0b01/SimGAN-Captcha
GAN-based approach to solve captchas using synthetic training data generated via adversarial refinement.
Not currently ranked — collecting fresh signals.
star history
This repository implements SimGAN to solve captchas without requiring manual label collection. It uses a generative adversarial network to refine synthetic captcha images so they resemble real ones, then trains a classifier on the refined synthetic data. The approach leverages Keras for the neural network implementation.
Frequently asked
- What is 0b01/SimGAN-Captcha?
- GAN-based approach to solve captchas using synthetic training data generated via adversarial refinement.
- Is SimGAN-Captcha open source?
- Yes — 0b01/SimGAN-Captcha is an open-source project tracked on heatdrop.
- What language is SimGAN-Captcha written in?
- 0b01/SimGAN-Captcha is primarily written in Jupyter Notebook.
- How popular is SimGAN-Captcha?
- 0b01/SimGAN-Captcha has 439 stars on GitHub.
- Where can I find SimGAN-Captcha?
- 0b01/SimGAN-Captcha is on GitHub at https://github.com/0b01/SimGAN-Captcha.