DeepISP: Learning End-to-End Image Processing Pipeline

We present DeepISP, a full end-to-end deep neural model of the camera image signal processing (ISP) pipeline. Our model learns a mapping from the raw low-light mosaiced image to the final visually compelling image and encompasses low-level tasks such as demosaicing and denoising as well as higher-level tasks such as color correction and image adjustment. The training and evaluation of the pipeline were performed on a dedicated dataset, the S7-ISP dataset1, containing pairs of lowlight and well-lit images captured by a Samsung S7 smartphone camera in both raw and processed JPEG formats. The proposed solution achieves state-of-the-art performance in objective evaluation of PSNR on the subtask of joint denoising and demosaicing. For the full end-to-end pipeline, it achieves better visual quality compared to the manufacturer ISP, in both a subjective human assessment and when rated by a deep model trained for assessing image quality. Well-lit (ground truth) Low-light raw input Samsung S7 DeepISP Figure 1: End-to-end low-light image processing. From left to right: a ground truth well-lit image, raw input low-light image, output of the Samsung S7 ISP, and of the proposed DeepISP. Dataset and full paper are available on the project page.