2018/01/31 by Eli Schwartz, Raja Giryes, Alex Bronstein +1 · 1 citation
Computer Science · Engineering · #Advanced Image Processing Techniques #Artificial intelligence #Color image #Computer science #Computer vision #Deep learning #Demosaicing #End-to-end principle #Image (mathematics) #Image Enhancement Techniques #Image and Signal Denoising Methods #Image processing #Image quality #JPEG #Noise reduction #Pipeline (software) #cs.CV #eess.IV
paper · pdf · doi:10.1109/tip.2018.2872858
published as IEEE Transactions on Image Processing 28.2 (2019): 912-923
openalex publication_date 2018/10/01 · arxiv created 2019/02/03 · arxiv updated 2019/02/05 · openalex created_date 2019/06/27 · openalex updated_date 2026/08/05
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 containing pairs of low-light 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.