diffgram
Linking multiple box/polygon annotations
I understand support for Multi-Modal Annotation is on the roadmap; but would it be possible to get a rough release date?
Can you tell us a bit more about your use case? Multi-modal can mean a lot of different things :) Our long term intent is to be able to mix and match various methods arbitrarily but there may be some faster shorter term options
My bad. I was specifically looking for a feature where one can link multiple box/polygon annotations and adding a description to the relationship between the annotations.
Another use case for this feature would be for creating OCR datasets where character annotations are linked together and given a word label.
All good, For the linking one we have this https://github.com/diffgram/diffgram/issues/891 I added your example line That issue is relatively easy compared to say annotating image and video as one file. We'd welcome pull requests on it :)