kalyani-28/Full_Stack_Dev_AI

This is an AI project which involves deep learning, computer vision and rule-based inference engine. The objective is to create an AI which can understand the hand-drawn forms and creates full-fledged application with front-end and back-end.

4990 Project: Full_Stack_Dev_AI

This is an AI project which involves deep learning, computer vision and rule-based inference engine. The objective is to create an AI which can understand the hand-drawn forms and creates full-fledged application with front-end and back-end.

Run the app

Dependencies

• python 3.8+
• google vision api

Instructions

1. Copy google cloud key to project root with name gcp-auth.json
2. $ pip install -r requirements.txt
3. $ cd src && uvicorn app:app
4. Access app frontend on local port 8000

Resources

Forms

• MDN Form Guide
• Form Input Types

OpenCV Shape Detection

• Simple shape detection tutorial
• Finding contours
• Hu Moments shape matching
• Zernike Moments shape matching
• Detect Lines with Hough Line Transform

OCR

• Text processing and removal tutorial
• Handwritten Text Recognition
• SimpleHTR

Tensorflow

• Image Classification Tutorial

You Only Look Once Object Detection (YOLO)

• You Only Look Once: Unified, Real-Time Object Detection
• Yolov5 by Ultralytics
• Train Yolov5 Model
• Dataset

The First Release:

• Lyndon Renaud, Ian Straatman and Hong Qian Zhao have sucessfully completed the first part of the project.

The Second Release:

• Ben Sutherland, AJ Jraige and Jonathan Doyle have completed the second part of the project.