{"id":19178714,"url":"https://github.com/meetps/tf-3dgan","last_synced_at":"2025-04-09T15:07:37.139Z","repository":{"id":78188863,"uuid":"82636131","full_name":"meetps/tf-3dgan","owner":"meetps","description":"Tensorflow implementation of 3D Generative Adversarial Network.","archived":false,"fork":false,"pushed_at":"2019-01-03T20:21:23.000Z","size":82,"stargazers_count":286,"open_issues_count":18,"forks_count":79,"subscribers_count":17,"default_branch":"master","last_synced_at":"2025-04-09T15:07:35.012Z","etag":null,"topics":["deep-learning","generative-adversarial-network","tensorflow"],"latest_commit_sha":null,"homepage":"https://meetshah.dev/gan/deep-learning/tensorflow/visdom/2017/04/01/3d-generative-adverserial-networks-for-volume-classification-and-generation.html","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/meetps.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","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":"2017-02-21T04:31:53.000Z","updated_at":"2025-02-10T11:43:20.000Z","dependencies_parsed_at":"2023-04-20T01:46:48.824Z","dependency_job_id":null,"html_url":"https://github.com/meetps/tf-3dgan","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/meetps%2Ftf-3dgan","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/meetps%2Ftf-3dgan/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/meetps%2Ftf-3dgan/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/meetps%2Ftf-3dgan/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/meetps","download_url":"https://codeload.github.com/meetps/tf-3dgan/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248055284,"owners_count":21040157,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","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":["deep-learning","generative-adversarial-network","tensorflow"],"created_at":"2024-11-09T10:40:38.979Z","updated_at":"2025-04-09T15:07:37.122Z","avatar_url":"https://github.com/meetps.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# tf-3dgan\n\n[![license](https://img.shields.io/github/license/mashape/apistatus.svg)](https://github.com/meetshah1995/tf-3dgan/blob/master/LICENSE)\n[![arXiv Tag](https://img.shields.io/badge/arXiv-1610.07584-brightgreen.svg)](https://arxiv.org/abs/1610.07584)\n\n## Tensorflow implementation of 3D Generative Adversarial Network.\n\nThis is a tensorflow implementation of the paper \"Learning a Probabilistic Latent Space of Object Shapes \nvia 3D Generative-Adversarial Modeling\"\n\n![](http://3dgan.csail.mit.edu/images/model.jpg)\n\n[Blog Post with interactive volume plots](https://meetshah1995.github.io/gan/deep-learning/tensorflow/visdom/2017/04/01/3d-generative-adverserial-networks-for-volume-classification-and-generation.html)\n\n### Requirements\n\n* tensorflow\u003e=1.0\n* visdom\u003e=1.0.1 (for mesh visualization)\n* scipy\n* scikit-image\n* stl (optional)\n\n\n#### One-line installation\n    \n`pip install scipy scikit-image stl visdom`\n\n### Data\n\n* Download the training data from the 3D Shapenet [website](http://3dshapenets.cs.princeton.edu/3DShapeNetsCode.zip)\n* Extract the zip and modify the path appropriately in `dataIO.py`\n\n### Usage\n\nLaunch [visdom](https://github.com/facebookresearch/visdom#launch) by running\n\n```\npython -m visdom.server\n```\n\nTo train the model (visdom will show generated chairs after every 200 minibatches)\n\n```\npython 3dgan_mit_biasfree.py 0 \u003cpath_to_model_checkpoint\u003e\n```\n\nTo generate chairs\n\n```\npython 3dgan_mit_biasfree.py 1 \u003cpath_to_trained_model\u003e\n```\n\nSome sample generated chairs\n\n|            |              |            |          |          |\n|------------|--------------|------------|----------|----------|\n|![](http://homepages.iitb.ac.in/~meetshah1995/files/2_.png) | ![](http://homepages.iitb.ac.in/~meetshah1995/files/4_.png) |  ![](http://homepages.iitb.ac.in/~meetshah1995/files/1_.png) |  ![](http://homepages.iitb.ac.in/~meetshah1995/files/3_.png) |  ![](http://homepages.iitb.ac.in/~meetshah1995/files/5_.png) |\n\n\n### Source code files\n\n| File      | Description                                                                   |\n|-----------|-------------------------------------------------------------------------------|\n|3dgan_mit_biasfree.py      | 3dgan as mentioned in the paper, with same hyperparams. \n|3dgan.py                   | baseline 3dgan with fully connected layer at end of discriminator.\n|3dgan_mit.py               | 3dgan as mentioned in the paper with bias in convolutional layers.\n|3dgan_autoencoder.py       | 3dgan with support for autoencoder based pre-training.\n|3dgan_feature_matching.py  | 3dgan with additional loss of feature mathcing of last layers. \n|dataIO.py                  | data input output and plotting utilities.\n|utils.py                   | tensorflow utils like leaky_relu and batch_norm layer.\n\n\n### Todo\n\n* Host the trained models\n* Add argparser based interface\n* Add threaded dataloader\n* Release the pytorch and keras versions of the GAN.\n* Train for longer number of epochs to improve quality of generated chairs.\n\n### Contributors\n\n* @meetshah1995\n* @khushhallchandra\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmeetps%2Ftf-3dgan","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmeetps%2Ftf-3dgan","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmeetps%2Ftf-3dgan/lists"}