Edit model card
YAML Metadata Error: "datasets[0]" with value "https://www.robots.ox.ac.uk/~vgg/data/fgvc-aircraft/" is not valid. If possible, use a dataset id from https://hf.co/datasets.
Generated Real (for comparison)
image/png image/png

This GAN model is trained on the FGVC Aircraft dataset. The model uses Progressive Growing with Spectral Normalization.

The work builds up on https://huggingface.co/PrakhAI/AIPlane and https://huggingface.co/PrakhAI/AIPlane2.

This model was trained to generate 256x256 images of Aircrafts. The implementation in JAX on Colab can be found here.

Convolutional Architecture

A significant improvement over https://huggingface.co/PrakhAI/AIPlane2 is the elimination of "checkerboard" artifacts. This is done by using Image Resize followed by Convolution layer in the Generator instead of a Transposed Convolution where the kernel size is not divisible by the stride.

Transposed Convolution (kernel size not divisible by stride) Resize followed by convolution
image/png image/png

'Good' Generated Samples

image/png

ProGAN

Progressive Growing of GANs was proposed in Progressive Growing of GANs for improved Quality, Stability, and Variation

The idea is to start learning at lower resolutions, and growing the resolution of the GAN over time. This improves both:

  • Training Speed: At lower resolutions, the Generator and Discriminator have fewer layers and fewer parameters.
  • Convergence Speed: It is much easier to learn high-level details followed by higher granularity features, compared to learning both at the same time.

image/png

Spectral Normalization

Spectral Normalization for GANs was first suggested in Spectral Normalization for Generative Adversarial Networks.

Spectral Normalization constrains the Gradient Norm of the Discriminator with respect to the input, yielding a much smoother loss landscape for the Generator to navigate through.

image/gif

Latent Space Interpolation

Latent Space Interpolation can be an educational exercise to get deeper insight into the model.

It is observed below that several aspects of the generated image, such as the color of the sky, the grounded-ness of the plane, and the plane shape and color, are frequently continuous through the latent space.

image/png

Training Progression

Unfortunately after uploading, the first few seconds of the video are frozen. The full high resolution video is in model files.

Demo

The demo app for this model is at https://huggingface.co/spaces/PrakhAI/AIPlane3 (please "Restart this Space" if prompted).

Downloads last month
0
Unable to determine this model's library. Check the docs .

Space using PrakhAI/AIPlane3 1