A 3D sunken-relief generation method of human faces from depth images of feature lines

In this paper, we propose a line-based 3D sunken-relief generation method that considers the influence of light direction on the width and depth of engraved lines. Firstly, the height field of a human face model is compressed to generate the sunken-relief base mesh. Secondly, several feature lines representing the facial features are extracted. Next, the engraved width of lines is determined based on the curvature of the feature lines and the light direction. Our approach then adopts the 3D engraved stroke technique to vary the depth of feature lines in 2D image space. Finally, the depth information of the engraved lines is added to the base mesh to generate the 3D sunken-relief of a human face. Experimental results show that the sunken-relief generated by our method is better able to express the facial features and reflect illumination changes under different light directions, resulting in a more visually pleasing piece of art.