Code for Lightweight Deep Residue Learning for Joint Color Image Demosaicking and Denoising (ICPR2018 Oral)

train code

python main_ICPR.py --is_train 1 --train_dataset './data/WED/'

test code

python main_ICPR.py --is_train 0 --test_dataset './data/Kodak/'

avarage results for Kodak and McMaster dataset (PSNR and SSIM)

    | dataset | LCNN-DD       | LCNN-DD+      |
    | Kodak   | 42.55dB/99.57 | 45.66dB/99.82 |
    | McMaster| 39.25dB/99.01 | 42.03dB/99.68 |