https://github.com/erankitcs/AWS_PhotoSimilarity

This project is created to identify and provide similar faces recommendation based on AWS Serverless and Neptune Graph DB.
https://github.com/erankitcs/AWS_PhotoSimilarity

api-gateway aws aws-lambda aws-serverless cloudformation-template graph-database html iam javascript neptune-cluster neptune-db nodejs rekognition s3-bucket serverless vpc vpc-endpoints

Last synced: 9 months ago
JSON representation

This project is created to identify and provide similar faces recommendation based on AWS Serverless and Neptune Graph DB.

• Host: GitHub
• URL: https://github.com/erankitcs/AWS_PhotoSimilarity
• Owner: erankitcs
• Created: 2020-06-14T06:14:50.000Z (over 5 years ago)
• Default Branch: master
• Last Pushed: 2023-01-07T19:19:46.000Z (about 3 years ago)
• Last Synced: 2025-03-13T21:44:07.868Z (10 months ago)
• Topics: api-gateway, aws, aws-lambda, aws-serverless, cloudformation-template, graph-database, html, iam, javascript, neptune-cluster, neptune-db, nodejs, rekognition, s3-bucket, serverless, vpc, vpc-endpoints
• Language: JavaScript
• Homepage:
• Size: 3 MB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 30
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# AWS_PhotoSimilarity

This project is created to identify and provide similar faces recommendation based on AWS Serverless and Neptune Graph DB.

## Features:

### Neptune Stack

  - VPC creation with two Subnet for high availability.

  - Network Security Group and Route Table to open HTTPs call to Neptune Cluster.

  - Internet Gateway and an Elastic IP.

  - Neptune Database Cluster and Neptune DB Instance.

  - VPC Endpoint for Neptune to access S3 data files.

  - Output Cluster Details and IAM roles to API Stack.

### Rekognition Stack

  - S3 bucket name creation using shortUuid and dotEnv modules.

  - Lambda Function with API Gateway to create an rekognition collection for image processing.

  - Lambda Function with API Gateway to delete an rekognition collection.

  - Lambda Function to process images (indexFaces and searchFaces utility of rekognition) and save Node and Edge   details into CSV format for Neptune DB.

  - APl Gateway to trigger Lambda function when a new image gets uploaded.

  - Lambda function with API Gateway to provide list of photos from S3 bucket.

  - Output S3 bucket name for API Stack.

### API Stack

  - Query Lambda Function to run parameterised Graph Queries using Gremlin against Neptune Instance.

  - API Gateway for Query Lambda Function to pass through Query Parameter String.

  - Loader Lambda Function to load CSV files created by Rekognition Stack Lambda Function into Neptune DB.

### UI

  - Create Front-End Application based on Node and React.

  - Call List Lambda function of rekognition Stack to load S3 Image files.

  - Call Query Lambda Function of API Stack to find similar photos list based on given threshold and hops.

## Setup:

  - Run npm install in each stack to load Node modules.

  - Serverless deploy in Neptune Stack first.

  - Serverless deploy in Rekognition Stack.

  - Serverless deploy in API Stack.

  - Manual Step: Attach created IAM role to Neptune DB from Front End.

  - Use Serverless and run Create collection Lambda function.

  - Upload Image to S3 bucket and make them public.

  - Run npm install inside UI folder.

  - Update App.js with Gateway base URLs and S3 bucket name.

  - Run npm start

  - Application is up and running.

## Architecture Diagram:

![Screenshot](ArchitectureDiagram.jpeg)

## Application:

![Screenshot](PhotoSimilarity.gif)