Sign inSign up

echizen321/facial_recognition

By echizen321

Updated about 6 years ago

Image
0

275

echizen321/facial_recognition repository overview

TODO:
  1. write workflow of code

  2. explain deployment parameters

Usage

Run docker-compose up with the b/m docker-compose.yml and test endpoint with b/m API, body and check against expected results

docker-compose.yml
version: '3.3'

services:
    facial_recognition:
        ports:
            - '7777:7777'
        volumes:
            - '/home:/home'
        restart: unless-stopped
        environment:
            - MODEL_DIR=../20170512-110547
        image: echizen321/facial_recognition
docker-compose parameters
# workflow:
# 1) check if have vehicle
# 2) 
#   a) if have, check the vehicle for LP (tighter constraint for LP detection)
#   b) else check original image for LP anyway
# 3) if have LP, check LP for contents (OCR)
#
# env variables:
# - LP_THRESHOLD (default 0.2)
# - OCR_THRESHOLD (default 0.4)
# - MIN_LP_LENGTH (default 3)
API
Endpoint and Body
POST
http://172.17.0.1:7777/FacialRecognition
{
	"fileLocation": "/home/plato/Desktop/few_images/faces.jpg"
}
Expected Output
{
    "bounding_box_format": "top, left, bottom, right",
    "results": [
        {
            "boundingBox": [
                169,
                410,
                17,
                542
            ],
            "confidence": 0.055359518990338016,
            "embedding": [ ... ],
            "name": "Unknown"
        },
        ...
    ]
}
Workflow: how the code works

...

Tag summary

Content type

Image

Digest

Size

2.1 GB

Last updated

about 6 years ago

docker pull echizen321/facial_recognition