zsyzzsoft/co-mod-gan
A Generative Adversarial Network for filling in large missing regions in images.
Not currently ranked — collecting fresh signals.
star history
This is an implementation of Co-Modulated Generative Adversarial Networks (CoModGAN), a novel GAN architecture for image completion that co-modulates both conditional and stochastic style representations. The model bridges image-conditional and modulated unconditional generative architectures to handle large-scale missing regions that prior methods struggle with. It also introduces Paired/Unpaired Inception Discriminative Scores (P-IDS/U-IDS) as quantitative metrics for measuring perceptual fidelity of inpainted images.
Frequently asked
- What is zsyzzsoft/co-mod-gan?
- A Generative Adversarial Network for filling in large missing regions in images.
- Is co-mod-gan open source?
- Yes — zsyzzsoft/co-mod-gan is an open-source project tracked on heatdrop.
- What language is co-mod-gan written in?
- zsyzzsoft/co-mod-gan is primarily written in Python.
- How popular is co-mod-gan?
- zsyzzsoft/co-mod-gan has 472 stars on GitHub.
- Where can I find co-mod-gan?
- zsyzzsoft/co-mod-gan is on GitHub at https://github.com/zsyzzsoft/co-mod-gan.