Node Object Detection

This project uses the node-yolo, a Node.js Addon to detect objects in a picture.
Is part of the vapi-server.

Installing

npm install

Deployment

npm start

Config File

config.json

Docker

You can also use the node-object-detection with Docker

CUDA

Build the container

cd docker/cuda && docker-compose build

Run it

docker run --runtime=nvidia -v <Motion output folder>:<Motion output folder> vapi/node-object-detection-cuda

CPU

Build the container image for CPU

cd docker/cpu && docker-compose build

Run it

docker run -v <Motion output folder>:<Motion output folder> vapi/node-object-detection