RazanBayoumi/An-Intelligent-Hybrid-Text-To-Image-Synthesis-Model-for-Generating-Realistic-Human-Faces

An-Intelligent-Hybrid-Text-To-Image-Synthesis-Model-for-Generating-Realistic-Human-Faces

This project page provides pytorch code that implements the following paper:

Title: "An Intelligent Hybrid Text-To-Image Synthesis Model for Generating Realistic Human Faces"

Link: https://ieeexplore.ieee.org/document/9694194

How to use

Python

• Python2.7
• Pytorch0.4 (conda install pytorch=0.4.1 cuda90 torchvision=0.2.1 -c pytorch)
• tensorflow (pip install --ignore-installed --upgrade https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-1.12.0-cp27-none-linux_x86_64.whl)
• pip install easydict pathlib
• conda install requests nltk pandas scikit-image pyyaml cudatoolkit=9.0

Data

• Download metadata for faces and extract them to data/faces/
  • python google_drive.py 1ao16xGvVmBltWadn21fIpxk2NGytEBLg ../data/faces.zip

Pretrained Models

• DAMSM for faces : Download and extract it to DAMSMencoders/
  • python google_drive.py 1ao16xGvVmBltWadn21fIpxk2NGytEBLg DAMSMencoders/faces.zip

Training

1- go into code/folder
2- python main.py --cfg cfg/faces_DMGAN.yml --gpu 0

Validation

1- Image Generation
	1- go into code/ folder
	2- python main.py --cfg cfg/eval_faces.yml --gpu 0