tf-exercise-gan

Tensorflow implementation of different GANs and their comparisions

GAN implementations

• [x] DCGAN from 'Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks' (https://arxiv.org/abs/1511.06434)
• [x] WGAN from 'Wasserstein GAN' (https://arxiv.org/abs/1701.07875)
• [x] BEGAN from 'BEGAN: Boundary Equilibrium Generative Adversarial Networks' (https://arxiv.org/abs/1703.10717)
• [x] MAD-GAN from 'Multi-Agent Diverse Generative Adversarial Networks' (https://arxiv.org/abs/1704.02906)
• [x] GoGAN from 'Gang of GANs: Generative Adversarial Networks with Maximum Margin Ranking' (https://arxiv.org/abs/1704.04865)
• [ ] ... (To be added)

Tasks

• [x] Impl. DCGAN, GoGAN, WGAN
• [x] Impl. BEGAN, MAD-GAN
• [x] Reproduce GANs on MNIST and CelebA datasets
• [x] Impl. training & evaluation on synthetic datasets
• [x] Add sample results
• [ ] Impl. inference-only code for GANs (may require refactoring)
• [ ] Impl. better evaluation function for real images (e.g. IvOM, energy dist., ...)
• [ ] Impl. a result logger
• [x] Compare GANs (synthetic)
• [x] Compare GANs (MNIST and CelebA dataset)
• [ ] Add quantitative comparisons
• [ ] Add more GAN implementations

Experiments & Benchmarks

170718 / Comparison of different GAN models on synthetic datasets

• Done without any hyper-parameter search.
• MAD-GAN worked best in the tested datasets.
• 

170718 / Sample results on MNIST dataset

170809 / Sample results on CelebA dataset

Other References