This application detects objects using TensorFlow.js, the CocoSSD model is already trained, it has 80 different classes for detection, such as Person, Cat, Dog, Cell Phone, Chair, etc.
To run this application it is necessary to have a WebCam to capture the video, the detection process is in real time, and to display the geometric shape the HTML Canvas was used.
$ npm installIf you have any errors in the installation process, run the command below:
$ npm install --legacy-peer-deps# Development mode
$ npm run dev
# Production mode
$ npm run build
$ npm run preview- Development mode: http://localhost:3000
- Production mode: http://localhost:4173