Image-to-Image Translation
Image-to-Image Translation is a task in the field of computer vision and machine learning that aims to learn the mapping relationship between input images and output images to achieve specific goals, such as style transfer, data augmentation, or image restoration. By constructing complex models, this task can effectively transform the visual attributes of images, enhancing the diversity and flexibility of image processing, and has a wide range of application values.
SYNTHIA-to-Cityscapes
GTAV-to-Cityscapes Labels
ResNet101 65.1
Cityscapes Labels-to-Photo
pix2pix
ADE20K Labels-to-Photos
CoCosNet
COCO-Stuff Labels-to-Photos
PITI
ADE20K-Outdoor Labels-to-Photos
CoCosNet
IXI
ResViT
CelebA-HQ
StarGAN v2
Cityscapes-to-Foggy Cityscapes
MIC
cat2dog
U-GAT-IT
Cityscapes Photo-to-Labels
pix2pix
BCI
pyramidpix2pix
FLIR
horse2zebra
CycleGANAS
LLVIP
RaFD
StarGAN
selfie2anime
GNR
photo2vangogh
U-GAT-IT
vangogh2photo
U-GAT-IT
zebra2horse
CycleGAN
Aerial-to-Map
cGAN
AFHQ
StarGAN v2
anime-to-selfie
FQ-GAN
Deep-Fashion
CoCosNet
Object Transfiguration (sheep-to-giraffe)
InstaGAN
selfie-to-anime
FQ-GAN
SYNTHIA Fall-to-Winter
CyCADA
2017_test set
ADE-Indoor Labels-to-Photo
SB-GAN
AFHQ (Cat to Dog)
AFHQ (Wild to Dog)
EGSDE
Apples and Oranges
BRATS
dog2cat
KITTI Object Tracking Evaluation 2012
SRNet
photo2portrait
U-GAT-IT
portrait2photo
Zebra and Horses
Shared discriminator GAN