{"id":25478326,"url":"https://github.com/superrmurlocc/face-detector","last_synced_at":"2026-05-04T12:36:17.276Z","repository":{"id":277909583,"uuid":"933888778","full_name":"SuperrMurlocc/Face-Detector","owner":"SuperrMurlocc","description":"Face-Detector is a deep learning-based system designed for identifying and recognizing faces in images and video streams. Built upon the VGGFace2 model and fine-tuned with ArcFace and Triplet Loss, it achieves 97% accuracy on a subset of the CelebA dataset.","archived":false,"fork":false,"pushed_at":"2025-02-16T23:18:06.000Z","size":0,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-02-16T23:27:21.720Z","etag":null,"topics":["computer-vision","deep-learning","face-recognition","opencv","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":"gpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/SuperrMurlocc.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":"2025-02-16T22:46:46.000Z","updated_at":"2025-02-16T23:18:35.000Z","dependencies_parsed_at":"2025-02-16T23:37:25.287Z","dependency_job_id":null,"html_url":"https://github.com/SuperrMurlocc/Face-Detector","commit_stats":null,"previous_names":["superrmurlocc/face-detector"],"tags_count":1,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SuperrMurlocc%2FFace-Detector","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SuperrMurlocc%2FFace-Detector/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SuperrMurlocc%2FFace-Detector/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SuperrMurlocc%2FFace-Detector/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/SuperrMurlocc","download_url":"https://codeload.github.com/SuperrMurlocc/Face-Detector/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":239488748,"owners_count":19647260,"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":["computer-vision","deep-learning","face-recognition","opencv","pytorch"],"created_at":"2025-02-18T14:31:27.082Z","updated_at":"2025-11-06T09:30:29.219Z","avatar_url":"https://github.com/SuperrMurlocc.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Face-Detector\n\n![output](https://github.com/user-attachments/assets/2dae3f03-bd0a-43fe-8f52-08706b9f6b70)\n\nThis project presents a **Face Detector** designed for **detecting faces in images and associating them with previously seen identities**. The system is built upon a **VGGFace2 model**, fine-tuned using **ArcFace** and **Triplet** Losses to enhance recognition performance.  \n\nThe implementation provides:  \n- A **face bank** for storing and managing known identities.  \n- An **application** capable of detecting faces from images or a live webcam feed.  \n\nThe project leverages a **deep learning-based algorithm** implemented with the **PyTorch** framework, achieving **97% classification accuracy** on a subset of the CelebA dataset.\n\n---\n\n## Table of Contents  \n- [Overview](#overview)  \n- [Datasets](#datasets)\n- [System Workflow](#system-workflow)  \n- [Technologies Used](#technologies-used)  \n- [Results](#results)  \n- [Contributors](#contributors)\n  \n---\n\n## Overview  \nThe goal of this project was to create a system capable of accurately detecting and recognizing faces, The main functionalities include:  \n- Identifing and locating faces in images or video streams.\n- Aligning faces for the neural network.\n- Match detected faces against a database of known identities.\n- Managing known faces for recognition.\n- Performing detection and recognition on live webcam input. \n  \n---\n\n## Datasets  \n\n- **Training:** The model was trained using the **CelebA** dataset.  \n- **Testing \u0026 Evaluation:** A subset of CelebA was used for testing and performance metrics.  \n- **Presentation:** A custom **Players Dataset** was created for demonstration purposes, containing a few images of **Kylian Mbappé** and **Cristiano Ronaldo**.  \n\n⚠ **Disclaimer:** I do not claim ownership of the images used in the *Players Dataset*. They are included solely for demonstration purposes.  \n\n---\n\n## System Workflow  \n![DLF drawio](https://github.com/user-attachments/assets/7b474345-b120-4ddd-b65f-081ecf90aa1f)\n\n---\n\n## Technologies Used  \n- **Python 3**\n- **PyTorch** - for building and training deep learning models\n- **OpenCV** – for image processing and feature detection  \n- **NumPy** – for efficient numerical operations  \n- **Matplotlib** – for visualization  \n- **einops** – for tensor manipulation  \n- **Git/GitHub** – for version control and collaboration\n\n---\n\n## Results  \n\nThe face recognition system achieved **97% classification accuracy** on a subset of the **CelebA dataset**, demonstrating high reliability in identifying known faces.  \n\nTo further evaluate performance, we analyzed **False Rejection Rate (FRR)** and **False Acceptance Rate (FAR)** across different decision thresholds:  \n\n- **FRR (False Rejection Rate)** – Measures the percentage of genuine faces incorrectly rejected by the system.  \n- **FAR (False Acceptance Rate)** – Measures the percentage of impostor faces incorrectly accepted as known identities.  \n\nThe following figure presents the **FRR and FAR curves**, showing the trade-off between security and recognition accuracy:  \n\n![output](https://github.com/user-attachments/assets/1e307a47-2218-41ae-ac92-967ccf52a77d)\n\nA well-balanced threshold ensures both **low rejection of genuine users** and **high resistance to false acceptances**.  \n\n---\n\n## Contributors  \n- **Jakub Bednarski** – Conceptualization, Methodology, Software Development, Project Administration  \n- **Julia Komorowska** – Software Development, Investigation\n- **Adam Wasiela** – Software Development, Investigation  \n- **Hubert Woziński** – Software Development, Investigation  \n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsuperrmurlocc%2Fface-detector","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsuperrmurlocc%2Fface-detector","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsuperrmurlocc%2Fface-detector/lists"}