Depth Estimation on KITTI Eigen split unsupervised

Metric: Delta < 1.25^2 (higher is better)

LeaderboardDataset

Loading chart...

Results

Sort:

#	Model↕	Delta < 1.25^2▼	Extra Data	Paper	Date↕	Code
1	SPIdepth	0.973	No	SPIdepth: Strengthened Pose Information for Self...	2024-04-18	Code
2	Jasmine	0.972	No	Jasmine: Harnessing Diffusion Prior for Self-sup...	2025-03-20	-
3	SCIPaD	0.97	No	SCIPaD: Incorporating Spatial Clues into Unsuper...	2024-07-07	Code
4	NimbleD-LiteMono-8M	0.97	No	NimbleD: Enhancing Self-supervised Monocular Dep...	2024-08-26	Code
5	ProDepth	0.969	No	ProDepth: Boosting Self-Supervised Multi-Frame M...	2024-07-12	Code
6	DCPI-Depth (M+1024x320)	0.969	No	DCPI-Depth: Explicitly Infusing Dense Correspond...	2024-05-27	-
7	MonoViT(MS+1024x320)	0.969	No	MonoViT: Self-Supervised Monocular Depth Estimat...	2022-08-06	Code
8	NimbleD-LiteMono	0.969	No	NimbleD: Enhancing Self-supervised Monocular Dep...	2024-08-26	Code
9	Nimbled-SwiftDepth	0.969	No	NimbleD: Enhancing Self-supervised Monocular Dep...	2024-08-26	Code
10	CREMono(M + 1024x320 + Res50)	0.969	No	-	-	-
11	PlaneDepth (S + 1280x384)	0.968	No	PlaneDepth: Self-supervised Depth Estimation via...	2022-10-04	Code
12	Manydepth2(M+640x192)	0.968	No	Manydepth2: Motion-Aware Self-Supervised Multi-F...	2023-12-23	Code
13	Manydepth2-NF(M+640x192)	0.968	No	Manydepth2: Motion-Aware Self-Supervised Multi-F...	2023-12-23	Code
14	DIFFNet (MS+1024x320)	0.968	No	Self-Supervised Monocular Depth Estimation with ...	2021-10-18	Code
15	Nimbled-MD2-R50	0.968	No	NimbleD: Enhancing Self-supervised Monocular Dep...	2024-08-26	Code
16	Nimbled-SwiftDepth-S	0.968	No	NimbleD: Enhancing Self-supervised Monocular Dep...	2024-08-26	Code
17	DCPI-Depth (M+640x192)	0.967	No	DCPI-Depth: Explicitly Infusing Dense Correspond...	2024-05-27	-
18	TransDSSL	0.967	No	-	-	Code
19	ProDepth(M+640x192)	0.967	No	ProDepth: Boosting Self-Supervised Multi-Frame M...	2024-07-12	Code
20	NimbleD-LiteMono-S	0.967	No	NimbleD: Enhancing Self-supervised Monocular Dep...	2024-08-26	Code
21	Nimbled-MD2-R18	0.967	No	NimbleD: Enhancing Self-supervised Monocular Dep...	2024-08-26	Code
22	EPCDepth(S+1024x320)	0.966	No	Excavating the Potential Capacity of Self-Superv...	2021-09-26	Code
23	DS-Depth	0.966	No	DS-Depth: Dynamic and Static Depth Estimation vi...	2023-08-14	Code
24	VTDepthB2 (stereo supervision)	0.965	No	-	-	Code
25	X-Distill (M+1024x320)	0.965	No	X-Distill: Improving Self-Supervised Monocular D...	2021-10-24	-
26	DynamicDepth (M+640x192)	0.964	No	Disentangling Object Motion and Occlusion for Un...	2022-03-29	Code
27	SCIPaD(M+640x192)	0.964	No	SCIPaD: Incorporating Spatial Clues into Unsuper...	2024-07-07	Code
28	VTDepthB2 (monocular supervision)	0.964	No	-	-	Code
29	FeatDepth-MS	0.963	No	Feature-metric Loss for Self-supervised Learning...	2020-07-21	Code
30	EPCDepth(S+640x192)	0.963	No	Excavating the Potential Capacity of Self-Superv...	2021-09-26	Code
31	MonoDEVSNet	0.962	No	Monocular Depth Estimation through Virtual-world...	2021-03-22	Code
32	Dyna-DM	0.959	No	Dyna-DM: Dynamic Object-aware Self-supervised Mo...	2022-06-08	Code
33	SharinGAN	0.954	No	SharinGAN: Combining Synthetic and Real Data for...	2020-06-07	Code

#1SPIdepthSOTA
0.973
Delta < 1.25^2· 2024-04-18
SPIdepth: Strengthened Pose Information for Self-supervised Monocular Depth Estimation Code
#2Jasmine
0.972
Delta < 1.25^2· 2025-03-20
Jasmine: Harnessing Diffusion Prior for Self-supervised Depth Estimation
#3SCIPaD
0.97
Delta < 1.25^2· 2024-07-07
SCIPaD: Incorporating Spatial Clues into Unsupervised Pose-Depth Joint Learning Code
#4NimbleD-LiteMono-8M
0.97
Delta < 1.25^2· 2024-08-26
NimbleD: Enhancing Self-supervised Monocular Depth Estimation with Pseudo-labels and Large-scale Video Pre-training Code
#5ProDepth
0.969
Delta < 1.25^2· 2024-07-12
ProDepth: Boosting Self-Supervised Multi-Frame Monocular Depth with Probabilistic Fusion Code
#6DCPI-Depth (M+1024x320)
0.969
Delta < 1.25^2· 2024-05-27
DCPI-Depth: Explicitly Infusing Dense Correspondence Prior to Unsupervised Monocular Depth Estimation
#7MonoViT(MS+1024x320)SOTA
0.969
Delta < 1.25^2· 2022-08-06
MonoViT: Self-Supervised Monocular Depth Estimation with a Vision Transformer Code
#8NimbleD-LiteMono
0.969
Delta < 1.25^2· 2024-08-26
NimbleD: Enhancing Self-supervised Monocular Depth Estimation with Pseudo-labels and Large-scale Video Pre-training Code
#9Nimbled-SwiftDepth
0.969
Delta < 1.25^2· 2024-08-26
NimbleD: Enhancing Self-supervised Monocular Depth Estimation with Pseudo-labels and Large-scale Video Pre-training Code
#10CREMono(M + 1024x320 + Res50)
0.969
Delta < 1.25^2
No paper
#11PlaneDepth (S + 1280x384)
0.968
Delta < 1.25^2· 2022-10-04
PlaneDepth: Self-supervised Depth Estimation via Orthogonal Planes Code
#12Manydepth2(M+640x192)
0.968
Delta < 1.25^2· 2023-12-23
Manydepth2: Motion-Aware Self-Supervised Multi-Frame Monocular Depth Estimation in Dynamic Scenes Code
#13Manydepth2-NF(M+640x192)
0.968
Delta < 1.25^2· 2023-12-23
Manydepth2: Motion-Aware Self-Supervised Multi-Frame Monocular Depth Estimation in Dynamic Scenes Code
#14DIFFNet (MS+1024x320)SOTA
0.968
Delta < 1.25^2· 2021-10-18
Self-Supervised Monocular Depth Estimation with Internal Feature Fusion Code
#15Nimbled-MD2-R50
0.968
Delta < 1.25^2· 2024-08-26
NimbleD: Enhancing Self-supervised Monocular Depth Estimation with Pseudo-labels and Large-scale Video Pre-training Code
#16Nimbled-SwiftDepth-S
0.968
Delta < 1.25^2· 2024-08-26
NimbleD: Enhancing Self-supervised Monocular Depth Estimation with Pseudo-labels and Large-scale Video Pre-training Code
#17DCPI-Depth (M+640x192)
0.967
Delta < 1.25^2· 2024-05-27
DCPI-Depth: Explicitly Infusing Dense Correspondence Prior to Unsupervised Monocular Depth Estimation
#18TransDSSL
0.967
Delta < 1.25^2
No paperCode
#19ProDepth(M+640x192)
0.967
Delta < 1.25^2· 2024-07-12
ProDepth: Boosting Self-Supervised Multi-Frame Monocular Depth with Probabilistic Fusion Code
#20NimbleD-LiteMono-S
0.967
Delta < 1.25^2· 2024-08-26
NimbleD: Enhancing Self-supervised Monocular Depth Estimation with Pseudo-labels and Large-scale Video Pre-training Code
#21Nimbled-MD2-R18
0.967
Delta < 1.25^2· 2024-08-26
NimbleD: Enhancing Self-supervised Monocular Depth Estimation with Pseudo-labels and Large-scale Video Pre-training Code
#22EPCDepth(S+1024x320)SOTA
0.966
Delta < 1.25^2· 2021-09-26
Excavating the Potential Capacity of Self-Supervised Monocular Depth Estimation Code
#23DS-Depth
0.966
Delta < 1.25^2· 2023-08-14
DS-Depth: Dynamic and Static Depth Estimation via a Fusion Cost Volume Code
#24VTDepthB2 (stereo supervision)
0.965
Delta < 1.25^2
No paperCode
#25X-Distill (M+1024x320)
0.965
Delta < 1.25^2· 2021-10-24
X-Distill: Improving Self-Supervised Monocular Depth via Cross-Task Distillation
#26DynamicDepth (M+640x192)
0.964
Delta < 1.25^2· 2022-03-29
Disentangling Object Motion and Occlusion for Unsupervised Multi-frame Monocular Depth Code
#27SCIPaD(M+640x192)
0.964
Delta < 1.25^2· 2024-07-07
SCIPaD: Incorporating Spatial Clues into Unsupervised Pose-Depth Joint Learning Code
#28VTDepthB2 (monocular supervision)
0.964
Delta < 1.25^2
No paperCode
#29FeatDepth-MSSOTA
0.963
Delta < 1.25^2· 2020-07-21
Feature-metric Loss for Self-supervised Learning of Depth and Egomotion Code
#30EPCDepth(S+640x192)
0.963
Delta < 1.25^2· 2021-09-26
Excavating the Potential Capacity of Self-Supervised Monocular Depth Estimation Code
#31MonoDEVSNet
0.962
Delta < 1.25^2· 2021-03-22
Monocular Depth Estimation through Virtual-world Supervision and Real-world SfM Self-Supervision Code
#32Dyna-DM
0.959
Delta < 1.25^2· 2022-06-08
Dyna-DM: Dynamic Object-aware Self-supervised Monocular Depth Maps Code
#33SharinGANSOTA
0.954
Delta < 1.25^2· 2020-06-07
SharinGAN: Combining Synthetic and Real Data for Unsupervised Geometry Estimation Code