I’m a fan of this idea. In fact I tried a similar strategy earlier this year (for object detection but without the auto segmentation) and posted the code and results: Synthetic Object Detection Dataset Generation Code and Tutorial
Would be interested to have others give it a try and see if their results are intriguing enough to continue development into something that could be fully automated.
An example of a simple model trained on only generated images:
I was thinking this could be a good way to bootstrap a dataset (and speed up labeling tasks by giving you a starting point to correct instead of having to annotate from scratch).