Weight Demodulation

Introduced by Karras et al. in Analyzing and Improving the Image Quality of StyleGAN

Weight Modulation is an alternative to adaptive instance normalization for use in generative adversarial networks, specifically it is introduced in StyleGAN2. The purpose of instance normalization is to remove the effect of $s$ - the scales of the features maps - from the statistics of the convolution’s output feature maps. Weight modulation tries to achieve this goal more directly. Assuming that input activations are i.i.d. random variables with unit standard deviation. After modulation and convolution, the output activations have standard deviation of:

$$ \sigma_{j} = \sqrt{{\sum_{i,k}w_{ijk}'}^{2}} $$

i.e., the outputs are scaled by the $L_{2}$ norm of the corresponding weights. The subsequent normalization aims to restore the outputs back to unit standard deviation. This can be achieved if we scale (“demodulate”) each output feature map $j$ by $1/\sigma_{j}$ . Alternatively, we can again bake this into the convolution weights:

$$ w''_{ijk} = w'_{ijk} / \sqrt{{\sum_{i, k}w'_{ijk}}^{2} + \epsilon} $$

where $\epsilon$ is a small constant to avoid numerical issues.

Source: Analyzing and Improving the Image Quality of StyleGAN

Latest Papers

PAPER DATE
Unsupervised Image-to-Image Translation via Pre-trained StyleGAN2 Network
Jialu HuangJing LiaoSam Kwong
2020-10-12
MaterialGAN: Reflectance Capture using a Generative SVBRDF Model
Yu GuoCameron SmithMiloš HašanKalyan SunkavalliShuang Zhao
2020-09-30
GIF: Generative Interpretable Faces
| Partha GhoshPravir Singh GuptaRoy UzielAnurag RanjanMichael BlackTimo Bolkart
2020-08-31
Improving the Performance of Fine-Grain Image Classifiers via Generative Data Augmentation
Shashank ManjunathAitzaz NathanielJeff DruceStan German
2020-08-12
High Resolution Zero-Shot Domain Adaptation of Synthetically Rendered Face Images
Stephan J. GarbinMarek KowalskiMatthew JohnsonJamie Shotton
2020-06-26
Differentiable Augmentation for Data-Efficient GAN Training
| Shengyu ZhaoZhijian LiuJi LinJun-Yan ZhuSong Han
2020-06-18
Training Generative Adversarial Networks with Limited Data
| Tero KarrasMiika AittalaJanne HellstenSamuli LaineJaakko LehtinenTimo Aila
2020-06-11
Using Generative Models for Pediatric wbMRI
Alex ChangVinith M. SuriyakumarAbhishek MoturuNipaporn TewattanaratAndrea DoriaAnna Goldenberg
2020-06-01
Network Bending: Manipulating The Inner Representations of Deep Generative Models
| Terence BroadFrederic Fol LeymarieMick Grierson
2020-05-25
Recognizing Families through Images with Pretrained Encoder
Tuan-Duy H. NguyenHuu-Nghia H. NguyenHieu Dao
2020-05-24
StyleGAN2 Distillation for Feed-forward Image Manipulation
| Yuri ViazovetskyiVladimir IvashkinEvgeny Kashin
2020-03-07
Analyzing and Improving the Image Quality of StyleGAN
| Tero KarrasSamuli LaineMiika AittalaJanne HellstenJaakko LehtinenTimo Aila
2019-12-03

Tasks

TASK PAPERS SHARE
Image Generation 4 26.67%
Decision Making 1 6.67%
Domain Adaptation 1 6.67%
Image Manipulation 1 6.67%
Image Morphing 1 6.67%
Style Transfer 1 6.67%
Colorization 1 6.67%
Image-to-Image Translation 1 6.67%
Semantic Similarity 1 6.67%

Components

COMPONENT TYPE
Convolution
Convolutions

Categories