{"id":18386855,"url":"https://github.com/bchathoth-wt/face-classification","last_synced_at":"2026-04-30T14:35:35.803Z","repository":{"id":247801445,"uuid":"826885507","full_name":"bchathoth-wt/face-classification","owner":"bchathoth-wt","description":"A basic CNN to classify and verify the face images","archived":false,"fork":false,"pushed_at":"2024-07-10T15:39:23.000Z","size":10488,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-04-12T02:11:41.285Z","etag":null,"topics":["cmu","cnn","face-classification","image-classification","ml","pytorch"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/bchathoth-wt.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"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":"2024-07-10T15:16:54.000Z","updated_at":"2024-07-10T15:39:27.000Z","dependencies_parsed_at":"2024-07-10T18:55:16.788Z","dependency_job_id":null,"html_url":"https://github.com/bchathoth-wt/face-classification","commit_stats":null,"previous_names":["bchathoth-wt/face-classification"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bchathoth-wt%2Fface-classification","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bchathoth-wt%2Fface-classification/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bchathoth-wt%2Fface-classification/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bchathoth-wt%2Fface-classification/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/bchathoth-wt","download_url":"https://codeload.github.com/bchathoth-wt/face-classification/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248505928,"owners_count":21115354,"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":["cmu","cnn","face-classification","image-classification","ml","pytorch"],"created_at":"2024-11-06T01:23:37.918Z","updated_at":"2026-04-30T14:35:30.782Z","avatar_url":"https://github.com/bchathoth-wt.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Face Classification and Verification\nA basic CNN to classify and verify the facial images\n\n### Overview\nA sample project to help learn to build CNN-based architectures for face classification and\nface verification. The homework will instruct you on two key concepts:\n- How to build effective convolutional neural networks.\n- How to generate discriminative and generalizable feature representations for data.\n\n#### Face Classification\nFace classification is a closed-set multiclass classification problem where the subjects in the\ntest set has also been seen in the training set, although the precise pictures in the test set\nwill not be in the training set. For this to achieve high accuracy, it is only required that the\nembeddings for (all pictures of) the subjects in our “vocabulary” be linearly separable from\neach other.\n\n#### Face Verification\nFace verification refers to determining whether two face images are of the same\nperson without necessarily knowing who the person is. Face verification is an instance of a\nlarger class of problems where we attempt to determine if two data instances belong to the\nsame class without necessarily knowing (or having a model for) the class itself.\nThis is a common problem used in a variety of situations, for instance, when your laptop\nuses facial recognition to identify you. You would have “enrolled” yourself with an enrollment\nimage, and later when you try to login, your system compares a picture it takes of your face\nto the stored enrollment image to determine if both are from the same person. \n\n### Data Description\nWe will work with a small batch of data provided by Tom Mitchell from CMU and ponder over how convolution networks perform when benchmarked against a fully connected network. The full data set is provided in the zipped file below. Make sure to test your model with your own pictures (preferably in low-res). The data set can be downloaded \nfrom Tom Mitchell's website at CMU or the UCI ML repo:\n\nhttp://archive.ics.uci.edu/ml/datasets/cmu+face+images\nhttp://www.cs.cmu.edu/afs/cs.cmu.edu/user/mitchell/ftp/faces.html\n\nNote: In the faces dataset you will see variations of the same image, for instance:\n\nan2i_left_angry_open.pgm: indicate a full-resolution image (128 columns by 120 rows); \nan2i_left_angry_open_2.pgm: indicates a half-resolution image (64 by 60) and\nan2i_left_angry_open_4.pgm: indicates a quarter-resolution image (32 by 30).\nWhen designing your network, you will have to choose work with images of a fixed resolution.\n\nThis is a relatively small dataset for \"big data\" standards. However a well-designed convolution neural network, such as LeNet5 could make a dent on this problem. Bear in mind the risk of overfitting, and if your network has too many parameters, try to adjust it to accommodate the smaller dataset. \n\n### Problem Statement\nConsider two tasks:\n- Task 1: Is this image a picture of Mitchell? \n- Task 2: What's the facial expression in the picture?\n\nGoal: Build a CNN using LeNet-5 architecture and report how your network performs at tasks Task 1 and Task 2.\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbchathoth-wt%2Fface-classification","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fbchathoth-wt%2Fface-classification","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbchathoth-wt%2Fface-classification/lists"}