Image

Method: General system overview. From a seeded user sketch, the silhouette image, the normal and depth maps are translated and a base mesh is determined. Using those, a differentiable renderer is used in order to predict a 3d model.
Method : General system overview. From a seeded user sketch, the silhouette image, the normal and depth maps are translated and a base mesh is determined. Using those, a differentiable renderer is used in order to predict a 3d model.

Publication

This image has been taken from the following publication:
2023
General system overview. From a seeded user sketch, the silhouette image, the normal and depth maps are translated and a base mesh is determined. Using those, a differentiable renderer is used in order to predict a 3d model. Kerstin Hofer (Department Creative Technologies, University of Applied Sciences Salzburg, Salzburg, Austria)
Sketch to 3d-Model using Deep Learning and Differentiable Rendering
[Method] [Thesis] [Source Code]		 Master Thesis