Deep Fusion Generative Adversarial Networks for Text-to-Image Synthesis

A PyTorch implementation of Deep Fusion GAN by Ming Tao, Hao Tang, Songsong Wu, Nicu Sebe, Xiaoyuan Jing, Fei Wu, Bingkun Bao.

Dependencies

Dependencies:

python~=3.7.9
torch~=1.8.0
numpy~=1.21.4
pandas~=1.2.2
torchvision~=0.9.0
Pillow~=7.2.0
matplotlib~=3.3.4
tqdm~=4.62.3

To install required packages use:

pip install -r requirements.txt

Experiments

Use train_example.ipynb, metrics_evaluation.ipynb and eval_example.ipynb to train, eval and generation.

Deep Fusion GAN architecture

The architecture of the proposed DF-GAN for text-to-image synthesis. DF-GAN generates high-resolution images directly by one pair of generator and discriminator and fuses the text information and visual feature maps through multiple Deep text-image Fusion Blocks (DFBlock) in UPBlocks.