The vvas_xmetaconvert plug-in is used to convert VVAS ML inference metadata, which is attached to an input buffer, into the overlay metadata, GstVvasOverlayMeta, that is needed by the vvas_xoverlay plug-in.
The vvas_xmetaconvert plug-in requires some static configuration parameters, such as font type and size, which are provided in the configuration file in JSON format. The config-location property allows you to specify the location of the configuration file that contains the necessary parameters.
By using these two plug-ins together, you can create an overlay, like bounding box, text, and so on, on a video frame/image that displays the ML inference results in real time.
For insight into the implementation details, refer to the vvas_xmetaconvert source code located at <vvas>/vvas-gst-plugins/gst/metaconvert/.