I suggest watching the lecture recording for this slide as it's animated. You'll be able to see how the simulated cloth drops and matches the photograph!
pranavkolluri
Is there a particular reason that the cloth starts out curved? How does one typically construct a default/starting shape to begin with? Presumably certain starting conditions are more optimal than others.
rcorona
@pranavkolluri I think that's a great question, somehow it feels like the experimental setup is somewhat underspecified?
I skimmed the figure and its description in the paper and the starting conditions don't appear to be described.
I feel like from the slack in the sheet, one could imagine that the sheet could also bend towards the wall rather than towards the camera under certain starting conditions. Makes me wonder that if the starting conditions of the simulator aren't matched closely to the starting conditions of the real-world sheet, then the comparison being made might not be completely fair?
GarciaEricS
It's amazing that such good looking simulations can be made with relatively simple assumptions that materials are fundamentally masses and springs. Especially because intuitively seems like to would make zero sense and not apply to the situation. I suppose the assumption just captures the fact that you can treat a point on the material as only depending on the state of other points around it, and this spring set up supports this rather well.
stang085
In the photo, the cloth also has a lot of wrinkles compared to the simulation... I wonder how you would be able to simulate that in 3D
I suggest watching the lecture recording for this slide as it's animated. You'll be able to see how the simulated cloth drops and matches the photograph!
Is there a particular reason that the cloth starts out curved? How does one typically construct a default/starting shape to begin with? Presumably certain starting conditions are more optimal than others.
@pranavkolluri I think that's a great question, somehow it feels like the experimental setup is somewhat underspecified?
I skimmed the figure and its description in the paper and the starting conditions don't appear to be described.
I feel like from the slack in the sheet, one could imagine that the sheet could also bend towards the wall rather than towards the camera under certain starting conditions. Makes me wonder that if the starting conditions of the simulator aren't matched closely to the starting conditions of the real-world sheet, then the comparison being made might not be completely fair?
It's amazing that such good looking simulations can be made with relatively simple assumptions that materials are fundamentally masses and springs. Especially because intuitively seems like to would make zero sense and not apply to the situation. I suppose the assumption just captures the fact that you can treat a point on the material as only depending on the state of other points around it, and this spring set up supports this rather well.
In the photo, the cloth also has a lot of wrinkles compared to the simulation... I wonder how you would be able to simulate that in 3D