An open API service indexing awesome lists of open source software.

https://github.com/soumilgit/cc-mini_project

Takes footballer images as input, and outputs based on AWS Rekognition their names. Also, added a form at the end to take user feedback for adding other footballers to our custom dataset.
https://github.com/soumilgit/cc-mini_project

Last synced: 4 months ago
JSON representation

Takes footballer images as input, and outputs based on AWS Rekognition their names. Also, added a form at the end to take user feedback for adding other footballers to our custom dataset.

Awesome Lists containing this project

README

          

# AWS Serverless Architecture

![image](https://github.com/user-attachments/assets/443505c9-234d-4458-b8d4-d8a30500355e)

# Footballer Face Recognition – Serverless AI Project

End-to-end serverless web application that identifies professional football players using facial recognition powered by **Amazon Rekognition**. Built with a modern frontend and deployed on a secure, scalable AWS architecture using **Lambda, S3, API Gateway, CloudFront, and Route 53**.

## Features

- Real-time footballer face recognition
- Secure HTTPS hosting via **CloudFront + Route 53**
- Ultra-fast, low-latency responses using **AWS Lambda**
- 90% threshold matching via **Amazon Rekognition**
- Hosted static frontend using **Amazon S3**
- Formspree integration for dataset suggestions
- Mobile-responsive UI with **TailwindCSS**
- Automatic HTTP to HTTPS redirection

---

## Architecture

**Tech Stack:**

| Component | Technology |
|----------------|----------------------------|
| Frontend | HTML, CSS, JS, TailwindCSS |
| Hosting | Amazon S3 + CloudFront |
| Backend Logic | AWS Lambda (Python) |
| API Gateway | AWS API Gateway (REST) |
| Image Matching | Amazon Rekognition |
| Auth & Roles | IAM |
| Suggestions | Formspree.io |
| Domain + HTTPS | Route 53 + CloudFront |

---

## Flow Overview

1. User uploads a footballer image via web form.
2. Image is encoded in Base64 and sent to API Gateway.
3. Lambda function decodes it and calls Rekognition's `search_faces_by_image`.
4. Player name is matched and returned to the frontend.
5. Users can also suggest new players via a Formspree-powered form.