https://hwang26.github.io/image-classifier/
A interface which displays a doodle prompt by fetching one the 345 categories utilized in the DoodleNet classifier. Users can attempt to draw the prompt in the provided p5 JS drawing pane which will dynamically return classification results from ml5 JS. Users can clear the pane to reattempt the drawing or refresh the page to be given a new prompt.
- Figma for prototyping
- Storybook for component design
- React, Node JS
- Next JS for static site generation for React
- SASS styling
- p5 JS for drawing functionality
- ml5 JS for image classification using DoodleNet model
- Github Actions + Github Pages for deployment