Rawformer: Unpaired Raw-to-Raw Translation for Learnable Camera ISPs

arXiv:2404.10700v1 Announce Type: cross
Abstract: Modern smartphone camera quality heavily relies on the image signal processor (ISP) to enhance captured raw images, utilizing carefully designed modules to produce final output images encoded in a standard color space (e.g., sRGB). Neural-based end-to-end learnable ISPs offer promising advancements, potentially replacing traditional ISPs with their ability to adapt without requiring extensive tuning for each new camera model, as is often the case for nearly every module in traditional ISPs. However, the key challenge …

abstract arxiv color cs.cv cs.lg eess.iv image images modern modules processor quality raw signal smartphone space standard translation type