RPS_Game :

A rock paper scissor game on web

Deployment

Deployed on Azure : https://gamerps.azurewebsites.net/

Progress :

• ML Model : Done
• Tensorflow js conversion : Done
• Basic Web Page : Done
• Tensorflow js integration : Done
• Classification on test images : Working (Done)
• Node JS setup and Routes : Done
• Deployment : Tested on Google Cloud Using a Virtual machine (Done)

Requirements :

1 . ML Model

• 1 . To develop a Machine Learning algorithm to classify and detect the rock paper and scissor hand gestures
• 2 . Model conversion so that we can use it for tensorflow js

2 . Web App

• 1 . Front -End : UI for the Web Game ( It should contain a DIV which can render live camera feed so that a frame can be craptured for playing the game)
• 2 . Back-End : for Routes and integrating the Tensorflow JS

3 . Deployment

• 1 . localhost:8000
• 2 . Virtual Machine ( whether in GCloud or Azure )
• 3 . Docker Image

Privacy Concern

As the user is required to upload the image . We respect the Privacy of the user . We will use the Images for training our ML Model

I am in the process of introducing the feature to delete your Data once you close the game

Tech Used

1 . Tensorflow , Tensorflow JS , Keras
2 . JavaScript , Python
3 . HTML , CSS
4 . NodeJs , Express , Express-fileupload

How to use ?

• Using LocalHost
  1 . clone this project
  2 . npm install
  3 . npm start
  4 . Just upload a photo by using the upload button and that's it !! You are good to go.
  
• Using Virtual Machine
  1 . Create a VM on Google cloud / Azure (os = linux)
  2 . Open the SSH terminal
  3 . Install Nodejs
  4 . clone this project and * cd RPC_Game *
  5 . npm start
  

Note you have to change the src for the image accordingly

Built With

• css
• html
• javascript
• jupyter-notebook
• tensorflow