OpenDataCam is a project based on Yolo (Darknet) to detect moving object real time. For more information, please check origin git and below video
| 👉 UI Walkthrough (2 min, OpenDataCam 3.0) | 👉 UI Walkthrough (4 min, OpenDataCam 2.0) | 👉 IoT Happy Hour #13: OpenDataCam 3.0 |
|---|---|---|
 |
 |
 |
This project enables you to run opendatacam on a GPU enabled Volterra CE. We assume you have basic knowledge of how to use Volterra
Create a fleet with the following config
Build your volterra site and attach above fleet
Download kubeconfig and use it for the following command
git clone https://github.com/risyou/ves-k8s-opendatacam.git
cd ves-k8s-opendatacam#Create configmap from config.json, it will be used in later deployment
kubectl create configmap opendatacam --from-file=config.json --dry-run -o yaml | kubectl apply -f -
#Create mongodb to store data
kubectl apply -f opendatacam/0001-mongodb-deployment.yaml
#Create opendataacm deployment
kubectl apply -f opendatacam/0002-opendatacam-deployment.yamlYou will have the following service and deployments
[shali@:~]$kubectl get service
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
mongo ClusterIP 192.168.205.208 <none> 27017/TCP 2d17h
opendatacam ClusterIP 192.168.22.123 <none> 80/TCP,8071/TCP,8090/TCP 2d17h
[shali@:~]$kubectl get deployment
NAME READY UP-TO-DATE AVAILABLE AGE
opendatacam 1/1 1 1 6h1m
opendatacam-mongo 1/1 1 1 2d17h
Create origin pool and loadbalancer, see ves-example.json for example
Access your loadbalancer, by default it will shows you an demo provided by opendatacam, just drag and drop your video there!
config.json contains the parameter you can customeize, for example
- VIDEO_INPUT: Chose from VIDEO_INPUTS_PARAMS{key} and input parameter in the {value}
- DISPLAY_CLASSES: Chose from yolo object store
After update the config.json you need to rebuild configmap and rebuild pod
kubectl create configmap opendatacam --from-file=config.json --dry-run -o yaml | kubectl apply -f -