Monocular panoramic depth estimation on Stanford2D3D benchmark leaderboard · SOTA2 Research

Monocular panoramic depth estimation on Stanford2D3D

93.94Delta 1 AccuracyPanoFormer

Evaluation Results

Method	Links
PanoFormer2022.03		93.94	98.38	99.41	0.3083	0.0405	0.0619	—
Neural Contourlet Network2022.08		91.4	97.2	99.03	0.3528	0.0558	0.0982	0.0339
HoHoNet2022.08		90.54	96.93	98.86	0.3834	0.1014	0.2027	0.0668
SliceNet2022.03		90.31	97.23	98.94	—	0.0744	0.1048	—
SliceNet2022.08		90.29	96.26	98.44	0.3684	0.0744	0.1048	0.0823
UniFuse2022.03		87.11	96.64	98.82	0.3691	—	0.2082	—
Unifuse2022.08		87.11	96.64	98.82	0.3691	—	0.2082	0.0721
Bifuse2022.03		86.6	95.8	98.6	0.4142	0.1209	0.2343	—
Bifuse2022.08		86.6	95.8	98.6	0.4142	0.1209	0.2343	0.0787
FCRN2022.03		72.3	92.07	97.31	0.5774	0.1837	0.3428	—
FCRN2022.08		72.3	92.07	97.31	0.5774	0.1837	0.3428	0.11
OmniDepth2022.03		68.77	88.91	95.78	0.6152	0.1996	0.3743	—
OmniDepth2022.08		68.77	88.91	95.78	0.6152	0.1996	0.3743	0.1212