Generative Models

Pix2Pix is a conditional image-to-image translation architecture that uses a conditional GAN objective combined with a reconstruction loss. The conditional GAN objective for observed images $x$, output images $y$ and the random noise vector $z$ is:

$$ \mathcal{L}_{cGAN}\left(G, D\right) =\mathbb{E}_{x,y}\left[\log D\left(x, y\right)\right]+ \mathbb{E}_{x,z}\left[log(1 − D\left(x, G\left(x, z\right)\right)\right] $$

We augment this with a reconstruction term:

$$ \mathcal{L}_{L1}\left(G\right) = \mathbb{E}_{x,y,z}\left[||y - G\left(x, z\right)||_{1}\right] $$

and we get the final objective as:

$$ G^{*} = \arg\min_{G}\max_{D}\mathcal{L}_{cGAN}\left(G, D\right) + \lambda\mathcal{L}_{L1}\left(G\right) $$

The architectures employed for the generator and discriminator closely follow DCGAN, with a few modifications:

  • Concatenated skip connections are used to "shuttle" low-level information between the input and output, similar to a U-Net.
  • The use of a PatchGAN discriminator that only penalizes structure at the scale of patches.
Source: Image-to-Image Translation with Conditional Adversarial Networks

Papers


Paper Code Results Date Stars

Tasks


Task Papers Share
Translation 39 18.40%
Image-to-Image Translation 33 15.57%
Image Generation 16 7.55%
Semantic Segmentation 8 3.77%
Style Transfer 6 2.83%
Colorization 6 2.83%
Anatomy 4 1.89%
Conditional Image Generation 3 1.42%
Image Classification 3 1.42%

Categories