Exploiting GAN Internal Capacity for High-Quality Reconstruction of Natural Images

Code for reproducing experiments in "Exploiting GAN Internal Capacity for High-Quality Reconstruction of Natural Images"

This directory contains associated source code to invert BigGAN generator for 128x128 resolution. Requires Tensorflow.

Generation of Random Samples:

Generate 1000 random samples of BigGAN generator:

  $> python random_sample.py random_sample.json

Inversion of the Generator:

The optimization is split into two steps according to the paper: First step, invesion to the latent space:

  $> python inversion.py params_latent.json

Second step, inversion to the dense layer:

  $> python inversion.py params_dense.json

Interpolation:

Generate interpolations between the inverted images and generated images:

  $> python interpolation.py params_dense.json

Segmentation:

Segment inverted images by clustering the attention map:

  $> python segmentation.py params_dense.json

Note: to replicate the experiments on real images from ImageNet, first a hdf5 file must be created with random images from the dataset, similar to the procedure in "random_sample.py". Then, the two step of optimization must be executed (modify the "dataset:" parameter in params_latent.json to consider custom datasets).