Docker Deployment Patterns

Each new SmartFace release for the Docker environment is available on our Github, specifically on the SmartFace repository https://github.com/innovatrics/smartface. For the ease of use we have prepared premade Docker Deployment Patterns. These represent the most commonly used situations and use cases.

When you deploy the repository to your local directory, you can see each Deployment Pattern has it’s own folder within the smartface/sf-docker/ directory.

Currently we provide these sample Docker Deployment Patterns:

• all-in-one
• LFIS
• nvidia-jetson

To enable hassle-free update between versions, please consider keeping the production deployment outside of the local repository clone. This way an updating the local repository to the latest version will not overwrite your specific deployment configuration. Changes related to each update are available in the Table of Releases and the Upgrades section of the Releases.

You can find more information about the scaling of services and updating the environment variables.

all-in-one#

The all-in-one is the most universal Deployment Pattern as it provides each type of SmartFace Platform service. On top of the SmartFace Platform services It also provides the SmartFace Station and the Access Controller module with sample configurations.

This pattern provides a scalable server platform for facial and pedestrian detection which can be easily integrated with third-party systems. It can process and analyze multiple real-time video streams in parallel.

SmartFace Platform covers most typical use cases when it comes to facial recognition and biometrics. It allows you to process video files, images or live video streams from RTSP cameras or smart embedded devices running SmartFace Embedded. Results of processing are stored into a database in the form of structured data together with images and can be accessed through an available REST API. Information is also provided through real-time notifications.

To find out how to alter this preset to better suit your project, please take a look at Maintenance for Docker guide. For information how to install this Deployment Pattern from the scratch, please read the Docker Installation steps.

Content#

LFIS#

The LFIS Deployment Patters represents the Lightweight Facial Identification Service.

The Lightweight Facial Identification Service is a lightweight, powerful, scalable, multi-platform and easy-to-deploy solution for facial biometrics use cases easily integrated to any third-party system via the REST API. Available for fast cloud and on premise deployment.

It supports advanced identity management, identities enrollment, verification, identification, and liveness check.

To find out how to alter this preset to better suit your project, please take a look at Maintenance for Docker guide. For information how to install this Deployment Pattern from the scratch, please read the Docker Installation steps.

Content#

nvidia-jetson#

NVIDIA Jetson is based on ARM architecture, specifically the ARM Cortex-A series of processors. It’s designed for low power consumption and optimized for AI and edge computing tasks. Jetson devices are known for their power efficiency, making them suitable for embedded and edge computing applications where power consumption is a critical factor.

Due to a different nature of the nvidia-jetson devices the SmartFace Platform services in the nvidia-jetson Deployment Pattern are using different docker images from the SmartFace images repository. They have a prefix sf-jetson- so they are clearly distinquishable.

Content#