This is an object detection app made using TensorFlow COCO SSD and ReactJS. It can detect objects in real time from the webcam.
Requirements Node.js ReactJS TensorFlow COCO SSD model Installation Clone the repository Install the dependencies npm install
npm start
The app uses the TensorFlow COCO SSD model to detect objects in real time. The model is pre-trained on the COCO dataset, which contains over 200,000 images of objects from 80 different categories.
When the app is started, it loads the TensorFlow model and initializes the ReactJS server. The ReactJS server renders the app's UI, which includes a live preview of the webcam feed.
The app continuously captures frames from the webcam and sends them to the TensorFlow model for inference. The model then returns a list of bounding boxes and classes for each object detected in the frame.
The ReactJS server then renders the bounding boxes and classes on the app's UI.
I hope this is helpful!
