3 ms·Generative Omnimatte Learning to Decompose Video into Layers1 points by xnx 2y agoxnx 2y agoVery impressive demo video of the capabilities of this model: https://www.youtube.com/watch?v=SD-VCNvTBg4 https://www.youtube.com/watch?v=SD-VCNvTBg4