Skip to content

README.md

Latest commit

 

History

History
50 lines (32 loc) · 1.57 KB

README.md

File metadata and controls

50 lines (32 loc) · 1.57 KB
Raw

Faces Client

Overview | Get the MLrun Demos | Go to the Client Directory | Configure the Client Initialization | Run the Client

Overview

The faces-demo client demonstrates ingestion of video images from a webcam to the Iguazio Data Science Platform ("the platform") using web APIs.

Clone the MLRun Demos

Clone the mlrun/demos repository into your laptop (Git clone URL — https://github.com/mlrun/demos.git). it is required in order to use your laptop camera and stream its frames into the iguazio system

Go to the Client Directory

cd mlrun/demos/realtime-face-recognition/client

Configure the Client Initialization

Edit the config/init.ini client-initialization file to configure the client initialization; replace the <nuclio endpoint> placeholders to match your specific environment. set the nuclio endpoint into nuclio-api-serving-function the function can be found under Projects(tab)->faces->

[app]
log_level=info

[nuclio]
url = <nuclio endpoint>

Run the Client

python video_capture.py

Client Response

unless you are a famous actor the expected response from the client is : [{"coords": [75, 889, 259, 706], "name": "unknown", "label": -1, "confidence": 0.3509389671540804, "encoding": [-0.20812971889972687, 0.13344672322273254, ....]}]'