Doodle Me repository is a platform designed to transform the way individuals communicate by leveraging the simplicity of doodles. This Flutter-based project aims to bridge communication gaps through an intuitive and accessible application that interprets doodles into recognizable categories.
Doodle Me uses advanced machine learning algorithms, particularly a Convolutional Neural Network (CNN), to accurately recognize and classify doodles. The app is engineered to assist individuals who face challenges with traditional language-based communication, including those with disabilities and non-native language speakers. By converting simple sketches into predefined categories, Doodle Me enhances communication inclusivity and offers a new avenue for expression.
The application is built using Flutter, ensuring a cross-platform functionality. It integrates a CNN model developed in a separate repository dedicated to AI models (doodle me AI models), which processes and classifies the doodles.
- Framework: The application is developed in Flutter, which allows for a single codebase for both iOS and Android platforms.
- Machine Learning: Employs a CNN model for the recognition and classification of doodles. The model is trained and managed in a separate repository and is central to the app's functionality.
- Web Service: Utilizes a cloud-based web service deployed on Heroku for the preprocessing of doodle data. This service ensures that doodle inputs are appropriately formatted and ready for analysis by the CNN model (doodle me flask).
Doodle Me is available for download on the App Store. You can access it through the following link: Doodle Me on the App Store