Image To Image Translation
Image-to-Image Translation 是计算机视觉与机器学习领域的一项任务,旨在学习输入图像与输出图像之间的映射关系,以实现特定目标,如风格迁移、数据增强或图像修复。该任务通过构建复杂的模型,能够有效转换图像的视觉属性,提升图像处理的多样性和灵活性,具有广泛的应用价值。
2017_test set
ADE-Indoor Labels-to-Photo
SB-GAN
ADE20K Labels-to-Photos
INADE
ADE20K-Outdoor Labels-to-Photos
DP-GAN
Aerial-to-Map
cGAN
AFHQ
StarGAN v2
AFHQ (Cat to Dog)
AFHQ (Wild to Dog)
EGSDE
anime-to-selfie
FQ-GAN
Apples and Oranges
BCI
pyramidpix2pix
BRATS
cat2dog
U-GAT-IT
CelebA-HQ
StarGAN v2
Cityscapes Labels-to-Photo
OASIS
Cityscapes Photo-to-Labels
pix2pix
Cityscapes-to-Foggy Cityscapes
MIC
COCO-Stuff Labels-to-Photos
PITI
Deep-Fashion
INADE
dog2cat
FLIR
GTAV-to-Cityscapes Labels
ResNet101 65.1
horse2zebra
CycleGANAS
IXI
ResViT
KITTI Object Tracking Evaluation 2012
SRNet
LLVIP
Object Transfiguration (sheep-to-giraffe)
InstaGAN
photo2portrait
U-GAT-IT
photo2vangogh
U-GAT-IT
portrait2photo
RaFD
StarGAN
selfie-to-anime
FQ-GAN
selfie2anime
GNR
SYNTHIA Fall-to-Winter
CyCADA
SYNTHIA-to-Cityscapes
vangogh2photo
U-GAT-IT
Zebra and Horses
Shared discriminator GAN
zebra2horse
CycleGAN