Genetic Feature Fusion for Object Skeleton Detection

Table 1

Performance comparison of the state-of-the-art approaches on commonly used skeleton\symmetry detection datasets.

Methods	Datasets
	SK-LARGE [30]	SK-SMALL [5]	WH-SYMMAX [31]
MIL [35]	0.353	0.392	0.365
HED [4]	0.497	0.541	0.732
RCF [36]	0.626	0.613	0.751
FSDS [5]	0.633	0.623	0.769
SRN [6]	0.658	0.632	0.443
LSN [10]	0.668	0.633	0.797
Hi-Fi [8]	0.724	0.681	0.805
DeepFlux [37]	0.732	0.695	0.840
GFF (ours)	0.736	0.723	0.837