GANs for Vision

Module: Computer Vision | Difficulty: Advanced

GAN Training Objective

The minimax game between generator and discriminator :

Wasserstein GAN (WGAN)

Replaces JS divergence with Wasserstein distance:

Spectral Normalization

Stabilizes training by constraining the Lipschitz constant:

FID Score

import torch
import torch.nn as nn

class Generator(nn.Module):
    def __init__(self, z_dim=128, channels=3):
        super().__init__()
        self.net = nn.Sequential(
            nn.ConvTranspose2d(z_dim, 512, 4, 1, 0),
            nn.BatchNorm2d(512), nn.ReLU(True),
            nn.ConvTranspose2d(512, 256, 4, 2, 1),
            nn.BatchNorm2d(256), nn.ReLU(True),
            nn.ConvTranspose2d(256, 128, 4, 2, 1),
            nn.BatchNorm2d(128), nn.ReLU(True),
            nn.ConvTranspose2d(128, channels, 4, 2, 1),
            nn.Tanh()
        )
    def forward(self, z):
        return self.net(z.view(-1, z.size(1), 1, 1))

class Discriminator(nn.Module):
    def __init__(self, channels=3):
        super().__init__()
        self.net = nn.Sequential(
            nn.Conv2d(channels, 128, 4, 2, 1),
            nn.LeakyReLU(0.2, True),
            nn.Conv2d(128, 256, 4, 2, 1),
            nn.BatchNorm2d(256), nn.LeakyReLU(0.2, True),
            nn.Conv2d(256, 512, 4, 2, 1),
            nn.BatchNorm2d(512), nn.LeakyReLU(0.2, True),
            nn.Conv2d(512, 1, 4, 1, 0),
        )
    def forward(self, x):
        return self.net(x).view(-1)

Key Takeaways

• GANs learn implicit data distributions for high-quality synthesis
• Spectral normalization and WGAN improve training stability
• FID is the standard metric for evaluating generated image quality