Art Suggester is an AI-based art recommendation tool designed to optimize image recognition techniques using a convolutional neural network (CNN). It identifies various art mediums such as paint, pencil crayons, and markers, providing personalized recommendations based on user preferences.
- Train an AI model capable of determining the medium of an artwork.
- Develop a database system to store users' favorited images for personalized recommendations.
The project uses a Convolutional Neural Network (CNN) built with TensorFlow to classify art mediums. The model was trained using images from various online databases consisting of paintings and drawings across different mediums such as paint, pencil crayons, and markers.
The datasets used for training the model were collected from publicly available online databases featuring artwork in different mediums. This diverse dataset ensures the model can effectively identify and classify various artistic styles and tools.
The database and user information, including favorited images, are hosted using MongoDB for efficient storage and retrieval.
Colab notebooks used to create the models can be found in the backend folder of the project’s GitHub repository.