Inverse Tone Mapping

Inverse tone mapping (ITM) aims to reconstruct high dynamic range (HDR) images from their compressed standard dynamic range (SDR) counterparts, recovering detail lost during capture. Current research focuses on improving the accuracy and efficiency of ITM, employing techniques like generative adversarial networks (GANs), look-up tables (LUTs) with AI-driven optimization, and lightweight neural networks such as improved residual networks. These advancements are crucial for enhancing the visual quality of existing SDR media on HDR displays and for enabling HDR content creation from readily available SDR sources, impacting both media production and consumer experiences.