Statements (47)
Predicate | Object |
---|---|
gptkbp:instanceOf |
gptkb:convolutional_neural_network
|
gptkbp:architecture |
gptkb:GAN
|
gptkbp:author |
gptkb:Jaakko_Lehtinen
gptkb:Samuli_Laine gptkb:Tero_Karras gptkb:Timo_Aila Erik H"ark"onen Miika Aittala |
gptkbp:citation |
1000+
|
gptkbp:developedBy |
gptkb:NVIDIA
|
https://www.w3.org/2000/01/rdf-schema#label |
StyleGAN3
|
gptkbp:improves |
gptkb:StyleGAN2
|
gptkbp:input |
latent vector
|
gptkbp:language |
gptkb:Python
|
gptkbp:license |
gptkb:NVIDIA_Source_Code_License
|
gptkbp:mainActivity |
artificial intelligence research
image synthesis deepfake generation |
gptkbp:notableFeature |
open source
reduced memory usage improved training stability better temporal consistency continuous image synthesis improved alias-free image synthesis reduced artifacts rotation and translation equivariance |
gptkbp:notableFor |
data augmentation
style transfer domain adaptation art generation research in generative models generating animal faces generating high-quality human faces video synthesis |
gptkbp:notablePublication |
Alias-Free Generative Adversarial Networks
https://arxiv.org/abs/2106.12423 |
gptkbp:output |
synthetic image
|
gptkbp:platform |
gptkb:PyTorch
|
gptkbp:predecessor |
gptkb:StyleGAN2
|
gptkbp:releaseYear |
2021
|
gptkbp:repository |
https://github.com/NVlabs/stylegan3
|
gptkbp:successor |
StyleGAN-T
|
gptkbp:trainer |
gptkb:AFHQ
gptkb:FFHQ MetFaces |
gptkbp:bfsParent |
gptkb:StyleGAN
|
gptkbp:bfsLayer |
6
|