0
\$\begingroup\$

I saw this video: Video
And now I am curious about the following: How can you achieve dynamic forms like they are shown in this video.
What I mean is: Do you need to have some "base" models which can be just linked together? Or is it possible to generate these 3D models "on the fly"?

If yes: How could that be done?
And: Is something like that as well possible in 2D?

And: How can you do animations, like (some) in the video? Because it seems as if the body parts of the creatures work together more or less correctly. As well here is the question: Would this be possible in 2D?

\$\endgroup\$
0
\$\begingroup\$

It's not possible to create them on the fly. What you can do is copy the same model thousand times in the memory, and make them slightly different. Then you just choose correct model for each frame.

One copy of the model per frame eats large amount of memory. To reduce memory consumption, you can choose one model per 2 seconds of animation, and interpolate the frames in between. You'll need a time range [a..b] and a function m(a) and m(b) gives you the models for the keyframes. Then shader interpolates the points between m(a) and m(b).

Same features work in 2d and 3d.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.