paarthneekhara/text-to-image
A TensorFlow implementation that generates images from text captions using Generative Adversarial Networks with Skip Thought Vector encoding.
Not currently ranked — collecting fresh signals.
star history
This repository implements a text-to-image synthesis system using the GAN-CLS algorithm from the paper Generative Adversarial Text-to-Image Synthesis. Text captions are encoded into thought vectors using Skip Thought Vectors, and these representations guide a DCGAN-based generator to synthesize corresponding images. The model was trained on the flowers dataset.
Frequently asked
- What is paarthneekhara/text-to-image?
- A TensorFlow implementation that generates images from text captions using Generative Adversarial Networks with Skip Thought Vector encoding.
- Is text-to-image open source?
- Yes — paarthneekhara/text-to-image is open source, released under the MIT license.
- What language is text-to-image written in?
- paarthneekhara/text-to-image is primarily written in Python.
- How popular is text-to-image?
- paarthneekhara/text-to-image has 2.2k stars on GitHub.
- Where can I find text-to-image?
- paarthneekhara/text-to-image is on GitHub at https://github.com/paarthneekhara/text-to-image.