Neural scene representation and rendering (GQN)
Branch: master
Clone or download
Latest commit bf98baf Dec 23, 2018
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
experiments Improve code Dec 20, 2018
gqn Improve code Dec 20, 2018
.gitignore 🚀 Oct 23, 2018
LICENSE Create LICENSE Aug 5, 2018
README.md Update README Dec 23, 2018
hyperparams.py Improve code Dec 13, 2018
model.py Improve code Dec 13, 2018
optimizer.py Improve code Dec 13, 2018
scheduler.py Improve code Dec 13, 2018
train.py Fix a bug Dec 22, 2018
train_mn.py Improve code Dec 13, 2018

README.md

🚧 Work in Progress 🚧

Neural scene representation and rendering

https://deepmind.com/blog/neural-scene-representation-and-rendering/

Current training progress:

Shepard-Matzler 64x64

shepard_matzler

https://gfycat.com/ForthrightBrokenCanadagoose

shepard_matzler

https://gfycat.com/MajorSeriousKittiwake

shepard_matzler

https://gfycat.com/PartialYellowishFritillarybutterfly

shepard_matzler_predictions_6 shepard_matzler_predictions_7 shepard_matzler_predictions_9

Requirements

  • Python 3
  • Chainer 4+ pip3 install chainer

Network Architecture

gqn_conv_draw

gqn_representation

Dataset

deepmind/gqn-datasets

Datasets used to train GQN in the paper are available to download.

https://github.com/deepmind/gqn-datasets

You need to convert .tfrecord files to NumPy .npy format before starting training.

https://github.com/musyoku/gqn-datasets-translator

gqn-dataset-renderer

I am working on a ray tracer for rendering GQN dataset.

https://github.com/musyoku/gqn-dataset-renderer

  • Shepard-Matzler

shepard_matzler

  • Rooms

anim

  • MNIST Dice

mnist_dice