SearchArt is an innovative web application developed from scratch, utilizing foundational web technologies such as HTML, CSS, PHP, and JavaScript without relying on any frameworks. This project aims to provide a unique user experience by allowing users to upload their own dataset of images and interact through voice commands. When a user pronounces a word, it is recognized as a label, and the app then displays artworks containing objects semantically represented by that word.

At the core of SearchArt is the integration of a neural network using the YOLO (You Only Look Once) library, which enables the efficient identification of objects within the images. This feature allows for an immersive and interactive way of exploring art based on specific elements or themes.

Additionally, the application includes a testing feature with a pre-loaded set of images already tagged with labels. This functionality enables users to experience and interact with the app, exploring its capabilities and the power of neural network-based image recognition in the context of art discovery.

• User-friendly interface to upload custom datasets of images.
• Voice recognition capability for intuitive user interaction.
• Neural network integration using YOLO for accurate object detection within artworks.
• A pre-loaded image set for immediate testing and exploration.
• Developed using basic but powerful web technologies, ensuring broad accessibility and ease of use.

SearchArt is designed not only as a tool for art enthusiasts and researchers but also as an educational platform demonstrating the practical application of neural networks in recognizing and classifying elements within visual art. It stands as an example of the potential of combining traditional web development with advanced machine learning techniques to create novel and engaging user experiences.

For a detailed overview of the project, please refer to the project presentation available in this repository: Presentation PDF