Image meshing via hierarchical optimization

Hao XIE,Ruo-feng TONG

Fig. 1 Inputs and outputs. Given an image, 'Lena' (a), and the sampling ratio (1% in this example), our algorithm generates a mesh-based vector graphic (d) with its rendered result (b). Detail regions of both (a) and (b) are enlarged and shown in (c), in which upper and lower parts correspond to (a) and (b), respectively. References to color refer to the online version of this figure