Usage of the Physarum Latent Walk Dataset to generate and interpolate images. The images are created with a latent walk, an interpolation between vectors in a latent space. The latent space is learned with an autoencoder neural network architecture.
More Examples in higher quality can be found here: https://piebro.github.io/physarum-latent-walk/index.html
An example of how to use the the dataset and generate a latent walk with the model can be found in example.ipynb. To open the notebook in google colab click here