Dynamic Range (HDR) visual world. Despite limited dynamic range, these LDR images are often captured with different exposures, implicitly containing information about the underlying HDR image distribution. Inspired by this intuition, in this work we present, to the best of our knowledge, the first method for learning a generative model of HDR images from in-the-wild LDR image collections in a fully unsupervised manner. The key idea is to train a generative adversarial network (GAN) to generate HDR images which, when projected to LDR under various exposures, are indistinguishable from real LDR images. The projection from HDR to LDR is achieved via a camera model that captures the stochasticity in exposure and camera response function. Experiments show that our method GlowGAN can synthesize photorealistic HDR images in many challenging cases such as landscapes, lightning, or windows, where previous supervised generative models produce overexposed images. We further demonstrate the new application of unsupervised inverse tone mapping (ITM) enabled by GlowGAN. Our ITM method does not need HDR images or paired multi-exposure images for training, yet it reconstructs more plausible information for overexposed regions than state-of-the-art supervised learning models trained on such data.