High-res multi-view results

This table lists the benchmark results for the high-res multi-view scenario. The following metrics are evaluated:

Accuracy [%]: The fraction of the reconstruction which is closer to the ground truth than the evaluation threshold distance (*). Larger is better.
Completeness [%]: The fraction of the ground truth which is closer to the reconstruction than the evaluation threshold distance (*). Larger is better.
F₁ score [%]: The harmonic mean of accuracy and completeness, used to rank methods based on both metrics. Larger is better.
Time [s]: The runtime of the method. Smaller is better.

(*) For exact definitions, detailing how potentially incomplete ground truth is taken into account, see our paper.

The datasets are grouped into different categories, and result averages are computed for a category and method if results of the method are available for all datasets within the category. Note that the category "all" includes both the high-res multi-view and the low-res many-view scenarios.

Methods with suffix _ROB may participate in the Robust Vision Challenge.

Click a dataset result cell to show a visualization of the reconstruction. For training datasets, ground truth and accuracy / completeness visualizations are also available. The visualizations may not work with mobile browsers.

Method	all	high-res multi-view	indoor	outdoor	courty.	delive.	electro	facade	kicker	meadow	office	pipes	playgr.	relief	relief.	terrace	terrai.

SF-MVS		77.13 9	81.70 9	71.79 3	79.32 1	80.76 5	82.96 2	57.49 11	82.82 10	64.79 5	83.69 11	84.46 6	62.83 14	77.13 27	75.21 33	83.35 7	87.85 17

SED-MVS		79.21 1	84.74 2	72.75 1	78.66 2	81.96 2	82.94 4	58.16 4	86.82 2	67.66 1	88.29 4	86.15 3	66.42 2	80.41 4	78.65 6	82.65 17	90.92 2

DVP-MVS		78.77 2	84.89 1	71.62 4	78.04 3	82.05 1	81.64 11	57.96 7	87.53 1	66.64 2	88.77 1	86.31 1	63.16 11	79.99 7	78.56 7	82.31 24	91.03 1
Zhenlong Yuan, Jinguo Luo, Fei Shen, Zhaoxin Li, Cong Liu, Tianlu Mao, Zhaoqi Wang: DVP-MVS: Synergize Depth-Edge and Visibility Prior for Multi-View Stereo. AAAI2025
APDe-MVS		77.58 6	82.93 7	71.35 6	77.86 4	80.55 6	83.38 1	57.61 10	85.57 5	61.11 22	86.10 8	82.98 9	64.76 5	78.29 15	76.78 22	83.40 6	90.20 5
Zhaojie Zeng, Yuesong Wang, Tao Guan: Matching Ambiguity-Resilient Multi-View Stereo via Adaptive Patch Deformation. Pattern Recognition
DPE-MVS		78.11 4	83.84 4	71.42 5	77.82 5	81.32 4	82.92 5	57.81 9	84.04 9	63.19 12	88.63 2	86.00 4	64.52 6	78.71 11	77.63 15	82.26 25	90.57 3
Kehua Chen, Zhenlong Yuan, Tianlu Mao, Zhaoqi Wang: Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization. AAAI2025
ME-MVS		77.51 7	83.11 6	70.98 9	77.52 6	80.15 9	82.81 7	57.37 13	84.71 8	59.56 28	87.31 6	85.76 5	65.28 4	78.00 18	76.77 23	83.31 8	89.06 9

SteuartSystems-ACMMP		75.23 14	79.60 16	70.13 14	76.81 7	78.92 13	81.74 9	57.99 6	75.30 30	59.83 27	75.69 19	78.45 25	60.48 30	80.58 2	79.77 2	83.94 3	88.52 11
Skip Steuart: Steuart Systems R&D Blog.
DVP-MVS++		78.76 3	84.44 3	72.14 2	76.63 8	81.93 3	82.92 5	58.10 5	85.86 4	66.53 3	87.86 5	86.16 2	66.45 1	80.46 3	78.53 9	82.19 29	90.29 4

HPM-MVS++		76.01 11	80.47 13	70.81 10	76.60 9	74.96 37	80.18 18	58.36 2	81.96 11	63.50 11	78.80 15	82.53 12	64.40 7	78.74 10	78.84 5	81.81 35	87.46 19

MCST-MVS		73.67 25	77.39 27	69.33 19	76.26 10	78.19 18	78.77 27	54.54 31	75.33 28	59.99 26	67.96 33	79.23 23	62.43 17	78.00 18	75.71 31	84.02 2	87.30 20

CNVR-MVS		75.62 13	79.91 15	70.61 11	75.76 11	78.82 15	81.66 10	57.12 14	79.77 17	63.04 13	70.69 26	81.15 17	62.99 12	80.23 5	79.54 3	83.11 9	89.16 8

NCCC		74.27 20	77.83 25	70.13 14	75.70 12	77.41 24	80.51 16	57.09 15	78.25 21	62.28 18	65.54 38	78.26 26	62.18 19	79.13 8	78.51 10	83.01 11	87.68 18

CSCG		74.68 17	79.22 17	69.40 18	75.69 13	80.01 10	79.12 25	52.83 42	79.34 18	63.99 9	70.49 27	82.02 13	60.35 33	77.48 25	77.22 19	84.38 1	87.97 16

DPM-MVS		72.80 27	75.90 31	69.19 21	75.51 14	77.68 22	81.62 12	54.83 27	75.96 26	62.06 19	63.96 50	76.58 32	58.55 41	76.66 34	76.77 23	82.60 20	83.68 41

TPM-MVS					75.48 15	76.70 31	79.31 22			62.34 17	64.71 43	77.88 29	56.94 55			81.88 33	83.68 41
Ray Leroy Khuboni and Hongjun Xu: Textureless Resilient Propagation Matching in Multiple View Stereosis (TPM-MVS). SATNAC 2025
SMA-MVS		77.32 8	82.51 8	71.26 7	75.43 16	80.19 8	82.22 8	58.26 3	84.83 7	64.36 7	78.19 16	83.46 7	63.61 9	81.00 1	80.28 1	83.66 4	89.62 6
Yufeng Yin; Xiaoyan Liu; Zichao Zhang: SMA-MVS: Segmentation-Guided Multi-Scale Anchor Deformation Patch Multi-View Stereo. IEEE Transactions on Circuits and Systems for Video Technology
APD-MVS		75.80 12	80.90 12	69.86 16	75.42 17	78.48 17	81.43 14	57.44 12	80.45 15	59.32 29	85.28 9	80.82 19	63.96 8	76.89 29	76.08 29	81.58 40	88.30 13
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
MSP-MVS		77.82 5	83.46 5	71.24 8	75.26 18	80.22 7	82.95 3	57.85 8	85.90 3	64.79 5	88.54 3	83.43 8	66.24 3	78.21 17	78.56 7	80.34 47	89.39 7
Zhenlong Yuan, Cong Liu, Fei Shen, Zhaoxin Li, Jingguo luo, Tianlu Mao and Zhaoqi Wang: MSP-MVS: Multi-granularity Segmentation Prior Guided Multi-View Stereo. AAAI2025
train_agg		73.89 22	78.25 23	68.80 24	75.25 19	72.27 52	79.75 19	56.05 21	74.87 33	58.97 30	81.83 12	79.76 22	61.05 26	77.39 26	76.01 30	81.71 38	85.61 31

TSAR-MVS + MP.		75.22 15	80.06 14	69.56 17	74.61 20	72.74 50	80.59 15	55.70 24	80.80 14	62.65 15	86.25 7	82.92 10	62.07 20	76.89 29	75.66 32	81.77 37	85.19 34
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
SD-MVS		74.43 18	78.87 19	69.26 20	74.39 21	73.70 46	79.06 26	55.24 26	81.04 13	62.71 14	80.18 13	82.61 11	61.70 22	75.43 41	73.92 44	82.44 23	85.22 33
Zhenlong Yuan, Jiakai Cao, Zhaoxin Li, Hao Jiang and Zhaoqi Wang: SD-MVS: Segmentation-driven Deformation Multi-View Stereo with Spherical Refinement and EM optimization. AAAI2024
ACMMP_NAP		76.15 10	81.17 10	70.30 12	74.09 22	79.47 11	81.59 13	57.09 15	81.38 12	63.89 10	79.02 14	80.48 20	62.24 18	80.05 6	79.12 4	82.94 12	88.64 10

HFP-MVS		74.87 16	78.86 21	70.21 13	73.99 23	77.91 19	80.36 17	56.63 17	78.41 20	64.27 8	74.54 21	77.75 30	62.96 13	78.70 12	77.82 13	83.02 10	86.91 22

AdaColmap		67.89 46	68.85 61	66.77 30	73.73 24	74.30 44	75.28 42	53.58 37	70.24 48	57.59 37	51.19 121	59.19 110	60.74 29	75.33 43	73.72 46	79.69 55	77.96 77

MP-MVS		74.31 19	78.87 19	68.99 22	73.49 25	78.56 16	79.25 24	56.51 18	75.33 28	60.69 24	75.30 20	79.12 24	61.81 21	77.78 22	77.93 12	82.18 31	88.06 15
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
TSAR-MVS + ACMM		72.56 29	79.07 18	64.96 42	73.24 26	73.16 49	78.50 28	48.80 68	79.34 18	55.32 43	85.04 10	81.49 16	58.57 40	75.06 44	73.75 45	75.35 119	85.61 31

CDPH-MVS		71.47 33	75.82 33	66.41 33	72.97 27	77.15 26	78.14 31	54.71 28	69.88 50	53.07 63	70.98 25	74.83 38	56.95 54	76.22 35	76.57 25	82.62 18	85.09 35

PGM-MVS		72.89 26	77.13 28	67.94 26	72.47 28	77.25 25	79.27 23	54.63 30	73.71 37	57.95 36	72.38 24	75.33 36	60.75 28	78.25 16	77.36 18	82.57 21	85.62 30

ACMMPR		73.79 24	78.41 22	68.40 25	72.35 29	77.79 21	79.32 21	56.38 19	77.67 24	58.30 34	74.16 22	76.66 31	61.40 23	78.32 14	77.80 14	82.68 16	86.51 23

OPM-MVS		69.33 38	71.05 47	67.32 28	72.34 30	75.70 34	79.57 20	56.34 20	55.21 89	53.81 60	59.51 80	68.96 59	59.67 35	77.61 24	76.44 27	82.19 29	83.88 40
Ray L. Khuboni, Hongjun Xu: Octagram Propagation Matching for Multi-Scale View Stereopsis (OPM-MVS).
X-MVS		71.18 34	75.66 34	65.96 37	71.71 31	76.96 27	77.26 34	55.88 23	72.75 41	54.48 56	64.39 45	74.47 39	54.19 77	77.84 21	77.37 17	82.21 28	85.85 28

HQP-MVS		70.88 35	75.02 35	66.05 36	71.69 32	74.47 42	77.51 33	53.17 39	72.89 40	54.88 49	70.03 29	70.48 53	57.26 49	76.02 37	75.01 36	81.78 36	86.21 24

MAR-MVS		68.04 45	70.74 49	64.90 43	71.68 33	76.33 33	74.63 45	50.48 56	63.81 58	55.52 42	54.88 101	69.90 55	57.39 48	75.42 42	74.79 38	79.71 52	80.03 58
Zhenyu Xu, Yiguang Liu, Xuelei Shi, Ying Wang, Yunan Zheng: MARMVS: Matching Ambiguity Reduced Multiple View Stereo for Efficient Large Scale Scene Reconstruction. CVPR 2020
mPP-MVS					71.67 34							74.36 42

SR-MVS					71.46 35			54.67 29				81.54 15

MGCNet		72.45 30	77.44 26	66.61 31	71.08 36	77.81 20	76.74 35	49.30 62	73.12 39	61.17 20	73.70 23	78.08 27	58.78 38	76.75 33	76.52 26	82.61 19	86.14 26

CLD-MVS		67.02 50	71.57 44	61.71 52	71.01 37	74.81 39	71.62 53	38.91 180	71.86 44	60.70 23	64.97 42	67.88 68	51.88 104	76.77 32	74.98 37	76.11 107	69.75 141
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
CP-MVS		72.63 28	76.95 29	67.59 27	70.67 38	75.53 35	77.95 32	56.01 22	75.65 27	58.82 31	69.16 31	76.48 33	60.46 31	77.66 23	77.20 20	81.65 39	86.97 21

ACMM	60.30 7	67.58 48	68.82 62	66.13 35	70.59 39	72.01 54	76.54 37	54.26 33	65.64 56	54.78 53	50.35 124	61.72 99	58.74 39	75.79 39	75.03 35	81.88 33	81.17 53
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
XVS					70.49 40	76.96 27	74.36 46			54.48 56		74.47 39				82.24 26

X-MVStestdata					70.49 40	76.96 27	74.36 46			54.48 56		74.47 39				82.24 26

ACMMP		71.57 32	75.84 32	66.59 32	70.30 42	76.85 30	78.46 29	53.95 35	73.52 38	55.56 41	70.13 28	71.36 51	58.55 41	77.00 28	76.23 28	82.71 15	85.81 29
Qingshan Xu, Weihang Kong, Wenbing Tao, Marc Pollefeys: Multi-Scale Geometric Consistency Guided and Planar Prior Assisted Multi-View Stereo. IEEE Transactions on Pattern Analysis and Machine Intelligence
MVS_111021_HR		67.62 47	70.39 51	64.39 45	69.77 43	70.45 61	71.44 55	51.72 48	60.77 66	55.06 46	62.14 63	66.40 78	58.13 44	76.13 36	74.79 38	80.19 49	82.04 50

MSLP-MVS++		68.17 44	70.72 50	65.19 40	69.41 44	70.64 58	74.99 43	45.76 79	70.20 49	60.17 25	56.42 93	73.01 45	61.14 24	72.80 55	70.54 61	79.70 53	81.42 52

LGP-MVS_train		68.87 40	72.03 43	65.18 41	69.33 45	74.03 45	76.67 36	53.88 36	68.46 51	52.05 70	63.21 53	63.89 86	56.31 58	75.99 38	74.43 40	82.83 14	84.18 37

ACMP	61.42 5	68.72 43	71.37 45	65.64 39	69.06 46	74.45 43	75.88 40	53.30 38	68.10 52	55.74 40	61.53 69	62.29 93	56.97 53	74.70 47	74.23 42	82.88 13	84.31 36
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
DeepC-MVS_fast	65.08 3	72.00 31	76.11 30	67.21 29	68.93 47	77.46 23	76.54 37	54.35 32	74.92 32	58.64 33	65.18 40	74.04 44	62.62 15	77.92 20	77.02 21	82.16 32	86.21 24
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
CANet		68.77 41	73.01 38	63.83 46	68.30 48	75.19 36	73.73 49	47.90 69	63.86 57	54.84 52	67.51 35	74.36 42	57.62 45	74.22 49	73.57 48	80.56 45	82.36 47

TSAR-MVS + GP.		69.71 36	73.92 37	64.80 44	68.27 49	70.56 59	71.90 51	50.75 52	71.38 45	57.46 38	68.68 32	75.42 35	60.10 34	73.47 52	73.99 43	80.32 48	83.97 39

3Dnovator+	62.63 4	69.51 37	72.62 40	65.88 38	68.21 50	76.47 32	73.50 50	52.74 43	70.85 46	58.65 32	55.97 95	69.95 54	61.11 25	76.80 31	75.09 34	81.09 43	83.23 45

DeepC-MVS	66.32 2	73.85 23	78.10 24	68.90 23	67.92 51	79.31 12	78.16 30	59.28 1	78.24 22	61.13 21	67.36 36	76.10 34	63.40 10	79.11 9	78.41 11	83.52 5	88.16 14
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
DeepPCF-MVS	66.49 1	74.25 21	80.97 11	66.41 33	67.75 52	78.87 14	75.61 41	54.16 34	84.86 6	58.22 35	77.94 17	81.01 18	62.52 16	78.34 13	77.38 16	80.16 50	88.40 12

EC-MVSNet		67.01 51	70.27 54	63.21 48	67.21 53	70.47 60	69.01 67	46.96 73	59.16 75	53.23 62	64.01 49	69.71 57	60.37 32	74.92 45	71.24 56	82.50 22	82.41 46

UA-Net		58.50 108	64.68 93	51.30 137	66.97 54	67.13 96	53.68 179	45.65 82	49.51 124	31.58 165	62.91 55	68.47 61	35.85 190	68.20 114	67.28 101	74.03 132	69.24 151

PHI-MVS		69.27 39	74.84 36	62.76 51	66.83 55	74.83 38	73.88 48	49.32 61	70.61 47	50.93 74	69.62 30	74.84 37	57.25 50	75.53 40	74.32 41	78.35 70	84.17 38

3Dnovator	60.86 6	66.99 52	70.32 52	63.11 49	66.63 56	74.52 40	71.56 54	45.76 79	67.37 54	55.00 48	54.31 106	68.19 64	58.49 43	73.97 50	73.63 47	81.22 42	80.23 57

EPNet		65.14 62	69.54 58	60.00 67	66.61 57	67.67 86	67.53 76	55.32 25	62.67 62	46.22 95	67.74 34	65.93 80	48.07 127	72.17 58	72.12 50	76.28 103	78.47 70
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
OpenMVS	57.13 9	62.81 77	65.75 84	59.39 71	66.47 58	69.52 63	64.26 114	43.07 139	61.34 65	50.19 77	47.29 142	64.41 85	54.60 74	70.18 83	68.62 82	77.73 74	78.89 66

ACMH	52.42 13	58.24 115	59.56 135	56.70 97	66.34 59	69.59 62	66.71 86	49.12 63	46.08 158	28.90 178	42.67 189	41.20 207	52.60 96	71.39 65	70.28 63	76.51 99	75.72 108
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
MSDG		58.46 110	58.97 141	57.85 88	66.27 60	66.23 105	67.72 73	42.33 143	53.43 97	43.68 107	43.39 180	45.35 184	49.75 116	68.66 103	67.77 91	77.38 85	67.96 156

QAPM		65.27 58	69.49 59	60.35 63	65.43 61	72.20 53	65.69 99	47.23 71	63.46 59	49.14 79	53.56 107	71.04 52	57.01 52	72.60 57	71.41 54	77.62 78	82.14 49

CPTT-MVS		68.76 42	73.01 38	63.81 47	65.42 62	73.66 47	76.39 39	52.08 44	72.61 42	50.33 76	60.73 73	72.65 47	59.43 36	73.32 53	72.12 50	79.19 61	85.99 27

MS-PatchMatch		58.19 117	60.20 122	55.85 104	65.17 63	64.16 126	64.82 106	41.48 154	50.95 114	42.17 116	45.38 163	56.42 122	48.08 126	68.30 109	66.70 111	73.39 145	69.46 149

PCF-MVS	59.98 8	67.32 49	71.04 48	62.97 50	64.77 64	74.49 41	74.78 44	49.54 58	67.44 53	54.39 59	58.35 87	72.81 46	55.79 64	71.54 64	69.24 72	78.57 64	83.41 43
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
EG-PatchMatch MVS		56.98 124	58.24 148	55.50 106	64.66 65	68.62 68	61.48 123	43.63 122	38.44 211	41.44 120	38.05 202	46.18 179	43.95 145	71.71 63	70.61 60	77.87 71	74.08 121

Effi-MVS+-dtu		60.34 93	62.32 105	58.03 83	64.31 66	67.44 91	65.99 93	42.26 144	49.55 122	42.00 119	48.92 132	59.79 108	56.27 59	68.07 118	67.03 103	77.35 86	75.45 111

FC-MVSNet-train		58.40 111	63.15 102	52.85 127	64.29 67	61.84 142	55.98 163	46.47 75	53.06 101	34.96 153	61.95 65	56.37 124	39.49 166	68.67 102	68.36 84	75.92 113	71.81 129

Effi-MVS+		63.28 74	65.96 83	60.17 65	64.26 68	68.06 78	68.78 70	45.71 81	54.08 93	46.64 90	55.92 96	63.13 90	55.94 62	70.38 79	71.43 53	79.68 56	78.70 67

LS3D		60.20 94	61.70 106	58.45 78	64.18 69	67.77 83	67.19 79	48.84 67	61.67 64	41.27 123	45.89 157	51.81 141	54.18 78	68.78 100	66.50 119	75.03 123	69.48 147

ACMH+	53.71 12	59.26 100	60.28 119	58.06 81	64.17 70	68.46 69	67.51 77	50.93 51	52.46 109	35.83 149	40.83 195	45.12 188	52.32 99	69.88 86	69.00 77	77.59 81	76.21 105

GeoE		62.43 80	64.79 92	59.68 70	64.15 71	67.17 95	68.80 69	44.42 98	55.65 88	47.38 84	51.54 118	62.51 91	54.04 80	69.99 85	68.07 86	79.28 59	78.57 68

DELS-MVS		65.87 54	70.30 53	60.71 62	64.05 72	72.68 51	70.90 56	45.43 83	57.49 83	49.05 81	64.43 44	68.66 60	55.11 71	74.31 48	73.02 49	79.70 53	81.51 51
Christian Sormann, Emanuele Santellani, Mattia Rossi, Andreas Kuhn, Friedrich Fraundorfer: DELS-MVS: Deep Epipolar Line Search for Multi-View Stereo. Winter Conference on Applications of Computer Vision (WACV), 2023
sasdasda		65.62 55	72.06 41	58.11 79	63.94 73	71.05 56	64.49 111	43.18 135	74.08 34	47.35 85	64.17 47	71.97 48	51.17 109	71.87 60	70.74 58	78.51 67	80.56 55

canonicalmvs		65.62 55	72.06 41	58.11 79	63.94 73	71.05 56	64.49 111	43.18 135	74.08 34	47.35 85	64.17 47	71.97 48	51.17 109	71.87 60	70.74 58	78.51 67	80.56 55

EIA-MVS		61.53 88	63.79 98	58.89 76	63.82 75	67.61 87	65.35 102	42.15 147	49.98 119	45.66 98	57.47 91	56.62 120	56.59 57	70.91 72	69.15 73	79.78 51	74.80 115

Anonymous202405211			60.60 115		63.44 76	66.71 102	61.00 128	47.23 71	50.62 117		36.85 205	60.63 105	43.03 153	69.17 96	67.72 93	75.41 116	72.54 126

ETV-MVS		63.23 75	66.08 82	59.91 68	63.13 77	68.13 74	67.62 75	44.62 94	53.39 98	46.23 94	58.74 84	58.19 113	57.45 47	73.60 51	71.38 55	80.39 46	79.13 63

viewcassd2359sk11		64.22 63	67.08 67	60.87 59	63.08 78	68.05 80	70.51 61	43.92 113	59.80 69	55.05 47	62.49 61	66.89 72	55.09 72	69.39 92	66.19 126	77.60 79	76.77 94

E3new		64.18 65	67.01 70	60.89 57	63.07 79	68.08 76	70.57 59	43.95 110	59.33 72	54.87 51	61.94 67	66.76 75	55.16 69	69.60 89	66.42 122	77.70 75	76.92 87

E2		64.19 64	67.06 68	60.84 61	63.07 79	68.02 81	70.44 62	43.88 114	59.94 68	55.15 45	62.73 57	66.97 71	55.01 73	69.18 95	65.98 130	77.53 83	76.63 96

E3		64.18 65	67.01 70	60.89 57	63.07 79	68.07 77	70.57 59	43.94 111	59.32 73	54.88 49	61.95 65	66.78 74	55.16 69	69.60 89	66.43 121	77.70 75	76.92 87

E6		64.03 69	66.63 76	60.99 56	63.04 82	68.16 72	70.80 57	44.14 100	57.66 82	54.63 54	60.32 75	66.05 79	55.49 65	70.14 84	67.09 102	77.85 72	76.94 86

E4		64.06 68	66.79 75	60.87 59	63.03 83	68.11 75	70.61 58	44.00 108	58.24 79	54.56 55	61.00 72	66.64 76	55.22 67	69.80 87	66.69 112	77.81 73	77.07 85

casdiffmvs_mvg		65.26 59	69.48 60	60.33 64	62.99 84	69.34 64	69.80 65	45.27 85	63.38 60	51.11 73	65.12 41	69.75 56	53.51 85	71.74 62	68.86 78	79.33 57	78.19 74
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
viewdifsd2359ckpt13		63.83 70	67.03 69	60.10 66	62.56 85	68.92 66	69.73 66	43.49 127	57.96 80	52.16 69	61.09 71	65.39 83	55.20 68	70.36 80	67.48 98	77.48 84	78.00 76

viewdifsd2359ckpt09		65.38 57	68.69 64	61.53 53	62.15 86	71.64 55	71.84 52	47.45 70	58.95 77	51.79 72	61.73 68	65.71 82	57.08 51	72.17 58	70.82 57	78.87 62	79.79 59

viewdifsd2359ckpt07		61.71 84	65.49 86	57.31 91	62.12 87	65.52 111	68.53 71	38.21 190	56.37 85	48.07 83	61.11 70	65.85 81	52.82 94	68.34 108	64.46 153	74.08 131	76.80 91

gg-mvs-nofinetune		49.07 186	52.56 186	45.00 186	61.99 88	59.78 161	53.55 181	41.63 152	31.62 228	12.08 224	29.56 223	53.28 135	29.57 205	66.27 151	64.49 151	71.19 175	62.92 191

SPE-MVS-test		65.18 60	68.70 63	61.07 55	61.92 89	68.06 78	67.09 83	45.18 87	58.47 78	52.02 71	65.76 37	66.44 77	59.24 37	72.71 56	70.05 66	80.98 44	79.40 62

DCV-MVSNet		59.49 96	64.00 97	54.23 113	61.81 90	64.33 124	61.42 124	43.77 116	52.85 106	38.94 137	55.62 98	62.15 97	43.24 152	69.39 92	67.66 95	76.22 105	75.97 106

casdiffmvs		64.09 67	68.13 65	59.37 72	61.81 90	68.32 71	68.48 72	44.45 97	61.95 63	49.12 80	63.04 54	69.67 58	53.83 81	70.46 76	66.06 127	78.55 65	77.43 80
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
CS-MVS		65.88 53	69.71 57	61.41 54	61.76 92	68.14 73	67.65 74	44.00 108	59.14 76	52.69 64	65.19 39	68.13 65	60.90 27	74.74 46	71.58 52	81.46 41	81.04 54

Anonymous20231211		57.71 120	60.79 112	54.13 115	61.68 93	65.81 109	60.81 129	43.70 120	51.97 112	39.67 132	34.82 210	63.59 87	43.31 150	68.55 106	66.63 115	75.59 114	74.13 120

IS_MVSNet		57.95 118	64.26 95	50.60 139	61.62 94	65.25 116	57.18 149	45.42 84	50.79 115	26.49 192	57.81 89	60.05 107	34.51 194	71.24 68	70.20 65	78.36 69	74.44 117

viewmanbaseed2359cas		63.67 71	67.42 66	59.30 74	61.34 95	67.42 92	70.01 64	40.50 167	59.53 70	52.60 65	62.56 60	67.34 70	54.44 76	70.33 81	66.93 107	76.91 93	77.82 79

NR-MVSNet		55.35 141	59.46 136	50.56 140	61.33 96	62.97 134	57.91 145	51.80 46	48.62 139	20.59 207	51.99 116	44.73 194	34.10 197	68.58 104	68.64 81	77.66 77	70.67 138

MVS_111021_LR		63.05 76	66.43 79	59.10 75	61.33 96	63.77 130	65.87 96	43.58 123	60.20 67	53.70 61	62.09 64	62.38 92	55.84 63	70.24 82	68.08 85	74.30 128	78.28 73

viewmacassd2359aftdt		63.43 73	66.95 72	59.32 73	61.27 98	67.48 90	70.15 63	40.54 164	57.82 81	52.27 68	60.49 74	66.81 73	54.58 75	70.67 74	67.39 100	77.08 92	78.02 75

MGCFI-Net		61.46 89	69.72 56	51.83 134	61.00 99	66.16 106	56.50 156	40.73 162	73.98 36	35.18 150	64.23 46	71.42 50	42.45 155	69.22 94	64.01 157	75.09 122	79.03 65

EPP-MVSNet		59.39 99	65.45 87	52.32 131	60.96 100	67.70 85	58.42 142	44.75 92	49.71 121	27.23 186	59.03 81	62.20 96	43.34 149	70.71 73	69.13 74	79.25 60	79.63 61

TransMVSNet (Re)		51.92 167	55.38 166	47.88 171	60.95 101	59.90 160	53.95 176	45.14 88	39.47 205	24.85 197	43.87 175	46.51 175	29.15 206	67.55 128	65.23 142	73.26 150	65.16 181

gm-plane-assit		44.74 207	45.95 215	43.33 194	60.88 102	46.79 222	36.97 231	32.24 221	24.15 236	11.79 225	29.26 224	32.97 230	46.64 132	65.09 166	62.95 170	71.45 172	60.42 203

baseline1		54.48 150	58.69 142	49.57 146	60.63 103	58.29 177	55.70 165	44.95 90	49.20 127	29.62 174	54.77 102	54.75 129	35.29 191	67.15 138	64.08 155	71.21 174	62.58 196

test2506		55.82 136	59.57 134	51.46 135	60.39 104	64.55 122	58.69 140	48.87 65	53.91 94	26.99 187	48.97 130	41.72 206	37.71 176	70.96 70	69.49 69	76.08 108	67.37 161

ECVR-MVS		56.44 131	60.74 113	51.42 136	60.39 104	64.55 122	58.69 140	48.87 65	53.91 94	26.76 189	45.55 162	53.43 134	37.71 176	70.96 70	69.49 69	76.08 108	67.32 163

DI_MVS_pp		61.88 82	65.17 89	58.06 81	60.05 106	65.26 114	66.03 92	44.22 99	55.75 87	46.73 88	54.64 104	68.12 66	54.13 79	69.13 97	66.66 113	77.18 88	76.61 97

PVSNet_Blended_VisFu		63.65 72	66.92 73	59.83 69	60.03 107	73.44 48	66.33 89	48.95 64	52.20 111	50.81 75	56.07 94	60.25 106	53.56 83	73.23 54	70.01 67	79.30 58	83.24 44

UniMVSNet_NR-MVSNet		56.94 126	61.14 109	52.05 133	60.02 108	65.21 117	57.44 147	52.93 41	49.37 125	24.31 200	54.62 105	50.54 147	39.04 168	68.69 101	68.84 79	78.53 66	70.72 134

IterMVS-LS		58.30 114	61.39 108	54.71 109	59.92 109	58.40 174	59.42 134	43.64 121	48.71 136	40.25 130	57.53 90	58.55 112	52.15 101	65.42 164	65.34 139	72.85 153	75.77 107
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
test1111		55.24 142	59.98 127	49.71 145	59.80 110	64.10 127	56.48 157	49.34 60	52.27 110	21.56 205	44.49 170	51.96 140	35.93 189	70.59 75	69.07 75	75.13 121	67.40 159

MVS_Test		62.40 81	66.23 81	57.94 84	59.77 111	64.77 120	66.50 88	41.76 150	57.26 84	49.33 78	62.68 58	67.47 69	53.50 87	68.57 105	66.25 123	76.77 95	76.58 98

FA-MVS(training)		60.00 95	63.14 103	56.33 99	59.50 112	64.30 125	65.15 104	38.75 185	56.20 86	45.77 96	53.08 108	56.45 121	52.10 102	69.04 99	67.67 94	76.69 96	75.27 114

TranMVSNet+NR-MVSNet		55.87 134	60.14 124	50.88 138	59.46 113	63.82 129	57.93 144	52.98 40	48.94 131	20.52 208	52.87 110	47.33 166	36.81 184	69.12 98	69.03 76	77.56 82	69.89 140

Fast-Effi-MVS+		60.36 92	63.35 101	56.87 95	58.70 114	65.86 108	65.08 105	37.11 195	53.00 103	45.36 100	52.12 115	56.07 126	56.27 59	71.28 67	69.42 71	78.71 63	75.69 109

IB-MVS	54.11 11	58.36 113	60.70 114	55.62 105	58.67 115	68.02 81	61.56 121	43.15 137	46.09 157	44.06 106	44.24 172	50.99 146	48.71 121	66.70 144	70.33 62	77.60 79	78.50 69
Christian Sormann, Mattia Rossi, Andreas Kuhn and Friedrich Fraundorfer: IB-MVS: An Iterative Algorithm for Deep Multi-View Stereo based on Binary Decisions. BMVC 2021
Fast-Effi-MVS+-dtu		56.30 132	59.29 138	52.82 128	58.64 116	64.89 118	65.56 100	32.89 218	45.80 160	35.04 152	45.89 157	54.14 131	49.41 117	67.16 137	66.45 120	75.37 118	70.69 136

CNLPA		62.78 78	66.31 80	58.65 77	58.47 117	68.41 70	65.98 94	41.22 158	78.02 23	56.04 39	46.65 145	59.50 109	57.50 46	69.67 88	65.27 141	72.70 159	76.67 95

tpm cat1		53.30 155	53.41 179	53.17 124	58.16 118	59.15 168	63.73 117	38.27 189	50.73 116	46.98 87	45.57 161	44.00 200	49.20 118	55.90 215	54.02 214	62.65 206	64.50 186

v1144		58.88 103	60.16 123	57.39 90	58.03 119	67.26 93	67.14 81	44.46 96	45.17 163	44.33 105	47.81 139	49.92 152	53.20 93	67.77 124	66.62 116	77.15 89	76.58 98

v10		59.17 102	60.60 115	57.50 89	57.95 120	66.73 99	67.09 83	44.11 101	46.85 151	45.42 99	48.18 138	51.07 143	53.63 82	67.84 122	66.59 117	76.79 94	76.92 87

v1192		58.51 107	59.66 130	57.17 92	57.82 121	67.72 84	66.21 91	44.83 91	44.15 171	43.49 108	46.68 144	47.94 157	53.55 84	67.39 131	66.51 118	77.13 90	77.20 83

viewmambaseed2359dif		60.40 91	64.15 96	56.03 101	57.79 122	63.53 132	65.91 95	41.64 151	54.98 90	46.47 91	60.16 77	64.71 84	50.76 111	66.25 152	62.83 172	73.61 143	76.57 100

v144192		58.23 116	59.40 137	56.87 95	57.56 123	66.89 97	65.70 97	45.01 89	44.06 172	42.88 110	46.61 146	48.09 156	53.49 88	66.94 142	65.90 133	76.61 97	77.29 81

v1921920		57.89 119	59.02 140	56.58 98	57.55 124	66.66 103	64.72 108	44.70 93	43.55 176	42.73 111	46.17 154	46.93 170	53.51 85	66.78 143	65.75 135	76.29 102	77.28 82

Vis-MVSNet (Re-imp)		50.37 176	57.73 154	41.80 201	57.53 125	54.35 192	45.70 211	45.24 86	49.80 120	13.43 222	58.23 88	56.42 122	20.11 224	62.96 174	63.36 166	68.76 184	58.96 208

CostFormer		56.57 129	59.13 139	53.60 118	57.52 126	61.12 149	66.94 85	35.95 200	53.44 96	44.68 103	55.87 97	54.44 130	48.21 124	60.37 187	58.33 195	68.27 186	70.33 139

thres400		52.38 161	55.51 164	48.74 157	57.49 127	60.10 159	55.45 168	43.54 124	42.90 183	26.72 190	43.34 182	45.03 192	36.61 185	66.20 154	64.53 150	72.66 160	66.43 167

thres600view7		51.91 168	55.14 168	48.14 167	57.43 128	60.18 157	54.60 174	43.73 118	42.61 187	25.20 195	43.10 185	44.47 197	35.19 192	66.36 148	63.28 167	72.66 160	66.01 174

thres200		52.39 160	55.37 167	48.90 155	57.39 129	60.18 157	55.60 166	43.73 118	42.93 182	27.41 184	43.35 181	45.09 189	36.61 185	66.36 148	63.92 162	72.66 160	65.78 176

baseline2		55.89 133	57.82 151	53.64 117	57.36 130	61.09 150	59.75 133	40.45 168	47.38 149	41.26 124	51.23 120	46.90 171	48.11 125	65.63 161	64.38 154	74.90 124	68.16 155

v8		58.88 103	60.57 117	56.92 94	57.35 131	65.69 110	66.69 87	42.64 141	47.89 146	45.77 96	49.04 129	52.98 136	52.77 95	67.51 129	65.57 136	76.26 104	75.30 113

thres100view900		52.04 165	54.81 172	48.80 156	57.31 132	59.33 165	55.30 170	42.92 140	42.85 184	27.81 182	43.00 186	45.06 190	36.99 182	64.74 167	63.51 164	72.47 163	65.21 180

tfpn200view9		52.53 158	55.51 164	49.06 153	57.31 132	60.24 156	55.42 169	43.77 116	42.85 184	27.81 182	43.00 186	45.06 190	37.32 180	66.38 147	64.54 149	72.71 158	66.54 166

v1240		57.55 121	58.63 144	56.29 100	57.30 134	66.48 104	63.77 116	44.56 95	42.77 186	42.48 113	45.64 160	46.28 177	53.46 89	66.32 150	65.80 134	76.16 106	77.13 84

dmvs_re		52.07 163	55.11 169	48.54 162	57.27 135	51.93 201	57.73 146	43.13 138	43.65 174	26.57 191	44.52 169	50.00 151	36.53 187	66.58 146	62.15 178	69.97 180	66.91 164

CHOSEN 1792x2688		55.85 135	58.01 149	53.33 120	57.26 136	62.82 136	63.29 120	41.55 153	46.65 153	38.34 138	34.55 211	53.50 132	52.43 98	67.10 139	67.56 97	67.13 190	73.92 123

PVSNet_BlendedMVS		61.63 86	64.82 90	57.91 86	57.21 137	67.55 88	63.47 118	46.08 77	54.72 91	52.46 66	58.59 85	60.73 102	51.82 105	70.46 76	65.20 143	76.44 100	76.50 102

PVSNet_Blended		61.63 86	64.82 90	57.91 86	57.21 137	67.55 88	63.47 118	46.08 77	54.72 91	52.46 66	58.59 85	60.73 102	51.82 105	70.46 76	65.20 143	76.44 100	76.50 102

viewdifsd2359ckpt11		59.45 97	63.57 99	54.65 111	57.17 139	62.71 138	64.67 109	38.99 177	52.96 104	42.12 117	58.97 82	62.23 94	51.18 107	67.35 132	63.98 158	73.75 136	76.80 91

viewmsd2359difaftdt		59.45 97	63.57 99	54.65 111	57.17 139	62.71 138	64.67 109	38.99 177	52.96 104	42.12 117	58.97 82	62.22 95	51.18 107	67.35 132	63.98 158	73.75 136	76.80 91

v2v482		58.69 106	60.12 126	57.03 93	57.16 141	66.05 107	67.17 80	43.52 125	46.33 155	45.19 101	49.46 128	51.02 144	52.51 97	67.30 134	66.03 129	76.61 97	74.62 116

tfpnnormal		50.16 178	52.19 190	47.78 173	56.86 142	58.37 175	54.15 175	44.01 107	38.35 213	25.94 193	36.10 206	37.89 220	34.50 195	65.93 156	63.42 165	71.26 173	65.28 179

diffmvs_AUTHOR		61.79 83	66.80 74	55.95 102	56.69 143	63.92 128	67.27 78	41.28 156	59.32 73	46.43 92	63.31 52	68.30 63	50.56 112	68.30 109	66.06 127	73.48 144	78.36 71

diffmvs		61.64 85	66.55 78	55.90 103	56.63 144	63.71 131	67.13 82	41.27 157	59.49 71	46.70 89	63.93 51	68.01 67	50.46 113	67.30 134	65.51 137	73.24 151	77.87 78
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
HyFIR lowres test		56.87 127	58.60 145	54.84 108	56.62 145	69.27 65	64.77 107	42.21 145	45.66 161	37.50 144	33.08 214	57.47 118	53.33 90	65.46 163	67.94 87	74.60 125	71.35 131

v7n		55.67 137	57.46 156	53.59 119	56.06 146	65.29 113	61.06 127	43.26 134	40.17 202	37.99 141	40.79 196	45.27 187	47.09 131	67.67 126	66.21 124	76.08 108	76.82 90

dps		50.42 175	51.20 196	49.51 147	55.88 147	56.07 188	53.73 177	38.89 181	43.66 173	40.36 129	45.66 159	37.63 222	45.23 140	59.05 190	56.18 199	62.94 205	60.16 204

CANet_DTU		58.88 103	64.68 93	52.12 132	55.77 148	66.75 98	63.92 115	37.04 196	53.32 99	37.45 145	59.81 78	61.81 98	44.43 144	68.25 111	67.47 99	74.12 130	75.33 112

WR-MVS		48.78 188	55.06 170	41.45 202	55.50 149	60.40 155	43.77 219	49.99 57	41.92 190	8.10 235	45.24 166	45.56 182	17.47 225	61.57 182	64.60 148	73.85 134	66.14 173

UniMVSNet (Re)		55.15 146	60.39 118	49.03 154	55.31 150	64.59 121	55.77 164	50.63 53	48.66 138	20.95 206	51.47 119	50.40 148	34.41 196	67.81 123	67.89 88	77.11 91	71.88 128

test-LLR		49.28 182	50.29 200	48.10 168	55.26 151	47.16 217	49.52 191	43.48 129	39.22 206	31.98 161	43.65 178	47.93 158	41.29 161	56.80 206	55.36 205	67.08 191	61.94 197

test0.0.03 1		43.15 212	46.95 214	38.72 211	55.26 151	50.56 205	42.48 222	43.48 129	38.16 215	15.11 217	35.07 209	44.69 195	16.47 227	55.95 214	54.34 213	59.54 213	49.87 227

GA-MVS		55.67 137	58.33 146	52.58 130	55.23 153	63.09 133	61.08 126	40.15 173	42.95 181	37.02 147	52.61 112	47.68 161	47.51 129	65.92 157	65.35 138	74.49 127	70.68 137

DTE-MVSNet		48.03 194	53.28 181	41.91 200	54.64 154	57.50 183	44.63 218	51.66 49	41.02 196	7.97 236	46.26 151	40.90 208	20.24 223	60.45 186	62.89 171	72.33 166	63.97 187

pmmvs4		54.66 149	56.07 160	53.00 125	54.63 155	57.08 185	60.43 131	44.10 102	51.69 113	40.55 127	46.55 149	44.79 193	45.95 137	62.54 176	63.66 163	72.36 165	66.20 171

DU-MVS		55.41 140	59.59 131	50.54 141	54.60 156	62.97 134	57.44 147	51.80 46	48.62 139	24.31 200	51.99 116	47.00 169	39.04 168	68.11 116	67.75 92	76.03 112	70.72 134

Baseline_NR-MVSNet		53.50 153	57.89 150	48.37 165	54.60 156	59.25 167	56.10 159	51.84 45	49.32 126	17.92 215	45.38 163	47.68 161	36.93 183	68.11 116	65.95 131	72.84 154	69.57 145

v148		55.58 139	57.61 155	53.20 122	54.59 158	61.86 141	61.18 125	38.70 186	44.30 170	42.25 115	47.53 140	50.24 150	48.73 120	65.15 165	62.61 176	73.79 135	71.61 130

PEN-MVS		49.21 184	54.32 174	43.24 196	54.33 159	59.26 166	47.04 204	51.37 50	41.67 192	9.97 230	46.22 152	41.80 205	22.97 221	60.52 185	64.03 156	73.73 138	66.75 165

pm-mvs1		51.02 172	55.55 163	45.73 181	54.16 160	58.52 172	50.92 187	42.56 142	40.32 200	25.67 194	43.66 177	50.34 149	30.06 204	65.85 158	63.97 160	70.99 176	66.21 170

EPNet_dtu		52.05 164	58.26 147	44.81 187	54.10 161	50.09 208	52.01 185	40.82 161	53.03 102	27.41 184	54.90 100	57.96 117	26.72 212	62.97 173	62.70 175	67.78 188	66.19 172
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
tpm		48.82 187	51.27 195	45.96 180	54.10 161	47.35 216	56.05 160	30.23 222	46.70 152	43.21 109	52.54 113	47.55 164	37.28 181	54.11 220	50.50 223	54.90 223	60.12 205

CDS-MVSNet		52.42 159	57.06 158	47.02 176	53.92 163	58.30 176	55.50 167	46.47 75	42.52 188	29.38 176	49.50 127	52.85 137	28.49 210	66.70 144	66.89 108	68.34 185	62.63 195
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
test20.03		40.38 221	44.20 222	35.92 219	53.73 164	49.05 209	38.54 228	43.49 127	32.55 225	9.54 231	27.88 227	39.12 216	12.24 232	56.28 211	54.69 210	57.96 218	49.83 228

Vis-MVSNet		58.48 109	65.70 85	50.06 144	53.40 165	67.20 94	60.24 132	43.32 132	48.83 133	30.23 171	62.38 62	61.61 100	40.35 164	71.03 69	69.77 68	72.82 155	79.11 64
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
PLC	52.09 14	59.21 101	62.47 104	55.41 107	53.24 166	64.84 119	64.47 113	40.41 170	65.92 55	44.53 104	46.19 153	55.69 127	55.33 66	68.24 113	65.30 140	74.50 126	71.09 132
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
thisisatest0530		56.68 128	59.68 129	53.19 123	52.97 167	60.96 152	59.41 135	40.51 165	48.26 142	41.06 125	52.67 111	46.30 176	49.78 114	67.66 127	67.83 89	75.39 117	74.07 122

OMC-MVS		65.16 61	71.35 46	57.94 84	52.95 168	68.82 67	69.00 68	38.28 188	79.89 16	55.20 44	62.76 56	68.31 62	56.14 61	71.30 66	68.70 80	76.06 111	79.67 60

tpmrst		48.08 192	49.88 205	45.98 179	52.71 169	48.11 214	53.62 180	33.70 213	48.70 137	39.74 131	48.96 131	46.23 178	40.29 165	50.14 229	49.28 225	55.80 220	57.71 211

tttt0517		56.53 130	59.59 131	52.95 126	52.66 170	60.99 151	59.21 137	40.51 165	47.89 146	40.40 128	52.50 114	46.04 180	49.78 114	67.75 125	67.83 89	75.15 120	74.17 119

GBi-Net		55.20 143	60.25 120	49.31 148	52.42 171	61.44 144	57.03 150	44.04 104	49.18 128	30.47 167	48.28 134	58.19 113	38.22 171	68.05 119	66.96 104	73.69 139	69.65 142

test1		55.20 143	60.25 120	49.31 148	52.42 171	61.44 144	57.03 150	44.04 104	49.18 128	30.47 167	48.28 134	58.19 113	38.22 171	68.05 119	66.96 104	73.69 139	69.65 142

FMVSNet2		55.04 147	59.95 128	49.31 148	52.42 171	61.44 144	57.03 150	44.08 103	49.55 122	30.40 170	46.89 143	58.84 111	38.22 171	67.07 140	66.21 124	73.69 139	69.65 142

FMVSNet3		54.78 148	59.58 133	49.17 151	52.37 174	61.31 148	56.72 155	44.04 104	49.18 128	30.47 167	48.28 134	58.19 113	38.09 174	65.48 162	65.20 143	73.31 148	69.45 150

testgi		38.71 224	43.64 224	32.95 224	52.30 175	48.63 213	35.59 234	35.05 203	31.58 229	9.03 234	30.29 219	40.75 210	11.19 238	55.30 216	53.47 219	54.53 225	45.48 231

Anonymous20231206		42.28 213	45.89 216	38.07 213	51.96 176	48.98 210	43.66 220	38.81 184	38.74 210	14.32 220	26.74 228	40.90 208	20.94 222	56.64 209	54.67 211	58.71 214	54.59 216

pmmvs6		48.35 190	51.64 192	44.51 189	51.92 177	57.94 181	49.44 193	42.17 146	34.45 221	24.62 199	28.87 226	46.90 171	29.07 208	64.60 168	63.08 168	69.83 181	65.68 177

PatchmatchNet		49.92 180	51.29 194	48.32 166	51.83 178	51.86 202	53.38 182	37.63 194	47.90 145	40.83 126	48.54 133	45.30 185	45.19 141	56.86 205	53.99 216	61.08 211	54.57 217
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
WR-MVS_H		47.65 195	53.67 177	40.63 206	51.45 179	59.74 162	44.71 217	49.37 59	40.69 198	7.61 237	46.04 155	44.34 199	17.32 226	57.79 201	61.18 182	73.30 149	65.86 175

LTVRE_ROB	44.17 16	47.06 200	50.15 203	43.44 193	51.39 180	58.42 173	42.90 221	43.51 126	22.27 238	14.85 219	41.94 193	34.57 227	45.43 138	62.28 179	62.77 174	62.56 208	68.83 153
Andreas Kuhn, Heiko HirschmÃ¼ller, Daniel Scharstein, Helmut Mayer: A TV Prior for High-Quality Scalable Multi-View Stereo Reconstruction. International Journal of Computer Vision 2016
CP-MVSNet		48.37 189	53.53 178	42.34 198	51.35 181	58.01 180	46.56 206	50.54 54	41.62 193	10.61 226	46.53 150	40.68 211	23.18 219	58.71 195	61.83 179	71.81 169	67.36 162

PS-CasMVS		48.18 191	53.25 182	42.27 199	51.26 182	57.94 181	46.51 207	50.52 55	41.30 194	10.56 227	45.35 165	40.34 213	23.04 220	58.66 196	61.79 180	71.74 171	67.38 160

our_test_3					51.15 183	57.31 184	55.12 171

SCA		50.99 173	53.22 183	48.40 164	51.07 184	56.78 186	50.25 189	39.05 176	48.31 141	41.38 121	49.54 126	46.70 174	46.00 136	58.31 197	56.28 198	62.65 206	56.60 214

UniMVSNet_ETH3D		52.62 157	55.98 161	48.70 159	51.04 185	60.71 154	56.87 153	46.74 74	42.52 188	26.96 188	42.50 190	45.95 181	37.87 175	66.22 153	65.15 146	72.74 156	68.78 154

IterMVS-SCA-FT		52.18 162	57.75 153	45.68 182	51.01 186	62.06 140	55.10 172	34.75 204	44.85 164	32.86 159	51.13 122	51.22 142	48.74 119	62.47 177	61.51 181	51.61 230	71.02 133

FMVSNet1		54.08 151	58.68 143	48.71 158	50.90 187	61.35 147	56.73 154	43.94 111	45.91 159	29.32 177	42.72 188	56.26 125	37.70 178	68.05 119	66.96 104	73.69 139	69.50 146

pmmvs-eth3d		51.33 170	52.25 189	50.26 143	50.82 188	54.65 191	56.03 161	43.45 131	43.51 177	37.20 146	39.20 199	39.04 217	42.28 156	61.85 181	62.78 173	71.78 170	64.72 184

MVSTER		57.19 122	61.11 110	52.62 129	50.82 188	58.79 170	61.55 122	37.86 192	48.81 134	41.31 122	57.43 92	52.10 139	48.60 122	68.19 115	66.75 110	75.56 115	75.68 110

ambc			45.54 219		50.66 190	52.63 199	40.99 225		38.36 212	24.67 198	22.62 234	13.94 244	29.14 207	65.71 160	58.06 196	58.60 216	67.43 158

thisisatest0515		53.85 152	56.84 159	50.37 142	50.25 191	58.17 178	55.99 162	39.90 174	41.88 191	38.16 140	45.91 156	45.30 185	44.58 143	66.15 155	66.89 108	73.36 147	73.57 125

baseline		55.19 145	60.88 111	48.55 161	49.87 192	58.10 179	58.70 139	34.75 204	52.82 107	39.48 136	60.18 76	60.86 101	45.41 139	61.05 183	60.74 186	63.10 204	72.41 127

MDTV_nov1_ep13		50.32 177	52.43 188	47.86 172	49.87 192	54.70 190	58.10 143	34.29 208	45.59 162	37.71 142	47.44 141	47.42 165	41.86 158	58.07 200	55.21 207	65.34 198	58.56 209

IterMVS		53.45 154	57.12 157	49.17 151	49.23 194	60.93 153	59.05 138	34.63 206	44.53 166	33.22 155	51.09 123	51.01 145	48.38 123	62.43 178	60.79 185	70.54 178	69.05 152
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
COLMAP_ROB	46.52 15	51.99 166	54.86 171	48.63 160	49.13 195	61.73 143	60.53 130	36.57 197	53.14 100	32.95 158	37.10 203	38.68 218	40.49 163	65.72 159	63.08 168	72.11 168	64.60 185
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
CR-MVSNet		50.47 174	52.61 185	47.98 170	49.03 196	52.94 196	48.27 196	38.86 182	44.41 167	39.59 133	44.34 171	44.65 196	46.63 133	58.97 192	60.31 187	65.48 196	62.66 193

V42		56.97 125	60.14 124	53.28 121	48.16 197	62.78 137	66.30 90	37.93 191	47.44 148	42.68 112	48.19 137	52.59 138	51.90 103	67.46 130	65.94 132	72.72 157	76.55 101

MDTV_nov1_ep13_2view		47.62 196	49.72 206	45.18 185	48.05 198	53.70 194	54.90 173	33.80 212	39.90 204	29.79 173	38.85 200	41.89 204	39.17 167	58.99 191	55.55 204	65.34 198	59.17 207

EPMVS		44.66 208	47.86 212	40.92 204	47.97 199	44.70 226	47.58 201	33.27 215	48.11 144	29.58 175	49.65 125	44.38 198	34.65 193	51.71 224	47.90 227	52.49 228	48.57 229

FE-MVSNET3		51.41 169	55.78 162	46.30 178	47.91 200	59.47 163	52.99 184	42.13 148	48.17 143	24.88 196	40.95 194	48.18 155	35.95 188	64.48 169	64.49 151	73.94 133	64.75 183

RPMNet		46.41 201	48.72 208	43.72 191	47.77 201	52.94 196	46.02 210	33.92 210	44.41 167	31.82 164	36.89 204	37.42 224	37.41 179	53.88 221	54.02 214	65.37 197	61.47 199

TAPA-MVS	54.74 10	60.85 90	66.61 77	54.12 116	47.38 202	65.33 112	65.35 102	36.51 198	75.16 31	48.82 82	54.70 103	63.51 88	53.31 91	68.36 107	64.97 147	73.37 146	74.27 118
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
SixPastTwentyTwo		47.55 197	50.25 202	44.41 190	47.30 203	54.31 193	47.81 199	40.36 171	33.76 222	19.93 210	43.75 176	32.77 231	42.07 157	59.82 188	60.94 184	68.98 182	66.37 169

TAMVS		44.02 210	49.18 207	37.99 214	47.03 204	45.97 223	45.04 214	28.47 226	39.11 208	20.23 209	43.22 184	48.52 153	28.49 210	58.15 199	57.95 197	58.71 214	51.36 220

UGNet		57.03 123	65.25 88	47.44 174	46.54 205	66.73 99	56.30 158	43.28 133	50.06 118	32.99 157	62.57 59	63.26 89	33.31 199	68.25 111	67.58 96	72.20 167	78.29 72
Wanjuan Su, Qingshan Xu, Wenbing Tao: Uncertainty-guided Multi-view Stereo Network for Depth Estimation. IEEE Transactions on Circuits and Systems for Video Technology, 2022
TSAR-MVS + COLMAP		62.65 79	69.90 55	54.19 114	46.31 206	66.73 99	65.49 101	41.36 155	76.57 25	46.31 93	76.80 18	56.68 119	53.27 92	69.50 91	66.65 114	72.40 164	76.36 104

PatchMatch-RL		50.11 179	51.56 193	48.43 163	46.23 207	51.94 200	50.21 190	38.62 187	46.62 154	37.51 143	42.43 191	39.38 215	52.24 100	60.98 184	59.56 190	65.76 195	60.01 206

CMPMVS	37.70 17	49.24 183	52.71 184	45.19 184	45.97 208	51.23 204	47.44 202	29.31 223	43.04 180	44.69 102	34.45 212	48.35 154	43.64 146	62.59 175	59.82 189	60.08 212	69.48 147
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
pmnet_mix02		40.48 220	43.80 223	36.61 217	45.79 209	40.45 231	42.12 223	33.18 216	40.30 201	24.11 202	38.76 201	37.11 225	24.30 216	52.97 222	46.66 231	50.17 231	50.33 224

pmmvs5		47.07 199	51.02 198	42.46 197	45.18 210	51.47 203	48.23 198	33.09 217	38.17 214	28.62 180	46.60 147	43.48 201	30.74 202	58.28 198	58.63 194	68.92 183	60.48 202

USDC		51.11 171	53.71 176	48.08 169	44.76 211	55.99 189	53.01 183	40.90 159	52.49 108	36.14 148	44.67 168	33.66 229	43.27 151	63.23 172	61.10 183	70.39 179	64.82 182

FC-MVSNet-test		39.65 223	48.35 210	29.49 227	44.43 212	39.28 235	30.23 237	40.44 169	43.59 175	3.12 244	53.00 109	42.03 203	10.02 240	55.09 217	54.77 209	48.66 232	50.71 222

ADS-MVSNet		40.67 218	43.38 225	37.50 215	44.36 213	39.79 233	42.09 224	32.67 220	44.34 169	28.87 179	40.76 197	40.37 212	30.22 203	48.34 234	45.87 232	46.81 234	44.21 233

new-patchmatchnet		33.24 230	37.20 231	28.62 229	44.32 214	38.26 236	29.68 238	36.05 199	31.97 227	6.33 239	26.59 229	27.33 235	11.12 239	50.08 230	41.05 235	44.23 235	45.15 232

MVS-HIRNet		42.24 214	41.15 228	43.51 192	44.06 215	40.74 229	35.77 233	35.35 201	35.38 220	38.34 138	25.63 231	38.55 219	43.48 148	50.77 226	47.03 229	64.07 200	49.98 225

PatchT		48.08 192	51.03 197	44.64 188	42.96 216	50.12 207	40.36 226	35.09 202	43.17 179	39.59 133	42.00 192	39.96 214	46.63 133	58.97 192	60.31 187	63.21 203	62.66 193

CVMVSNet		46.38 203	52.01 191	39.81 208	42.40 217	50.26 206	46.15 208	37.68 193	40.03 203	15.09 218	46.56 148	47.56 163	33.72 198	56.50 210	55.65 203	63.80 202	67.53 157

TinyColmap		47.08 198	47.56 213	46.52 177	42.35 218	53.44 195	51.77 186	40.70 163	43.44 178	31.92 163	29.78 222	23.72 240	45.04 142	61.99 180	59.54 191	67.35 189	61.03 200

ET-MVSNet_ETH3D		58.38 112	61.57 107	54.67 110	42.15 219	65.26 114	65.70 97	43.82 115	48.84 132	42.34 114	59.76 79	47.76 160	56.68 56	67.02 141	68.60 83	77.33 87	73.73 124

MIMVSNet		43.79 211	48.53 209	38.27 212	41.46 220	48.97 211	50.81 188	32.88 219	44.55 165	22.07 203	32.05 215	47.15 167	24.76 215	58.73 194	56.09 201	57.63 219	52.14 218

WB-MVS		29.70 232	35.40 233	23.05 232	40.96 221	39.59 234	18.79 241	40.20 172	25.26 234	1.88 247	33.33 213	21.97 242	3.36 241	48.69 233	44.60 233	33.11 239	34.39 235

N_pmnet		32.67 231	36.85 232	27.79 230	40.55 222	32.13 237	35.80 232	26.79 229	37.24 217	9.10 232	32.02 216	30.94 233	16.30 228	47.22 235	41.21 234	38.21 237	37.21 234

anonymousdsp		52.84 156	57.78 152	47.06 175	40.24 223	58.95 169	53.70 178	33.54 214	36.51 219	32.69 160	43.88 174	45.40 183	47.97 128	67.17 136	70.28 63	74.22 129	82.29 48

FE-MVSNET2		45.69 205	49.95 204	40.72 205	40.11 224	56.16 187	46.59 205	41.89 149	36.97 218	13.66 221	29.00 225	37.59 223	28.96 209	63.26 171	63.93 161	73.13 152	62.72 192

EU-MVSNet		40.63 219	45.65 218	34.78 222	39.11 225	46.94 220	40.02 227	34.03 209	33.50 223	10.37 228	35.57 208	37.80 221	23.65 218	51.90 223	50.21 224	61.49 210	63.62 190

FPMVS		38.36 225	40.41 229	35.97 218	38.92 226	39.85 232	45.50 212	25.79 232	41.13 195	18.70 212	30.10 220	24.56 238	31.86 201	49.42 231	46.80 230	55.04 221	51.03 221

TESTMET0.1,1		46.09 204	50.29 200	41.18 203	36.91 227	47.16 217	49.52 191	20.32 236	39.22 206	31.98 161	43.65 178	47.93 158	41.29 161	56.80 206	55.36 205	67.08 191	61.94 197

FMVSNet5		40.96 216	45.81 217	35.29 221	34.30 228	44.55 227	47.28 203	28.84 225	40.76 197	21.62 204	29.85 221	42.44 202	24.77 214	57.53 202	55.00 208	54.93 222	50.56 223

PMMVS		49.20 185	54.28 175	43.28 195	34.13 229	45.70 224	48.98 194	26.09 231	46.31 156	34.92 154	55.22 99	53.47 133	47.48 130	59.43 189	59.04 193	68.05 187	60.77 201

PMVS	27.84 18	33.81 229	35.28 234	32.09 225	34.13 229	24.81 240	32.51 236	26.48 230	26.41 233	19.37 211	23.76 232	24.02 239	25.18 213	50.78 225	47.24 228	54.89 224	49.95 226
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
test-mter		45.30 206	50.37 199	39.38 209	33.65 231	46.99 219	47.59 200	18.59 237	38.75 209	28.00 181	43.28 183	46.82 173	41.50 160	57.28 203	55.78 202	66.93 193	63.70 189

CHOSEN 280x420		40.80 217	45.05 220	35.84 220	32.95 232	29.57 238	44.98 215	23.71 234	37.54 216	18.42 213	31.36 218	47.07 168	46.41 135	56.71 208	54.65 212	48.55 233	58.47 210

PM-MVS		44.55 209	48.13 211	40.37 207	32.85 233	46.82 221	46.11 209	29.28 224	40.48 199	29.99 172	39.98 198	34.39 228	41.80 159	56.08 213	53.88 218	62.19 209	65.31 178

FE-MVSNET		39.75 222	44.50 221	34.21 223	32.01 234	48.77 212	37.71 230	38.94 179	30.91 230	6.25 240	26.24 230	32.10 232	23.68 217	57.28 203	59.53 192	66.68 194	56.64 213

TDRefinement		49.31 181	52.44 187	45.67 183	30.44 235	59.42 164	59.24 136	39.78 175	48.76 135	31.20 166	35.73 207	29.90 234	42.81 154	64.24 170	62.59 177	70.55 177	66.43 167

Gipuma		25.87 233	26.91 236	24.66 231	28.98 236	20.17 241	20.46 239	34.62 207	29.55 231	9.10 232	4.91 244	5.31 248	15.76 229	49.37 232	49.10 226	39.03 236	29.95 237
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
MDA-MVSNet-bldmvs		41.36 215	43.15 226	39.27 210	28.74 237	52.68 198	44.95 216	40.84 160	32.89 224	18.13 214	31.61 217	22.09 241	38.97 170	50.45 228	56.11 200	64.01 201	56.23 215

E-PMN		15.09 236	13.19 240	17.30 234	27.80 238	12.62 244	7.81 245	27.54 227	14.62 242	3.19 242	6.89 241	2.52 251	15.09 230	15.93 240	20.22 239	22.38 240	19.53 240

MIMVSNet1		35.51 227	41.41 227	28.63 228	27.53 239	43.36 228	38.09 229	33.82 211	32.01 226	6.77 238	21.63 235	35.43 226	11.97 234	55.05 218	53.99 216	53.59 227	48.36 230

EMVS		14.49 237	12.45 241	16.87 236	27.02 240	12.56 245	8.13 244	27.19 228	15.05 241	3.14 243	6.69 242	2.67 250	15.08 231	14.60 242	18.05 240	20.67 241	17.56 242

pmmvs3		35.10 228	38.47 230	31.17 226	26.37 241	40.47 230	34.51 235	18.09 238	24.75 235	16.88 216	23.05 233	26.69 236	32.69 200	50.73 227	51.60 221	58.46 217	51.98 219

RPSCF		46.41 201	54.42 173	37.06 216	25.70 242	45.14 225	45.39 213	20.81 235	62.79 61	35.10 151	44.92 167	55.60 128	43.56 147	56.12 212	52.45 220	51.80 229	63.91 188

new_pmnet		23.19 234	28.17 235	17.37 233	17.03 243	24.92 239	19.66 240	16.16 240	27.05 232	4.42 241	20.77 236	19.20 243	12.19 233	37.71 236	36.38 236	34.77 238	31.17 236

PMMVS2		15.84 235	19.68 237	11.35 237	15.74 244	16.95 242	13.31 242	17.64 239	16.08 240	0.36 248	13.12 238	11.47 245	1.69 243	28.82 237	27.24 238	19.38 243	24.09 239

MVE	12.28 19	13.53 238	15.72 238	10.96 238	7.39 245	15.71 243	6.05 246	23.73 233	10.29 244	3.01 245	5.77 243	3.41 249	11.91 235	20.11 238	29.79 237	13.67 244	24.98 238
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
tmp_tt				5.40 240	3.97 246	2.35 248	3.26 248	0.44 243	17.56 239	12.09 223	11.48 240	7.14 246	1.98 242	15.68 241	15.49 241	10.69 245

test_method		12.44 239	14.66 239	9.85 239	1.30 247	3.32 247	13.00 243	3.21 241	22.42 237	10.22 229	14.13 237	25.64 237	11.43 237	19.75 239	11.61 242	19.96 242	5.79 243

GG-mvs-BLEND		36.62 226	53.39 180	17.06 235	0.01 248	58.61 171	48.63 195	0.01 244	47.13 150	0.02 249	43.98 173	60.64 104	0.03 244	54.92 219	51.47 222	53.64 226	56.99 212

uanet_test		0.00 242	0.00 244	0.00 241	0.00 249	0.00 249	0.00 251	0.00 245	0.00 247	0.00 250	0.00 247	0.00 252	0.00 247	0.00 245	0.00 245	0.00 247	0.00 246

sosnet-low-res		0.00 242	0.00 244	0.00 241	0.00 249	0.00 249	0.00 251	0.00 245	0.00 247	0.00 250	0.00 247	0.00 252	0.00 247	0.00 245	0.00 245	0.00 247	0.00 246

sosnet		0.00 242	0.00 244	0.00 241	0.00 249	0.00 249	0.00 251	0.00 245	0.00 247	0.00 250	0.00 247	0.00 252	0.00 247	0.00 245	0.00 245	0.00 247	0.00 246

testmvs		0.01 240	0.02 242	0.00 241	0.00 249	0.00 249	0.01 250	0.00 245	0.01 245	0.00 250	0.03 246	0.00 252	0.01 245	0.01 244	0.01 243	0.00 247	0.06 245

test123		0.01 240	0.02 242	0.00 241	0.00 249	0.00 249	0.00 251	0.00 245	0.01 245	0.00 250	0.04 245	0.00 252	0.01 245	0.00 245	0.01 243	0.00 247	0.07 244

RE-MVS-def										33.01 156

9.14												81.81 14

MTAPA										65.14 4		80.20 21

MTMP										62.63 16		78.04 28

Patchmatch-RL test							1.04 249

NP-MVS									72.00 43

Patchmtry						47.61 215	48.27 196	38.86 182		39.59 133

DeepMVS_CX						6.95 246	5.98 247	2.25 242	11.73 243	2.07 246	11.85 239	5.43 247	11.75 236	11.40 243		8.10 246	18.38 241

High-resolution multi-view benchmark