Title: HDR-cGAN: Single LDR to HDR Image Translation using Conditional GAN

Abstract: The prime goal of digital imaging techniques is to reproduce the realistic appearance of a scene. Low Dynamic Range (LDR) cameras are incapable of representing the wide dynamic range of the real-world scene. The captured images turn out to be either too dark (underexposed) or too bright (overexposed). Specifically, saturation in overexposed regions makes the task of reconstructing a High Dynamic Range (HDR) image from single LDR image challenging. In this paper, we propose a deep learning based approach to recover details in the saturated areas while reconstructing the HDR image. We formulate this problem as an image-to-image (I2I) translation task. To this end, we present a novel conditional GAN (cGAN) based framework trained in an end-to-end fashion over the HDR-REAL and HDR-SYNTH datasets. Our framework uses an overexposed mask obtained from a pre-trained segmentation model to facilitate the hallucination task of adding details in the saturated regions. We demonstrate the effectiveness of the proposed method by performing an extensive quantitative and qualitative comparison with several state-of-the-art single-image HDR reconstruction techniques.