https://github.com/brain-to/vae-gan-celeba

Code and notebooks related to the paper: "Reconstructing Faces from fMRI Patterns using Deep Generative Neural Networks" by VanRullen & Reddy, 2019

https://github.com/brain-to/vae-gan-celeba

Science Score: 10.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • CITATION.cff file
  • codemeta.json file
  • .zenodo.json file
  • DOI references
  • Academic publication links
    Links to: arxiv.org
  • Academic email domains
  • Institutional organization owner
  • JOSS paper metadata
  • Scientific vocabulary similarity
    Low similarity (6.8%) to scientific vocabulary
Last synced: 10 months ago · JSON representation

Repository

Code and notebooks related to the paper: "Reconstructing Faces from fMRI Patterns using Deep Generative Neural Networks" by VanRullen & Reddy, 2019

Basic Info
  • Host: GitHub
  • Owner: BRAIN-TO
  • License: mit
  • Default Branch: master
  • Homepage:
  • Size: 30.3 KB
Statistics
  • Stars: 0
  • Watchers: 1
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Fork of rufinv/VAE-GAN-CelebA
Created over 5 years ago · Last pushed over 6 years ago

https://github.com/BRAIN-TO/VAE-GAN-CelebA/blob/master/

# VAE-GAN-CelebA
Python code related to the paper: ["Reconstructing Faces from fMRI Patterns using Deep Generative Neural Networks"](https://arxiv.org/abs/1810.03856) by VanRullen & Reddy (2019)

### This folder contains:
* a [link to a pre-trained VAE-GAN model checkpoint 'vaegan_celeba.ckpt'](http://depotcerco.univ-tlse3.fr/vmcca78o) (~15 epochs on CelebA dataset=50,000 batches of 64 images)
* a set of .py functions for the VAE-GAN face decomposition/reconstruction model, in particular:
  * VAEGAN_image2latent.py => goes from any image file to the corresponding 1024D latent encoding (saved as a Matlab .mat file)
  * VAEGAN_latent2image.py => goes from a 1024D latent encoding (Matlab .mat file) to the corresponding image(s)
* (optional) [a link to download the fMRI datasets](https://openneuro.org/datasets/ds001761) (4 subjects, each saw > 8,000 faces in the scanner) and some Matlab analysis code

### Example usage:
    VAEGAN_image2latent.py -i example.jpg     #this will create example_z.mat with the 1024 latent vars
    VAEGAN_latent2image.py -i example_z.mat   #this will generate example_z_g.jpg (and also example_z_g.mat)

### Requirements:
* Python >= 3.4
* Tensorflow >= 1.8
* matplotlib, numpy, scipy, skimage

Owner

  • Name: Brain Research in Advanced Imaging and Neuromodeling - Toronto
  • Login: BRAIN-TO
  • Kind: organization

GitHub Events

Total
Last Year