Comparison to
Lai et al. 2009


We compare against method described in

Yu-Kun Lai and Shi-Min Hi and Ralph R. Martin,
Automatic and Topology-Preserving Gradient Mesh Generation for Image Vectorization,
ACM Transactions on Graphics 23(3), 2009

We used the author's implementation to create these results.


Boo
Super Mario World, © Nintendo

Input Nearest (16x) Our Method (16x) Lai et al., 2009


Mario and Yoshi 2
Super Mario World, © Nintendo

Input Nearest (8x) Our Method (8x) Lai et al., 2009


Bomber 4
Super Bomberman, © Hudson Soft

Input Nearest (16x) Our Method (16x) Lai et al., 2009


Yoshi 3
Super Mario World, © Nintendo

Input Nearest (16x) Our Method (16x) Lai et al., 2009