'Is there a dlib shape_predictor confidence value that can be used as a threshold for setting points?
I am trying to use dlib.shape_predictor to set points of interest and then measure them. However, there are instances where one or more points shouldn't be shown as the location of interest is not visible, making their placement incorrect. Is there a way to set a confidence threshold to the shape_predictor? :)
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|