StyleGAN
E290872
StyleGAN is a state-of-the-art generative adversarial network architecture known for producing highly realistic, controllable images by manipulating disentangled style representations at different layers of the network.
All labels observed (7)
| Label | Occurrences |
|---|---|
| StyleGAN canonical | 2 |
| A Style-Based Generator Architecture for Generative Adversarial Networks | 1 |
| StyleGAN family | 1 |
| StyleGAN-ADA | 1 |
| StyleGAN-XL | 1 |
| StyleGAN2 | 1 |
| StyleGAN3 | 1 |
How this entity was disambiguated
This entity first appeared as the object of triple T2703881 — resolving that mention is where its identity was fixed. The disambiguator weighed these candidate entities and picked the highlighted one (or “None”, minting a new entity). This is how homonymy is resolved: the same surface form can point to different entities.
NED1
Entity disambiguation (via context triple)
gpt-5-mini-2025-08-07
Target entity: StyleGAN Context triple: [Generative Adversarial Networks, notableVariant, StyleGAN]
-
A.
PixelRNN
PixelRNN is a deep generative model that uses recurrent neural networks to sequentially model and generate images pixel by pixel.
-
B.
PixelCNN
PixelCNN is a deep generative model that uses convolutional neural networks with autoregressive masking to model and generate images pixel by pixel.
-
C.
Generative Adversarial Networks
Generative Adversarial Networks are a class of machine learning models in which two neural networks compete to generate highly realistic synthetic data, such as images, audio, or text.
-
D.
DALL·E
DALL·E is an AI model developed by OpenAI that generates images from natural language descriptions, enabling text-to-image synthesis.
-
E.
Automatic Adam
Automatic Adam is the nickname of Adam Vinatieri, a legendary NFL placekicker renowned for his clutch, game-winning field goals in high-pressure situations.
- F. None of above. chosen
- G. Unsure - the case is ambiguous/there is not enough information to decide.
NED2
Entity disambiguation (via description)
gpt-5-mini-2025-08-07
Target entity: StyleGAN Target entity description: StyleGAN is a state-of-the-art generative adversarial network architecture known for producing highly realistic, controllable images by manipulating disentangled style representations at different layers of the network.
-
A.
PixelRNN
PixelRNN is a deep generative model that uses recurrent neural networks to sequentially model and generate images pixel by pixel.
-
B.
PixelCNN
PixelCNN is a deep generative model that uses convolutional neural networks with autoregressive masking to model and generate images pixel by pixel.
-
C.
Generative Adversarial Networks
Generative Adversarial Networks are a class of machine learning models in which two neural networks compete to generate highly realistic synthetic data, such as images, audio, or text.
-
D.
DALL·E
DALL·E is an AI model developed by OpenAI that generates images from natural language descriptions, enabling text-to-image synthesis.
-
E.
Automatic Adam
Automatic Adam is the nickname of Adam Vinatieri, a legendary NFL placekicker renowned for his clutch, game-winning field goals in high-pressure situations.
- F. None of above. chosen
Statements (50)
| Predicate | Object |
|---|---|
| instanceOf |
deep learning model
ⓘ
generative adversarial network architecture ⓘ image synthesis model ⓘ |
| applicationDomain |
art generation
ⓘ
data augmentation ⓘ face generation ⓘ image synthesis ⓘ |
| basedOn |
GAN
ⓘ
Progressive GAN ⓘ
surface form:
Progressive Growing of GANs
|
| controlsCoarseFeatures | high-level styles ⓘ |
| controlsFeaturesAt | different layers of the network ⓘ |
| controlsFineFeatures | low-level styles ⓘ |
| developedBy |
NVIDIA Corporation
ⓘ
surface form:
NVIDIA
|
| discriminatorType | convolutional neural network ⓘ |
| firstPublicReleaseYear | 2019 ⓘ |
| firstPublishedYear | 2018 ⓘ |
| generatorType | style-based generator ⓘ |
| hasComponent |
mapping network
ⓘ
noise inputs ⓘ style blocks ⓘ synthesis network ⓘ |
| hasKeyConcept |
adaptive instance normalization
ⓘ
disentangled latent representations ⓘ multi-scale style control ⓘ noise injection ⓘ stochastic variation ⓘ style modulation ⓘ style-based generator ⓘ |
| hasOfficialRepository | https://github.com/NVlabs/stylegan ⓘ |
| hasSuccessor |
StyleGAN
self-linksurface differs
ⓘ
surface form:
StyleGAN2
StyleGAN self-linksurface differs ⓘ
surface form:
StyleGAN3
|
| implementedIn | TensorFlow ⓘ |
| influenced |
StyleGAN
self-linksurface differs
ⓘ
surface form:
StyleGAN-ADA
StyleGAN self-linksurface differs ⓘ
surface form:
StyleGAN-XL
many subsequent GAN architectures ⓘ |
| introducedBy |
Samuli Laine
ⓘ
Tero Karras ⓘ Timo Aila ⓘ |
| introducedInPaper |
StyleGAN
self-linksurface differs
ⓘ
surface form:
A Style-Based Generator Architecture for Generative Adversarial Networks
|
| license | NVIDIA Source Code License ⓘ |
| mapsLatentSpaceWith | mapping network ⓘ |
| notableFor |
disentangled style representations
ⓘ
fine-grained control over image attributes ⓘ highly realistic images ⓘ |
| publishedAtConference |
IEEE Computer Society Conference on Computer Vision and Pattern Recognition
ⓘ
surface form:
CVPR 2019
|
| trainingDataset |
CelebA
ⓘ
surface form:
CelebA-HQ
FFHQ ⓘ |
| usesLatentSpace |
W space
ⓘ
Z space ⓘ |
| usesOperation | AdaIN ⓘ |
How these facts were elicited
The pipeline generated the facts above by prompting gpt-5.1 with this entity's name + description and the instruction below.
Instruction
You are a knowledge base construction expert. Given a subject entity and a description of it, return factual statements that you know for the subject as a JSON list of dictionaries(triples), where keys must be "subject", "predicate" and "object". The number of facts may be very high, between 25 to 50 or more, for very popular subjects. For less popular subjects, the number of facts can be very low, like 5 or 10. # Requirements - If you don't know the subject at all, return an empty list. - If the subject is not a named entity, return an empty list. - Include at least one triple where predicate is "instanceOf". - Do not get too wordy. - Separate several objects into multiple triples with one object.
Input
Subject: StyleGAN Description of subject: StyleGAN is a state-of-the-art generative adversarial network architecture known for producing highly realistic, controllable images by manipulating disentangled style representations at different layers of the network.
Referenced by (8)
Full triples — surface form annotated when it differs from this entity's canonical label.
this entity surface form:
StyleGAN family
this entity surface form:
A Style-Based Generator Architecture for Generative Adversarial Networks
this entity surface form:
StyleGAN2
this entity surface form:
StyleGAN3
this entity surface form:
StyleGAN-ADA
this entity surface form:
StyleGAN-XL