Hello Nvidia,
Is there a way to get a the omniverse replicator annotator to generate .xml format file (pascalVoc) ? I want train the generated dataset using tensorflow which requires .xml annotation. Or is there converter for this?
Hello,
Yes, you could translate the data yourself into xml by creating a Custom Writer
Data from annotators comes in a Python dictionary, and we mostly output to .json.
You could create your own custom writer or sub-class an existing writer, but override the write() function to output xml instead of json