aelnouby/Text-to-Image-Synthesis
A PyTorch implementation of GAN-based text-to-image synthesis that generates images from textual descriptions.
Not currently ranked — collecting fresh signals.
star history
This repository implements a conditional Generative Adversarial Network trained on text embeddings to generate images matching given descriptions. The architecture extends DCGAN with training stabilization techniques including feature matching, label smoothing, minibatch discrimination, and WGAN/WGAN-GP variants. The model is trained on Caltech-UCSD Birds 200 and Flowers datasets using pre-computed text embeddings.
Frequently asked
- What is aelnouby/Text-to-Image-Synthesis?
- A PyTorch implementation of GAN-based text-to-image synthesis that generates images from textual descriptions.
- Is Text-to-Image-Synthesis open source?
- Yes — aelnouby/Text-to-Image-Synthesis is open source, released under the GPL-3.0 license.
- What language is Text-to-Image-Synthesis written in?
- aelnouby/Text-to-Image-Synthesis is primarily written in Python.
- How popular is Text-to-Image-Synthesis?
- aelnouby/Text-to-Image-Synthesis has 411 stars on GitHub.
- Where can I find Text-to-Image-Synthesis?
- aelnouby/Text-to-Image-Synthesis is on GitHub at https://github.com/aelnouby/Text-to-Image-Synthesis.