Quoting the researcher D. Ross in her prior art explaining how self-supervised video representation learning from rotation works:
Upside down
Boy, you turn me
Inside out
And round and round
Is this real life, is this just fantasy
- Self supervised learning in virtual environment(s), Bahnhof, Gorli - Symposium of Localisation Agnostic Canonical Knowledge, 2023. Xberg