{"id":18086704,"url":"https://github.com/lubosmj/i2i-gans","last_synced_at":"2026-04-27T22:31:26.606Z","repository":{"id":110723720,"uuid":"342263301","full_name":"lubosmj/I2I-GANs","owner":"lubosmj","description":"GANs used for translating images of faces to preserve the privacy of individuals.","archived":false,"fork":false,"pushed_at":"2021-05-05T10:02:39.000Z","size":52,"stargazers_count":0,"open_issues_count":2,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-04-06T00:44:51.679Z","etag":null,"topics":["discogan","gcgan","image-to-image-translation","tensorflow","travelgan"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/lubosmj.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE.md","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2021-02-25T14:03:06.000Z","updated_at":"2022-04-29T04:13:21.000Z","dependencies_parsed_at":"2024-03-08T09:00:30.790Z","dependency_job_id":null,"html_url":"https://github.com/lubosmj/I2I-GANs","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/lubosmj/I2I-GANs","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/lubosmj%2FI2I-GANs","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/lubosmj%2FI2I-GANs/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/lubosmj%2FI2I-GANs/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/lubosmj%2FI2I-GANs/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/lubosmj","download_url":"https://codeload.github.com/lubosmj/I2I-GANs/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/lubosmj%2FI2I-GANs/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":32358509,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-27T20:07:02.737Z","status":"ssl_error","status_checked_at":"2026-04-27T20:07:00.910Z","response_time":128,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.5:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["discogan","gcgan","image-to-image-translation","tensorflow","travelgan"],"created_at":"2024-10-31T16:08:53.832Z","updated_at":"2026-04-27T22:31:26.591Z","avatar_url":"https://github.com/lubosmj.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# I2I-GANs\n\nCommon generative adversarial networks (GANs) implemented in TensorFlow 2.4.1. The GANs are suitable\nfor image-to-image translation tasks.\n\nThe repository was published as a part of the master's thesis (Generative Adversarial Networks Applied\nfor Privacy Preservation in Biometric-Based Authentication and Identification). Preliminary\nresults were presented at http://excel.fit.vutbr.cz/submissions/2021/031/31.pdf.\n\nThe following architectures are implemented:\n- TraVeLGAN (https://github.com/KrishnaswamyLab/travelgan)\n- DiscoGAN (https://github.com/SKTBrain/DiscoGAN)\n- GcGAN (https://github.com/hufu6371/GcGAN)\n\n### Setup\n1. Clone this repository:\n   ```\n   git clone https://github.com/lubosmj/I2I-GANs \u0026\u0026 cd I2I-GANs\n   ```\n2. Create a new virtual environment:\n   ```\n   python3 -m venv venv\n   source source venv/bin/activate\n   ```\n3. Install the packages:\n   ```\n   python3 setup.py install\n   ```\n4. Use the installed modules in your application:\n   ```python3\n   from i2i_gans import TraVeLGAN\n   \n   travelgan = TraVeLGAN(...)\n   travelgan.compile()\n   travelgan.load_weights(...)\n   \n   fake_images = travelgan.generator(...)\n   ```\n\n### Running the Examples\n1. Train a new TraVeLGAN model:\n   ```\n   python3 -m examples.travelgan_trainer train --domain_A \"path/to/dataset/A/*.png\" --domain_B \"path/to/dataset/B/*.png\" --dataset_size 5000 --batch_size=16 --checkpoints_freq 10 --parallel --samples_freq 10 --samples_dir samples --checkpoints_dir checkpoints --augment random_flip_left_right --epochs 250\n   ```\n2. Train a new DiscoGAN model:\n   ```\n   python3 -m examples.discogan_trainer train --domain_A \"path/to/dataset/A/*.png\" --domain_B \"path/to/dataset/B/*.png\" --dataset_size 5000 --batch_size=200 --checkpoints_freq 10 --parallel --samples_freq 10 --samples_dir samples --checkpoints_dir checkpoints --augment random_flip_left_right --epochs 200\n   ```\n3. Train a new GcGAN model:\n   ```\n   python3 -m examples.gcgan_trainer train --domain_A \"path/to/dataset/A/*.png\" --domain_B \"path/to/dataset/B/*.png\" --dataset_size 5000 --batch_size=12 --checkpoints_freq 10 --parallel --samples_freq 10 --samples_dir samples --checkpoints_dir checkpoints --augment random_flip_left_right --epochs 200\n   ```\n\n### Generated Images\n#### TraVeLGAN\nThe GAN was trained for 250 epochs with Adam optimizer (learning rate: 0.0002, batch size: 16, dataset size: 8,000).\n- Datasets:\n  - Augmented images from [CelebA](http://mmlab.ie.cuhk.edu.hk/projects/CelebA.html)\n  - [flowers102](https://www.robots.ox.ac.uk/~vgg/data/flowers/)\n\n\u003cimg src=\"https://user-images.githubusercontent.com/8740962/117123304-28234880-ad97-11eb-800e-35547f05d528.png\" width=\"60%\"\u003e\n\n#### DiscoGAN\nThe GAN was trained for 200 epochs with the same hyper-parameters as recommended in the original paper (dataset size: 20,000). Additionally, one convolution layer with 100 filters was inserted into the generators.\n- Datasets:\n  - [CelebA](http://mmlab.ie.cuhk.edu.hk/projects/CelebA.html)\n  - [UT Zappos50K](http://vision.cs.utexas.edu/projects/finegrained/utzap50k/)\n\n\u003cimg src=\"https://user-images.githubusercontent.com/8740962/117123568-80f2e100-ad97-11eb-8bd1-47f42b8a3c1b.png\" width=\"60%\"\u003e\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flubosmj%2Fi2i-gans","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Flubosmj%2Fi2i-gans","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flubosmj%2Fi2i-gans/lists"}