Link to the paper: <a href="https://arxiv.org/abs/1803.04758" rel="nofollow">https://arxiv.org/abs/1803.04758</a>
Oh how I hate marketing speech.<p>First of all, the title should include "video of a predefined 360° turn".<p>And then they say something along the lines of "average accuracy of about 5mm" for joining the constructed modeled joints to their model, while you see the body wobbling around happily.<p>This is an impressive demo, but gah!
Structure from motion is an existing technique. What is the contribution of ML in this case (it seems like joint positioning maybe?)?<p><a href="https://en.m.wikipedia.org/wiki/Structure_from_motion" rel="nofollow">https://en.m.wikipedia.org/wiki/Structure_from_motion</a>
How is this ML? They use a CNN for foreground segmentation, a minor step in their pipeline. But the major contribution seems to be putting the silhouettes in a common reference frame. I sincerely hope sciencemag isn’t putting ML in the title purely to jump on the bandwagon.
I wonder if will see a future soon where a director can fully edit the positions and physical actions of the actors at post production.<p>basically, the whole scenes will be transferred to believable 3d models seemlessly, and you can reanimate parts of everything. I feel like that's doing to happen for sure, for big Hollywood productions at least (like the Marvel stuff)