author: Tobias Schoch

A small web service running in a docker container that enables face recognition in images.

• changed to nas paths
• changed to facerec==0.1.5
• changed to service, working on windows
• add .env to gitignore
• added a compose for the service

• changed to python-facerec v0.1.0
• fixed config persistency
• added usage documentation
• exposed configuration /config/ (threshold parameter)
• changelog adapted

• added delete and patch faces and general RESTfulness
• fixed tests

• show number of stored faces at startup
• persistent data base configuration
• added tests for flask service
• working docker file for facerec server
• add working python-facerec
• added the service api for code and image requests
• added docker frame and package
• rerouted and nice rectangle and fontsize
• added image scaling and font size
• improved form, added routes
• basic teach and identify over web
• some small changes
• working upload
• adapted identify page
• added data directory
• update readme
• initial service draft

• initial version

docker run -v data:/data -p 8081:80  --name facerec-server shocki/facerec

Faces database is stored in /data. The web services are exposed on port 80.

To install use pip:

git clone https://github.com/toschoch/docker-facerec.git
pip install -r docker-facerec/requirement.txt
python facerec_service.py

Webinterface with browser on port 80.

REST-API Endpoints:

• /faces (GET) list of faces
• /faces/<id or name> (GET, PATCH, DELETE) get, modify or delete face entry
• /config (GET, PATCH) get or modify configuration parameters
• /image/indentify (POST) post an image and identify persons on it
• /image/teach (POST) post an image and a name to teach the system a face
• /facecode/identify (POST) post a facecode and identify persons on it
• /facecode/teach (POST) post a facecode and a name to teach the system a face

Web-API Endpoints:

• /, /identify browser interface for identification
• /teach browser interface for teaching