Reconstructing 3D Model from Single-View Sketch with Deep Neural Network

Table 5

Character models (human drawing).

	Ours	ShapeMVD	Nearest retrieval	Tatarchenko et al. [15]	[15]+U-net	Volumetric decoder	R2N2
Hausdorff distance	0.102	0.117	0.188	0.139	0.136	0.178	0.168
Chamfer distance	0.013	0.021	0.036	0.025	0.024	0.032	0.036
Normal distance	28.22	33.44	43.81	36.11	34.74	54.91	54.29
Depth map error	0.012	0.026	0.040	0.031	0.027	0.037	0.040
Volumetric distance	0.217	0.298	0.458	0.342	0.307	0.420	0.436