Depth Estimation on Stanford2D3D Panoramic

Metric: absolute relative error (lower is better)

LeaderboardDataset

Loading chart...

Results

Sort:

#	Model↕	absolute relative error▲	Extra Data	Paper	Date↕	Code
1	PanoFormer	0.0405	No	PanoFormer: Panorama Transformer for Indoor 360 ...	2022-03-17	Code
2	HiMODE	0.0532	No	HiMODE: A Hybrid Monocular Omnidirectional Depth...	2022-04-11	-
3	Neural Contourlet Network	0.0558	No	Neural Contourlet Network for Monocular 360 Dept...	2022-08-03	Code
4	NLFB	0.0649	No	Improving 360 Monocular Depth Estimation via Non...	2021-09-22	Code
5	SliceNet	0.0744	No	-	-	-
6	OmniFusion (2-iter)	0.095	No	OmniFusion: 360 Monocular Depth Estimation via G...	2022-03-02	Code
7	FreDSNet	0.0952	No	FreDSNet: Joint Monocular Depth and Semantic Seg...	2022-10-04	Code
8	PanoDepth	0.0972	No	PanoDepth: A Two-Stage Approach for Monocular Om...	2022-02-02	-
9	ACDNet	0.0984	No	ACDNet: Adaptively Combined Dilated Convolution ...	2021-12-29	Code
10	HoHoNet (ResNet-101)	0.1014	No	HoHoNet: 360 Indoor Holistic Understanding with ...	2020-11-23	Code
11	UniFuse with fusion	0.1114	No	UniFuse: Unidirectional Fusion for 360$^{\circ}$...	2021-02-06	Code
12	BiFuse++	0.1117	No	BiFuse++: Self-supervised and Efficient Bi-proje...	2022-09-07	Code
13	SphereDepth	0.1158	No	SphereDepth: Panorama Depth Estimation from Sphe...	2022-08-29	-
14	BiFuse with fusion	0.1209	No	-	-	Code
15	DisConv	0.176	No	-	-	-
16	OmniDepth	0.1996	No	OmniDepth: Dense Depth Estimation for Indoors Sp...	2018-07-25	Code

#1PanoFormerSOTA
0.0405
absolute relative error· 2022-03-17
PanoFormer: Panorama Transformer for Indoor 360 Depth Estimation Code
#2HiMODE
0.0532
absolute relative error· 2022-04-11
HiMODE: A Hybrid Monocular Omnidirectional Depth Estimation Model
#3Neural Contourlet Network
0.0558
absolute relative error· 2022-08-03
Neural Contourlet Network for Monocular 360 Depth Estimation Code
#4NLFBSOTA
0.0649
absolute relative error· 2021-09-22
Improving 360 Monocular Depth Estimation via Non-local Dense Prediction Transformer and Joint Supervised and Self-supervised Learning Code
#5SliceNet
0.0744
absolute relative error
No paper
#6OmniFusion (2-iter)
0.095
absolute relative error· 2022-03-02
OmniFusion: 360 Monocular Depth Estimation via Geometry-Aware Fusion Code
#7FreDSNet
0.0952
absolute relative error· 2022-10-04
FreDSNet: Joint Monocular Depth and Semantic Segmentation with Fast Fourier Convolutions Code
#8PanoDepth
0.0972
absolute relative error· 2022-02-02
PanoDepth: A Two-Stage Approach for Monocular Omnidirectional Depth Estimation
#9ACDNet
0.0984
absolute relative error· 2021-12-29
ACDNet: Adaptively Combined Dilated Convolution for Monocular Panorama Depth Estimation Code
#10HoHoNet (ResNet-101)SOTA
0.1014
absolute relative error· 2020-11-23
HoHoNet: 360 Indoor Holistic Understanding with Latent Horizontal Features Code
#11UniFuse with fusion
0.1114
absolute relative error· 2021-02-06
UniFuse: Unidirectional Fusion for 360$^{\circ}$ Panorama Depth Estimation Code
#12BiFuse++
0.1117
absolute relative error· 2022-09-07
BiFuse++: Self-supervised and Efficient Bi-projection Fusion for 360 Depth Estimation Code
#13SphereDepth
0.1158
absolute relative error· 2022-08-29
SphereDepth: Panorama Depth Estimation from Spherical Domain
#14BiFuse with fusion
0.1209
absolute relative error
No paperCode
#15DisConv
0.176
absolute relative error
No paper
#16OmniDepthSOTA
0.1996
absolute relative error· 2018-07-25
OmniDepth: Dense Depth Estimation for Indoors Spherical Panoramas Code