Monocular Depth Estimation on 360D (test) benchmark leaderboard · SOTA2 Research

Monocular Depth Estimation on 360D (test)

0.1113RMSERectNet

Evaluation Results

Method
RectNet2018.07	0.1113	0.008	0.0042	0.035	94.97	99.07	99.69	—	—
UResNet2018.07	0.1289	0.0097	0.0062	0.041	92.45	98.53	99.55	—	—
SliceNet2022.02	0.1323	—	—	0.0212	97.88	99.52	99.69	0.1134	—
SliceNet2022.08	0.1323	—	—	0.0212	97.88	99.52	99.69	0.1134	0.0467
ODE-CNN2022.02	0.1728	0.0467	—	—	98.14	99.67	99.89	—	—
Distortion-aware2022.02	0.1769	0.0406	—	—	98.65	99.66	99.87	—	—
GLPanoDepth(Complete)2022.02	0.1844	—	—	0.0313	98.34	99.65	99.87	0.0978	—
PanoDepth2022.02	0.1955	0.0456	—	—	98.3	99.57	99.84	—	—
GLPanoDepth(Concat)2022.02	0.2011	—	—	0.0344	97.86	99.61	99.87	0.1104	—
GLPanoDepth(ViT+CNN)2022.02	0.2316	—	—	0.0392	96.81	99.46	99.83	0.1279	—
SphereDepth2022.08	0.2364	—	—	0.0369	97.43	99.44	99.78	0.1145	0.055
BiFuse with fusion2022.02	0.244	0.0615	—	—	96.99	99.27	99.69	—	—
BiFuse2022.02	0.244	—	—	0.0428	96.99	99.27	99.69	0.1143	—
BiFuse2022.08	0.244	—	—	0.0428	96.99	99.27	99.69	0.1143	0.0615
FCRN2022.02	0.2833	0.0699	—	—	95.32	99.05	99.66	—	—
FCRN2022.02	0.2833	—	—	0.0473	95.32	99.05	99.66	0.1381	—
FCRN2022.08	0.2833	—	—	0.0473	95.32	99.05	99.66	0.1381	0.0699
RectNet2022.02	0.2911	0.0702	—	—	95.74	99.33	99.79	—	—
Mapped Convolution2022.02	0.2966	0.0965	—	—	90.68	98.54	99.67	—	—
Liu et al.2018.07	0.3048	0.0312	0.0532	0.107	60.3	84.12	93.38	—	—
Laina et al.2018.07	0.3152	0.03	0.0549	0.1033	63.53	86.16	94.12	—	—
OmniDepth2022.02	0.3171	—	—	0.0725	90.92	97.02	98.51	0.1706	—
OmniDepth2022.08	0.3171	—	—	0.0725	90.92	97.02	98.51	0.1706	0.0931
GLPanoDepth(CViT)2022.02	0.3555	—	—	0.0604	92.19	98.51	99.51	0.1791	—
Godard et al.2018.07	1.6559	0.0453	0.1743	0.1958	45.24	70.23	83.15	—	—