Monocular Depth Estimation On Kitti Eigen 1

评估指标

Delta u003c 1.25

Delta u003c 1.25^2

Delta u003c 1.25^3

Mono

RMSE

RMSE log

Sq Rel

absolute relative error

评测结果

各个模型在此基准测试上的表现结果

模型名称	Delta u003c 1.25	Delta u003c 1.25^2	Delta u003c 1.25^3	Mono	RMSE	RMSE log	Sq Rel	absolute relative error	Paper Title	Repository
TransDSSL	0.906	0.967	0.984	O	4.321	0.172	0.711	0.095	TransDSSL: Transformer based Depth Estimation via Self-Supervised Learning
pc4consistentdepth	-	-	-	-	-	-	-	0.113	Pose Constraints for Consistent Self-supervised Monocular Depth and Ego-motion
GCNDepth	-	-	-	-	4.494	0.181	0.720	0.104	GCNDepth: Self-supervised Monocular Depth Estimation based on Graph Convolutional Network
DCPI-Depth (M+832x256+SC-V3)	-	-	-	-	4.496	-	0.679	0.109	DCPI-Depth: Explicitly Infusing Dense Correspondence Prior to Unsupervised Monocular Depth Estimation	-
Manydepth2(M+640x192)	0.909	0.968	0.984	O	4.232	0.170	0.649	0.091	Manydepth2: Motion-Aware Self-Supervised Multi-Frame Monocular Depth Estimation in Dynamic Scenes
HR-Depth-MS-1024X320	-	-	-	-	-	-	-	0.101	HR-Depth: High Resolution Self-Supervised Monocular Depth Estimation
Lite-HR-Depth-T-1280x384	-	-	-	-	-	-	-	0.104	HR-Depth: High Resolution Self-Supervised Monocular Depth Estimation
SCIPaD(M+640x192)	0.897	0.964	0.983	O	4.391	0.175	0.732	0.098	SCIPaD: Incorporating Spatial Clues into Unsupervised Pose-Depth Joint Learning	-
DS-Depth	0.905	0.966	0.984	-	4.329	0.173	0.698	0.095	DS-Depth: Dynamic and Static Depth Estimation via a Fusion Cost Volume
PackNet-SfM M	-	-	-	-	-	-	-	0.107	3D Packing for Self-Supervised Monocular Depth Estimation
Nimbled-SwiftDepth-S	0.901	0.968	0.985	O	4.401	0.174	0.733	0.098	NimbleD: Enhancing Self-supervised Monocular Depth Estimation with Pseudo-labels and Large-scale Video Pre-training
Occlusion_mask_640x192	-	-	-	-	-	-	-	0.113	Improving Self-Supervised Single View Depth Estimation by Masking Occlusion
VTDepthB2 (stereo supervision)	0.904	0.965	0.983	-	4.439	0.178	0.743	0.099	Exploring Efficiency of Vision Transformers for Self-Supervised Monocular Depth Estimation	-
SCIPaD	0.918	0.970	0.985	-	4.056	0.166	0.650	0.090	SCIPaD: Incorporating Spatial Clues into Unsupervised Pose-Depth Joint Learning	-
HR-Depth-M-640x192	-	-	-	-	-	-	-	0.109	HR-Depth: High Resolution Self-Supervised Monocular Depth Estimation
MonoFormer	-	-	-	O	-	-	-	0.104	Deep Digging into the Generalization of Self-Supervised Monocular Depth Estimation
Monodepth S	-	-	-	-	-	-	-	0.133	Unsupervised Monocular Depth Estimation with Left-Right Consistency
Nimbled-MD2-R18	0.898	0.967	0.985	O	4.440	0.175	0.739	0.100	NimbleD: Enhancing Self-supervised Monocular Depth Estimation with Pseudo-labels and Large-scale Video Pre-training
Monodepth2 M	-	-	-	-	-	-	-	0.115	Digging Into Self-Supervised Monocular Depth Estimation
EPCDepth(S+1024x320)	0.901	0.966	0.983	X	4.207	0.176	0.646	0.091	Excavating the Potential Capacity of Self-Supervised Monocular Depth Estimation

0 of 54 row(s) selected.