Adaptive Instance Normalization is a normalization method that aligns the mean and variance of the content features with those of the style features.
Instance Normalization normalizes the input to a single style specified by the affine parameters. Adaptive Instance Normaliation is an extension. In AdaIN, we receive a content input $x$ and a style input $y$, and we simply align the channel-wise mean and variance of $x$ to match those of $y$. Unlike Batch Normalization, Instance Normalization or Conditional Instance Normalization, AdaIN has no learnable affine parameters. Instead, it adaptively computes the affine parameters from the style input:
$$ \textrm{AdaIN}(x, y)= \sigma(y)\left(\frac{x-\mu(x)}{\sigma(x)}\right)+\mu(y) $$
Source: Arbitrary Style Transfer in Real-time with Adaptive Instance NormalizationPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Image Generation | 77 | 12.85% |
Disentanglement | 36 | 6.01% |
Style Transfer | 25 | 4.17% |
Image-to-Image Translation | 23 | 3.84% |
Image Manipulation | 20 | 3.34% |
Face Recognition | 18 | 3.01% |
Face Generation | 18 | 3.01% |
Decoder | 18 | 3.01% |
Translation | 16 | 2.67% |
Component | Type |
|
---|---|---|
🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |