Table 1 Depth accuracy and error of different methods on the KITTI dataset.

Methods	Abs Rel	Sq Rel	RMSE	RMSE log	\(\delta\)< 1.25	\(\delta\)< 1.25²	\(\delta\)< 1.25³	Params
GoNet[20]	0.149	1.060	5.567	0.226	0.796	0.935	0.975	31.6 M
DDVO[21]	0.151	1.257	5.583	0.228	0.810	0.936	0.974	28.1 M
Monodepth[11]	0.148	1.344	5.927	0.247	0.803	0.922	0.964	20.2 M
Monodepth2[12]	0.115	0.903	4.863	0.193	0.877	0.959	0.981	14.3 M
FastDepth[15]	0.150	0.890	5.321	0.207	0.808	0.945	0.981	3.96 M
R-MSFM6[22]	0.112	0.806	4.704	0.191	0.878	0.960	0.981	3.5 M
Lite-HR-Depth[23]	0.116	0.845	4.841	0.190	0.866	0.957	0.982	3.1 M
MonoFormer[24]	0.104	0.846	4.580	0.183	0.891	0.962	0.982	23.9 M+
Lite-Mono[25]	0.107	0.765	4.561	0.183	0.886	0.963	0.983	3.1 M
Ours	0.105	0.810	4.506	0.183	0.891	0.963	0.983	3.0 M

Quick links

Search