mingyuliutw/UNIT
A PyTorch implementation of UNsupervised Image-to-Image Translation (UNIT) networks using GANs for cross-domain image generation.
Not currently ranked — collecting fresh signals.
star history
The project implements the NIPS 2017 paper on unsupervised image-to-image translation networks. It uses a shared latent space assumption combining variational autoencoders with GANs to learn bidirectional image translation between visual domains without paired training data. The codebase supports multi-scale discriminators and includes pre-trained models for domain-specific translation tasks.
Frequently asked
- What is mingyuliutw/UNIT?
- A PyTorch implementation of UNsupervised Image-to-Image Translation (UNIT) networks using GANs for cross-domain image generation.
- Is UNIT open source?
- Yes — mingyuliutw/UNIT is an open-source project tracked on heatdrop.
- What language is UNIT written in?
- mingyuliutw/UNIT is primarily written in Python.
- How popular is UNIT?
- mingyuliutw/UNIT has 2k stars on GitHub.
- Where can I find UNIT?
- mingyuliutw/UNIT is on GitHub at https://github.com/mingyuliutw/UNIT.