{ "info": { "author": "Ishwar Sawale", "author_email": "ishwar.code@gmail.com", "bugtrack_url": null, "classifiers": [ "Development Status :: 4 - Beta", "Intended Audience :: Developers", "License :: OSI Approved :: MIT License", "Natural Language :: English", "Programming Language :: Python :: 2", "Programming Language :: Python :: 2.6", "Programming Language :: Python :: 2.7", "Programming Language :: Python :: 3", "Programming Language :: Python :: 3.3", "Programming Language :: Python :: 3.4", "Programming Language :: Python :: 3.5", "Programming Language :: Python :: 3.6" ], "description": "Face Recognition\n================\n\nFace Recognition Based on Facenet\n\nBuilt using `Facenet `__'s\nstate-of-the-art face recognition built with deep learning. The model\nhas an accuracy of 99.2% on the `Labeled Faces in the\nWild `__ benchmark.\n\nFeatures\n--------\n\n- Out of Box Working Face Recognition\n- Choose Any Pre-Trained Model from Facenet\n- For training just provide the proper folder structure\n- Faster than other available solutions\n\nPrerequisites\n~~~~~~~~~~~~~\n\n- You need Python(2.6 to 3.5) installed\n- X-based System supported *(does work on Windows but not tested)*\n\nInstalling\n~~~~~~~~~~\n\n.. code:: python\n\n pip install facenet_recognition\n\nSetup\n^^^^^\n\n**Create setup as follows:**\n\n1. Create input directory eg: input\\_images\n2. Create aligned images directory eg: aligned\\_images *Create this\n directory we will store aligned images here*\n3. Create pre-trained model directory eg: pretrained\\_facenet\\_model\n *Download Pre-Trained model from\n `Facenet`* and keep it\n in the pre\\_model directory\n4. Create my trained classifier directory eg: my\\_classifier *In this\n directory we will save our trained model*\n\nLet's Begin\n-----------\n\nFor Facial Recognition we need to align images as follows:\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n\n.. code:: python\n\n import facenet_recognition\n facenet_recognition.align_input('input_images','aligned_images')\n\n*Above command will create our input images into aligned format and save\nit in given aligned images folder*\n\nTrain & Test Classifier on Images\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n\nAfter we have aligned images now we can train our classifier.\n\n.. code:: python\n\n pre_model='./pretrained_facenet_model/20170511-185253.pb' #locaiton of pret-trained model from Facenet\n my_class ='./my_classifier/my_classifier.pkl' #location where we want to save\n test_classifier_type = 'svm' #type of model either svm or nn\n weight= './my_classifier/model_small.yaml' #local stored weights\n\n facenet_recognition.test_train_classifier(aligned_images,pre_model,my_class,weight,test_classifier_type,nrof_train_images_per_class=30, seed=102)\n\n*Mininum Required Image per person*: *1* *Number of Images for Training\nper Person*: *30 (configurable)*\n\nTrain Classifer on Images(only Training)\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n\nThis API is used to Train our Classifier on Aligned Images\n\n.. code:: python\n\n pre_model='./pretrained_facenet_model/20170511-185253.pb' #locaiton of pret-trained model from Facenet\n my_class ='./my_classifier/my_classifier.pkl' #location where we want to save\n test_classifier_type = 'nn' #type of model either svm or nn\n weight= './my_classifier/model_small.yaml' #local stored weights\n\n facenet_recognition.create_classifier(aligned_images,pre_model,my_class,weight,test_classifier_type)\n\n*Mininum Required Image per person*: *1* *Number of Images for Training\nper Person*: *30 (fixed)*\n\nTest Classifer on Images\n~~~~~~~~~~~~~~~~~~~~~~~~\n\nThis API is used to test our Trained Classifer\n\n.. code:: python\n\n pre_model='./pretrained_facenet_model/20170511-185253.pb' #locaiton of pret-trained model from Facenet\n my_class ='./my_classifier/my_classifier.pkl' #location where we want to save\n test_classifier_type = 'nn' #type of model either svm or nn\n weight= './my_classifier/model_small.yaml' #local stored weights\n\n facenet_recognition.test_classifier(aligned_images,pre_model,my_class,weight,test_classifier_type)\n\n*Mininum Required Image per person*: *1*\n\nAuthors\n-------\n\n- **Ishwar Sawale** -- `Visit Portfolio `__\n\nLicense\n-------\n\nThis project is licensed under the MIT License - see the\n`LICENSE.md `__ file for details\n\nAcknowledgments\n---------------\n\n- Big Thanks to David Sandberg for Facent\n- Inspired by Dlib based library face\\_recognition\n\n\nHistory\n=======\n\n0.1.4 (2018-28-03)\n------------------\n\n* First beta release.\n", "description_content_type": "", "docs_url": null, "download_url": "", "downloads": { "last_day": -1, "last_month": -1, "last_week": -1 }, "home_page": "http://github.com/ishwarsawale/facenet_recognition", "keywords": "", "license": "MIT", "maintainer": "", "maintainer_email": "", "name": "facenet_recognition", "package_url": "https://pypi.org/project/facenet_recognition/", "platform": "", "project_url": "https://pypi.org/project/facenet_recognition/", "project_urls": { "Homepage": "http://github.com/ishwarsawale/facenet_recognition" }, "release_url": "https://pypi.org/project/facenet_recognition/0.1.4/", "requires_dist": null, "requires_python": "", "summary": "Face recognition based on Facenet", "version": "0.1.4" }, "last_serial": 3717224, "releases": { "0.1.4": [ { "comment_text": "", "digests": { "md5": "b7c1b2a51028a951c07483a1dbc6cd4a", "sha256": "756f8879990c33e07f6e78db258587e79bf7ee63097061691736b21ffd6633e4" }, "downloads": -1, "filename": "facenet_recognition-0.1.4.tar.gz", "has_sig": false, "md5_digest": "b7c1b2a51028a951c07483a1dbc6cd4a", "packagetype": "sdist", "python_version": "source", "requires_python": null, "size": 1884807, "upload_time": "2018-03-29T14:19:51", "url": "https://files.pythonhosted.org/packages/f4/ca/bc3517970eb7b2e1b5bad99603f1c0313d3a6bf1a0aeaf60edf7973ac821/facenet_recognition-0.1.4.tar.gz" } ] }, "urls": [ { "comment_text": "", "digests": { "md5": "b7c1b2a51028a951c07483a1dbc6cd4a", "sha256": "756f8879990c33e07f6e78db258587e79bf7ee63097061691736b21ffd6633e4" }, "downloads": -1, "filename": "facenet_recognition-0.1.4.tar.gz", "has_sig": false, "md5_digest": "b7c1b2a51028a951c07483a1dbc6cd4a", "packagetype": "sdist", "python_version": "source", "requires_python": null, "size": 1884807, "upload_time": "2018-03-29T14:19:51", "url": "https://files.pythonhosted.org/packages/f4/ca/bc3517970eb7b2e1b5bad99603f1c0313d3a6bf1a0aeaf60edf7973ac821/facenet_recognition-0.1.4.tar.gz" } ] }