subinsk/solution-challenge-classroom-x

CLASSROOM-X

"Low-Bandwidth AI Teaching Platform for overcoming internet access inequity"

Problem-Statement

Slow Internet Connections or Limited Access From Homes in Rural/Hilly/Remote Areas Can Contribute to Students Falling Behind Academically During Online-Era in The Prevailing COVID-Setups, According to a New Report From Michigan State University’s Quello Center.

Our-Hack

A Robust E-Classroom Platform With Lowest Data-Transmission Leveraging on AI for Overcoming the Internet Access Inequity Our Approach Was to Take Shortest Path Instead of Fastest Vehicle , So Instead of Direct Streaming We Regenerated Teacher's Screen on Student Console With Bare-Minimum Data-Streaming and Client-Side AI Monitoring Students Presence.

Presentation-Slides

Presentation-Video

Achievement

Able to Operate at 8 Kbps Speed With 28.8 Mb Hourly Data Consumption as Compared to Google Meet With Requirements of 90 Kbps Speed & 324 Mb/hour.

Data-Transmitted

• PDF -ANNOTATION
  • STATIC : PDF WHILE PAGE LOADS INITIALLY
  • DYNAMIC : PAGE_NO , COORDINATES_ANNOTATION
• WHITEBOARD
  • DYNAMIC : COORDINATES_ANNOTATION , CLEAR_FLAG
• AUDIO-CHANNEL
  • DYNAMIC : VOICE_BULB
• CHAT-ROOM:
  • DYNAMIC : TEXT
• SYSTEM:
  • DYNAMIC : ROOM_ID , MODE
• MONITOR-STUDENTS:
  • DYNAMIC : NONE
  • CLIENT-SIDE AI MODEL MONITORS STUDENTS THROUGH WEB-CAM AND ALERTS THE TEACHER ABOUT STUDENT'S PRESENCE.

Technologies Used

• Backend:
  • Python/Flask
  • Socket.io
• Frontend:
  • HTML/CSS/JS
  • JQuery/Bootstrap
  • Tensorflow.js
    • BlazeFace
• Database:
  • MongoDB
• Deployment:
  • Heroku

Configuration

Create a .env file in the folder.

Note: Make sure that .env file is at the same place where app.py exists.

Then, enter your MONGO_URI there:

MONGO_URI= **Your Mongo URi**

Installation

Install the Dependencies and devDependencies first:

pip install -r requirements.txt
flask run