sahilsync07 / capty

Capty

Identify new things smartly!

About The Project

Aim of our project is to create a ML based Software Application that intelligently identifies the patterns in the given Image and converts it into a one line description in the form of text. It predicts the results on the basis of the patterns it has learnt during its Training phase.

Target audience

• Visually Impaired people
• People who have trouble identifying objects or remembering the name

Unique Selling Point

Never was an app made to help people recognize the scene in front of them. This app focuses on these scenarios:

• Blind people who can't see
• Color blind people who can't recognize color of object in front of them
• People who don't know the object in front of them

Tech Stack Used

• Flutter
• Firebase
• Flask
• Keras

Functionalities

• to click a photo and get a caption of what the photo is about

Packages used

UI

Prototype

Figma link: http://bit.ly/FigmaDesignCapty

Prototype link: http://bit.ly/PrototypeCapty

Project Report document: http://bit.ly/CaptyProjectReport

Usage

👉 Click a photo of anything, and let the application generate caption for you!

Future aspects

👉 It has future plans of getting integrated with Google search, and letting you know more information about it.

👉 A helmet with camera can be connected with the app via Bluetooth, enabling visually impaired people to get to know what is the scene in front of them via listening to the caption generated.

👉 Improving accuracy of caption generation.

👧 Contributors 👦

Gaurav Poosarla
Akansh Choudhary
Margi Bhatt
Sahil Kumar