Monocular Depth Estimation on KITTI Eigen split unsupervised

110개 결과 · ⬇ CSV · JSON

absolute relative error 낮을수록 좋음

Rank	Model	absolute relative error	RMSE	RMSE log	Sq Rel	Delta < 1.25	Delta < 1.25^2	Delta < 1.25^3	Resolution	Paper	Code	Year
1	SPIdepth	0.071	3.662	0.153	0.531	0.94	0.973	0.985	1024x320	SPIdepth: Strengthened Pose Information for Self-supervised Monocular Depth Estimation	Lavreniuk/SPIdepth	2024
2	PlaneDepth (S + 1280x384)	0.084	3.981	0.169	0.549	0.911	0.968	0.984	1280x384	PlaneDepth: Self-supervised Depth Estimation via Orthogonal Planes	svip-lab/planedepth	2022
3	ProDepth	0.086	4.139	0.166	0.629	0.918	0.969	0.984	640x192	ProDepth: Boosting Self-Supervised Multi-Frame Monocular Depth with Probabilistic Fusion	sungmin-woo/ProDepth	2024
4	Jasmine	0.09	3.944	0.161	0.581	0.919	0.972	0.986	1024x320	Jasmine: Harnessing Diffusion Prior for Self-supervised Depth Estimation		2025
4	SCIPaD	0.090	4.056	0.166	0.650	0.918	0.970	0.985	640x192	SCIPaD: Incorporating Spatial Clues into Unsupervised Pose-Depth Joint Learning	fengyi233/SCIPaD	2024
4	DCPI-Depth (M+1024x320)	0.090	4.113	0.167	0.655	0.914	0.969	0.985	–	DCPI-Depth: Explicitly Infusing Dense Correspondence Prior to Unsupervised Monocular Depth Estimation		2024
7	EPCDepth(S+1024x320)	0.091	4.207	0.176	0.646	0.901	0.966	0.983	1024x320	Excavating the Potential Capacity of Self-Supervised Monocular Depth Estimation	prstrive/EPCDepth	2021
7	Manydepth2(M+640x192)	0.091	4.232	0.170	0.649	0.909	0.968	0.984	640x192	Manydepth2: Motion-Aware Self-Supervised Multi-Frame Monocular Depth Estimation in Dynamic Scenes	kaichen-z/rad	2023
9	NimbleD-LiteMono-8M	0.092	4.194	0.165	0.646	0.910	0.970	0.986	640x192	NimbleD: Enhancing Self-supervised Monocular Depth Estimation with Pseudo-labels and Large-scale Video Pre-training	xapaxca/nimbled	2024
10	MonoViT(MS+1024x320)	0.093	4.202	0.169	0.671	0.912	0.969	0.985	1024x320	MonoViT: Self-Supervised Monocular Depth Estimation with a Vision Transformer	zxcqlf/monovit	2022
11	Manydepth2-NF(M+640x192)	0.094	4.246	0.170	0.676	0.909	0.968	0.985	640x192	Manydepth2: Motion-Aware Self-Supervised Multi-Frame Monocular Depth Estimation in Dynamic Scenes	kaichen-z/rad	2023
11	DIFFNet (MS+1024x320)	0.094	4.250	0.172	0.678	0.911	0.968	0.984	–	Self-Supervised Monocular Depth Estimation with Internal Feature Fusion	brandleyzhou/diffnet	2021
13	DCPI-Depth (M+640x192)	0.095	4.274	0.170	0.662	0.902	0.967	0.985	–	DCPI-Depth: Explicitly Infusing Dense Correspondence Prior to Unsupervised Monocular Depth Estimation		2024
13	TransDSSL	0.095	4.321	0.172	0.711	0.906	0.967	0.984	–	TransDSSL: Transformer based Depth Estimation via Self-Supervised Learning	sejong-rcv/2021.Paper.TransDSSL	2022
13	DS-Depth	0.095	4.329	0.173	0.698	0.905	0.966	0.984	–	DS-Depth: Dynamic and Static Depth Estimation via a Fusion Cost Volume	xingy038/ds-depth	2023
13	ProDepth(M+640x192)	0.095	4.345	0.172	0.693	0.902	0.967	0.985	640x192	ProDepth: Boosting Self-Supervised Multi-Frame Monocular Depth with Probabilistic Fusion	sungmin-woo/ProDepth	2024
17	CADepth-Net (MS+1024x320)	0.096	4.264	0.173	0.694	–	–	–	–	Channel-Wise Attention-Based Network for Self-Supervised Monocular Depth Estimation	kamiLight/CADepth-master	2021
17	NimbleD-LiteMono	0.096	4.304	0.171	0.684	0.903	0.969	0.986	640x192	NimbleD: Enhancing Self-supervised Monocular Depth Estimation with Pseudo-labels and Large-scale Video Pre-training	xapaxca/nimbled	2024
17	Nimbled-SwiftDepth	0.096	4.333	0.171	0.697	0.905	0.969	0.986	640x192	NimbleD: Enhancing Self-supervised Monocular Depth Estimation with Pseudo-labels and Large-scale Video Pre-training	xapaxca/nimbled	2024
17	DynamicDepth (M+640x192)	0.096	4.458	0.175	0.720	0.897	0.964	0.984	–	Disentangling Object Motion and Occlusion for Unsupervised Multi-frame Monocular Depth	AutoAILab/DynamicDepth	2022