I wondered whether it is possible with the help of a neural network to form a picture based on the frame on which the 3D model is depicted. Something like the one in the picture. enter image description here I have only one idea - to type lines primitives, and then try to approximate the 3D model with a set of these primitives, are there any other approaches to solving this problem?

  • For these purposes, well-suited "classical mechanics." It is done using various shaders. DirectX has a similar example in sdk (if I 'm not confusing anything) - ArchDemon
  • But how realistic is the result? - Oleg Belogay
  • I just wouldn’t like to get a loot from Ninja Batman as a result - Oleg Belogay
  • unity3d.ru/distribution/viewtopic.php?f=35&t=38806 may find something similar - ArchDemon
  • Thanks for the help, here the idea is not in simple drawing, but in imitation of drawing from scratch, as in the attached image - Oleg Belogai

0