Camunda Team Blog


Listen up, keep your privates private!

Written by Rob Parker on , under Execution category.
In a recent blog post, I described a use case based on facial image recognition to perform an authentication task. In an ideal process, the recognition system either passes or fails the recognition task and the process proceeds via straight through processing to completion. In practice there is a fuzzy boundary between these two outcomes where human judgement via a human task is injected into the process. Now consider the case where say an image of a driver licence is used as a reference image.