Moodscape is an Android application built with Flutter that uses AI technology to analyze handwritten or printed text and facial expressions from images. It classifies both into psychological traits and provides personalized suggestions and motivation based on the results.
- Flutter – Main framework for building the app
- TensorFlow Lite – For running machine learning models locally on Android devices
- Teachable Machine by Google – Used to train and export image classification models
- Camera / Gallery Input – For capturing or selecting images from the device
- The app detects text from images using Optical Character Recognition (OCR).
- Then it classifies the detected text into one of 4 DISC personality types:
- Dominance
- Compliance
- Steadiness
- Influence
- Detects faces in an image and classifies the expression into one of the 7 emotional categories:
- Anger
- Sadness
- Surprise
- Happy
- Disgust
- Fear
- Neutral
- Based on both classification results (text and facial expression):
- Provides descriptions of two personality traits.
- Offers personalized suggestions and motivational insights.
- Take or select a photo that includes facial expression and/or text.
- Moodscape analyzes the image using on-device AI.
- Receive trait analysis and personalized motivation instantly.
-
Models are trained using Teachable Machine.
-
Exported as TensorFlow Lite models and integrated into the Flutter app for local inference.
-
The app requires access to:
-
Camera
Storage (to pick images from gallery)
📄 License MIT License