Workflows¶
Overview¶
Workflows are a pre-defined set of rules and parameters used to perform automatic collection of subject data for determining liveness. Workflows are created by the Face Capture library using their name. Capture Sessions are started using a Workflow and Camera to manage its settings. Some settings are available for modification through the Workflow such as:
A username to identify the subject in the transmitted JSON for reporting purposes.
A timeout duration which stops the Capture Session once the specified amount of time has passed. Setting a zero second timeout disbales the Capture Session from timing out.
A capture profile specifying the criteria used to determine when to automatically capture the subject’s face.
Workflows are split into two categories: passive and active. Passive liveness workflows only require the user to position themselves in the frame so that a good quality capture can be completed. An active liveness workflow requires the user to perform a specific action in order to pass the liveness check.
Please see the (Face Capture Passive Workflow Summary
) table below for a general overview of the available passive liveness workflows.
Feature |
Charlie |
Foxtrot |
---|---|---|
Device Type |
Mobile Device |
Mobile Device |
Device Type - The type of device to use this workflow with. Different workflows perform captures differently based on device type. For example, using a mobile device workflow when capturing with a kiosk will result in inaccurate or poor performance.
Workflow Package Types¶
After a successful capture session, the back-end request package will be available through the Workflow. There are three different types of back-end package that can be obtained:
High Usability - A request for the back-end focussed on usability.
High Security - A request for the back-end focussed on security.
Balanced - A request for the back-end with a mix of security and usability.
It is recommended to start with a Balanced approach and adjust to either High Usability or High Security based on testing.
Active Liveness¶
Active liveness workflows are provided in Face Capture as a way of capturing the user doing a particular action Active liveness is provided primarily as a means of complying with laws and regulations that require active liveness to be used. Active liveness is less accurate than passive liveness and will generally lower the usability of an application without appreciably increasing the security. Due to this, we only recommend implementing active liveness when it is absolutely required. Further, we require that active liveness be used together with passive liveness to ensure that a full liveness check has been done. A server package can be created by using the Get Combined Server Package or Get Encrypted Combined Server Package functions with the passive liveness and active liveness workflows that successfully completed.
The actions that Face Capture supports are having the user look in one of four different directions. It is important to capture the user’s movement from start to finish, so we recommend providing the user written and visual instructions and have them initiate the capture when they are ready. This will provide the best results.
Active liveness workflows do not use a capture profile and will ignore any provided. Active liveness workflows are unaffected by setting the Enable Autocapture function.
These are the currently supported active liveness workflows:
active_up - The subject needs to look upwards.
active_down - The subject needs to look downwards.
active_left - The subject needs to look left.
active_right - The subject needs to look right.