TABLE 4.

Quantitative comparison results between the proposed method and other methods on nyud-v2 test set.

Method	Input	ODS	OIS	AP
gPb-UCM (Arbelaez et al., 2010)	RGB	0.631	0.661	0.562
SE (Dollár and Zitnick, 2014)		0.695	0.708	0.719
gPb + NG (Gupta et al., 2013)		0.687	0.716	0.629
SE + NG + (Gupta et al., 2014)		0.706	0.734	0.549
OEF (Hallman and Fowlkes, 2015)		0.651	0.667	0.653
HED (Xie and Tu, 2015)	RGB	0.717	0.732	0.704
	HHA	0.681	0.695	0.674
	RGB-HHA	0.741	0.757	0.749
RCF (Liu et al., 2017)	RGB	0.729	0.742	0.693
	HHA	0.705	0.715	0.650
	RGB-HHA	0.757	0.771	0.749
LPCB (Deng et al., 2018)	RGB	0.739	0.754	–
	HHA	0.707	0.719	–
	RGB-HHA	0.762	0.778	–
DRC (Cao et al., 2020)	RGB	0.749	0.762	0.718
	HHA	0.711	0.722	0.677
	RGB-HHA	0.769	0.782	0.771
LRC (Lin et al., 2020)	RGB	0.737	0.750	0.686
	HHA	0.697	0.708	0.642
	RGB-HHA	0.759	0.771	0.748
BDCN (He et al., 2019)	RGB	0.748	0.763	0.770
	HHA	0.707	0.719	0.731
	RGB-HHA	0.765	0.781	0.813
AMH-Net-ResNet50 (Xu et al., 2018)	RGB	0.744	0.758	0.765
	HHA	0.716	0.729	0.734
	RGB-HHA	0.771	0.786	0.802
PiDiNet (Su et al., 2021)	RGB-HHA	0.756	0.773	–
EDTER (Pu et al., 2022)	RGB	0.774	0.789	0.797
MEDNet	RGB	0.752	0.766	0.723
	HHA	0.711	0.723	0.681
	RGB-HHA	0.772	0.787	0.776

The best two results are marked with red and blue, respectively.