Perceptual Losses for Real-Time Style Transfer and Super-Resolution

27 Mar 2016Justin JohnsonAlexandre AlahiLi Fei-Fei

We consider image transformation problems, where an input image is transformed into an output image. Recent methods for such problems typically train feed-forward convolutional neural networks using a \emph{per-pixel} loss between the output and ground-truth images... (read more)

PDF Abstract

Code


Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT LEADERBOARD
Image Super-Resolution BSD100 - 4x upscaling Perceptual Loss PSNR 24.95 # 39
SSIM 0.6317 # 39
Nuclear Segmentation Cell17 FnsNet F1-score 0.7413 # 4
Dice 0.6165 # 5
Hausdorff 25.9102 # 5
Image Super-Resolution Set5 - 4x upscaling Perceptual Loss PSNR 27.09 # 45
SSIM 0.7680 # 39

Methods used in the Paper


METHOD TYPE
🤖 No Methods Found Help the community by adding them if they're not listed; e.g. Deep Residual Learning for Image Recognition uses ResNet