Skip to content

A project on an ARKit based IOS app using text detection and manipulation for people with reading disabilities.

License

Notifications You must be signed in to change notification settings

ShivendraAgrawal/EduAR

Repository files navigation

EduAR

The main focus of this project is to aid individuals with learning disabilities.

CSCI-5413 VR Project

Hardware Requirement

  1. A Macbook
  2. Xcode installed
  3. An iOS device beyond iphone 6

Installing the app

  1. Install python 3.6

  2. Install the required libraries -

    Flask==0.12.2

    Flask-Cors==3.0.3

    numpy==1.14.2

    opencv-python==3.4.1.15

    Pillow==5.1.0

    pytesseract==0.2.5

  3. Start the Python server by running the app

  4. Find you public IP address by using ifconfig (on Unix based systems)

  5. Change the IP address on Line 36 of ViewController.swift

  6. Open the Xcode project (EduAR.xcodeproj)

  7. Set the team ID as shivendra.agrawal@gmail.com or create a new one.

  8. Provide a unique budle identifier.

  9. Connect your ipad or iphone and build the project

Preparing ARKit

  • First you have to spatially map the surrounding, the yellow dots are feature points which show that the environment is being perceived and mapped.

img1

Parsing and Visualizing the Text in a Scene

  • Then, EduAR reads from the paper with the LSTM mode of the Tesseract OCR. We visalise this with 3D text on a floating plane in AR anchored on the real text.
  • The 3D anchoring location is obtained by the performing a hit test with the obtained 2D location of the text and the feature points identified by ARKit.

img2

Visual Aide Text Editor

  • This will help learners to generate visually anchored 3D text for all the texts in a scene.
    Then they can tap the desired text and enter the visual aide of EduAR.
  • Text Editor Features

    • The text editor has all the standard copy, share, define options as well as the ability to modify the parsed text.
    • Unique read mode - This allows the users to just focus on one paragraph, sentence or word at a time and swipe to the next segment.
    • The font size option is also provided to change the font size in real-time as per convinience.

img3

3D Objects

  • To encourage visual learning we have enabled a double tap to visualize feature.
  • When the user double taps anywhere on the screen it spawns a 3D object on the plane with hit test (with planes this time).
  • The 3D object is one that should be present in our small database of objects and also mentioned in the text we scanned.
  • The objects can be rotated and their size can be changed with a pinch (they can't be moved).
  • This should aid visual learners grasp the concepts better.

img4

Now download the code and have fun!

About

A project on an ARKit based IOS app using text detection and manipulation for people with reading disabilities.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published