Skip to content

Latest commit

 

History

History
67 lines (47 loc) · 3.03 KB

Assistance.md

File metadata and controls

67 lines (47 loc) · 3.03 KB
Raw

ASSISTANCE

Team members

  • Aayushi Joshi
  • Aayush Shah
  • Nishtha Sainger
  • Priyanka Avadh
  • Vedika Mamidwar

Mentor

  • Anuj Raghani

Description

Assistance is an app made for the visually impaired with a motive to give them a feeling of independence. Assistance is a mobile application which serves as a “guide dog”. Whenever they step out on the road, all they have to do is open the application and let it guide them as they proceed. The app is supported with a ML model at the backend which detects the objects around. The objects hence detected are told to the user as a speech output which helps them to walk freely and confidently.

  • Object detection model at the backend
  • User friendly UI
  • Output in the form of speech

Along with this, add the link of the drive folder that contains the app APK/Screenshots/Screen Recordings. If you have hosted your project on the web, add that link as well.

Technology stack

Tools and technologies that you learnt and used in the project.

  1. Flutter and Dart (Frontend)
  2. Python (Backend)
  3. Flask (Backend + Integration)
  4. YOLO (ML model for object detection)

Applications

The project ASSISTANCE aims at helping visionless people. Imagine a day in your life without your eyes. You won't be able to read books, attend online lectures, scroll social media and most importantly you would require constant help and support from others, you would be dependent on someone throughout your day. This is definitely not a great feeling. We understood their pain and we made an attempt to help the blind people in taking their first step towards being independent.

What did you learn from this project?

  1. Aayushi Joshi - Machine Learning (CNN) and Flask, team work.
  2. Aayush Shah - Machine Learning (CNN) and Flask
  3. Nishtha Sainger - Learnt Flutter, worked with a team to bring an idea to life.
  4. Priyanka Avadh - Flutter
  5. Vedika Mamidwar - Flask API, Integration with flutter using flask.

Future scope

We will deploy the app which will make the user experience much better. We can connect this detection-model to a wearable-physical camera which will make a new set of 'eyes' for the visually imapired. We can increase precision of the object-detection model.

Screenshots

These are photos of some the obstacles that the app can detect.

Screeshot of Sofa image

Screenshot of cat image

Screenshot of oven image

Screenshot of dog image