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.

HSP-MVS		87.45 3	90.22 3	84.22 7	90.00 18	91.80 3	90.59 4	75.80 4	89.93 4	78.35 15	92.54 2	89.18 3	80.89 1	87.99 11	86.29 25	89.70 35	93.85 8

APDe-MVS		88.00 2	90.50 2	85.08 2	90.95 5	91.58 4	92.03 1	75.53 8	91.15 1	80.10 9	92.27 3	88.34 6	80.80 2	88.00 10	86.99 14	91.09 4	95.16 2

ESAPD		88.46 1	91.07 1	85.41 1	91.73 2	92.08 1	91.91 2	76.73 1	90.14 3	80.33 8	92.75 1	90.44 1	80.73 3	88.97 4	87.63 8	91.01 6	95.48 1

APD-MVS		86.84 7	88.91 9	84.41 4	90.66 8	90.10 7	90.78 3	75.64 5	87.38 12	78.72 13	90.68 6	86.82 10	80.15 4	87.13 20	86.45 23	90.51 15	93.83 9
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
TSAR-MVS + MP.		86.88 6	89.23 5	84.14 8	89.78 21	88.67 26	90.59 4	73.46 21	88.99 6	80.52 7	91.26 4	88.65 4	79.91 5	86.96 25	86.22 26	90.59 13	93.83 9

HPM-MVS++		87.09 4	88.92 8	84.95 3	92.61 1	87.91 34	90.23 9	76.06 3	88.85 7	81.20 4	87.33 9	87.93 7	79.47 6	88.59 5	88.23 4	90.15 28	93.60 15

CNVR-MVS		86.36 9	88.19 12	84.23 6	91.33 4	89.84 9	90.34 6	75.56 6	87.36 13	78.97 12	81.19 23	86.76 11	78.74 7	89.30 2	88.58 1	90.45 21	94.33 5

SD-MVS		86.96 5	89.45 4	84.05 10	90.13 15	89.23 17	89.77 12	74.59 9	89.17 5	80.70 5	89.93 7	89.67 2	78.47 8	87.57 15	86.79 17	90.67 12	93.76 11

MCST-MVS		85.13 18	86.62 19	83.39 13	90.55 11	89.82 11	89.29 16	73.89 18	84.38 26	76.03 23	79.01 26	85.90 15	78.47 8	87.81 12	86.11 28	92.11 1	93.29 17

HFP-MVS		86.15 10	87.95 13	84.06 9	90.80 6	89.20 18	89.62 14	74.26 11	87.52 10	80.63 6	86.82 12	84.19 23	78.22 10	87.58 14	87.19 12	90.81 7	93.13 19

MPTG		85.71 12	86.88 18	84.34 5	90.54 12	87.11 38	89.77 12	74.17 13	88.54 8	83.08 2	78.60 27	86.10 13	78.11 11	87.80 13	87.46 10	90.35 24	92.56 21

NCCC		85.34 15	86.59 20	83.88 11	91.48 3	88.88 20	89.79 11	75.54 7	86.67 16	77.94 18	76.55 30	84.99 19	78.07 12	88.04 8	87.68 7	90.46 20	93.31 16

TSAR-MVS + GP.		83.69 25	86.58 21	80.32 30	85.14 48	86.96 39	84.91 44	70.25 35	84.71 24	73.91 30	85.16 16	85.63 16	77.92 13	85.44 35	85.71 31	89.77 32	92.45 22

MSLP-MVS++		82.09 31	82.66 35	81.42 24	87.03 37	87.22 37	85.82 36	70.04 36	80.30 38	78.66 14	68.67 56	81.04 37	77.81 14	85.19 38	84.88 38	89.19 44	91.31 32

ACMMPR		85.52 13	87.53 15	83.17 17	90.13 15	89.27 15	89.30 15	73.97 16	86.89 15	77.14 20	86.09 13	83.18 26	77.74 15	87.42 16	87.20 11	90.77 8	92.63 20

DeepC-MVS_fast	78.24 3	84.27 24	85.50 26	82.85 18	90.46 13	89.24 16	87.83 27	74.24 12	84.88 21	76.23 22	75.26 33	81.05 36	77.62 16	88.02 9	87.62 9	90.69 11	92.41 23
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
DeepC-MVS	78.47 2	84.81 21	86.03 24	83.37 14	89.29 26	90.38 6	88.61 21	76.50 2	86.25 18	77.22 19	75.12 34	80.28 38	77.59 17	88.39 6	88.17 5	91.02 5	93.66 13
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
DeepPCF-MVS	79.04 1	85.30 16	88.93 7	81.06 26	88.77 29	90.48 5	85.46 40	73.08 22	90.97 2	73.77 31	84.81 17	85.95 14	77.43 18	88.22 7	87.73 6	87.85 66	94.34 4

MP-MVS		85.50 14	87.40 16	83.28 15	90.65 9	89.51 14	89.16 18	74.11 14	83.70 28	78.06 17	85.54 15	84.89 21	77.31 19	87.40 17	87.14 13	90.41 22	93.65 14

SteuartSystems-ACMMP		85.99 11	88.31 11	83.27 16	90.73 7	89.84 9	90.27 8	74.31 10	84.56 25	75.88 24	87.32 10	85.04 18	77.31 19	89.01 3	88.46 2	91.14 3	93.96 7
Skip Steuart: Steuart Systems R&D Blog.
CP-MVS		84.74 22	86.43 22	82.77 19	89.48 24	88.13 33	88.64 20	73.93 17	84.92 20	76.77 21	81.94 21	83.50 24	77.29 21	86.92 26	86.49 22	90.49 16	93.14 18

3Dnovator+	75.73 4	82.40 29	82.76 34	81.97 23	88.02 31	89.67 12	86.60 31	71.48 30	81.28 37	78.18 16	64.78 69	77.96 44	77.13 22	87.32 18	86.83 16	90.41 22	91.48 31

PGM-MVS		84.42 23	86.29 23	82.23 21	90.04 17	88.82 22	89.23 17	71.74 29	82.82 31	74.61 27	84.41 18	82.09 28	77.03 23	87.13 20	86.73 19	90.73 10	92.06 27

ACMMP_Plus		86.52 8	89.01 6	83.62 12	90.28 14	90.09 8	90.32 7	74.05 15	88.32 9	79.74 10	87.04 11	85.59 17	76.97 24	89.35 1	88.44 3	90.35 24	94.27 6

AdaColmap		79.74 41	78.62 52	81.05 27	89.23 27	86.06 46	84.95 43	71.96 27	79.39 42	75.51 25	63.16 73	68.84 79	76.51 25	83.55 49	82.85 48	88.13 58	86.46 62

CPTT-MVS		81.77 32	83.10 33	80.21 31	85.93 44	86.45 44	87.72 28	70.98 32	82.54 33	71.53 43	74.23 39	81.49 33	76.31 26	82.85 56	81.87 52	88.79 51	92.26 25

train_agg		84.86 20	87.21 17	82.11 22	90.59 10	85.47 49	89.81 10	73.55 20	83.95 27	73.30 32	89.84 8	87.23 9	75.61 27	86.47 29	85.46 33	89.78 31	92.06 27

ACMMP		83.42 26	85.27 27	81.26 25	88.47 30	88.49 29	88.31 25	72.09 26	83.42 29	72.77 35	82.65 19	78.22 42	75.18 28	86.24 32	85.76 30	90.74 9	92.13 26
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
TSAR-MVS + ACMM		85.10 19	88.81 10	80.77 29	89.55 23	88.53 28	88.59 22	72.55 24	87.39 11	71.90 37	90.95 5	87.55 8	74.57 29	87.08 22	86.54 21	87.47 71	93.67 12

CNLPA		77.20 54	77.54 57	76.80 50	82.63 58	84.31 57	79.77 61	64.64 71	85.17 19	73.18 33	56.37 107	69.81 70	74.53 30	81.12 72	78.69 90	86.04 127	87.29 59

MVS_0304		81.73 33	83.86 30	79.26 36	86.22 43	89.18 19	86.41 32	67.15 56	75.28 49	70.75 47	74.59 36	83.49 25	74.42 31	87.05 23	86.34 24	90.58 14	91.08 35

OPM-MVS		79.68 42	79.28 51	80.15 32	87.99 32	86.77 41	88.52 23	72.72 23	64.55 81	67.65 54	67.87 60	74.33 55	74.31 32	86.37 31	85.25 35	89.73 34	89.81 44

3Dnovator	73.76 5	79.75 40	80.52 45	78.84 39	84.94 53	87.35 35	84.43 46	65.54 66	78.29 43	73.97 29	63.00 75	75.62 51	74.07 33	85.00 39	85.34 34	90.11 29	89.04 47

CSCG		85.28 17	87.68 14	82.49 20	89.95 19	91.99 2	88.82 19	71.20 31	86.41 17	79.63 11	79.26 24	88.36 5	73.94 34	86.64 27	86.67 20	91.40 2	94.41 3

PLC	68.99 11	75.68 59	75.31 69	76.12 53	82.94 57	81.26 76	79.94 60	66.10 61	77.15 46	66.86 59	59.13 88	68.53 80	73.73 35	80.38 82	79.04 87	87.13 81	81.68 123
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
ACMM	72.26 8	78.86 49	78.13 53	79.71 34	86.89 38	83.40 63	86.02 34	70.50 33	75.28 49	71.49 44	63.01 74	69.26 73	73.57 36	84.11 44	83.98 42	89.76 33	87.84 54
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
OMC-MVS		80.26 36	82.59 36	77.54 46	83.04 56	85.54 48	83.25 51	65.05 69	87.32 14	72.42 36	72.04 44	78.97 40	73.30 37	83.86 45	81.60 55	88.15 57	88.83 49

MVS_111021_HR		80.13 37	81.46 39	78.58 41	85.77 45	85.17 53	83.45 50	69.28 43	74.08 55	70.31 48	74.31 38	75.26 52	73.13 38	86.46 30	85.15 36	89.53 38	89.81 44

PHI-MVS		82.36 30	85.89 25	78.24 43	86.40 41	89.52 13	85.52 38	69.52 42	82.38 34	65.67 60	81.35 22	82.36 27	73.07 39	87.31 19	86.76 18	89.24 42	91.56 30

CANet		81.62 34	83.41 31	79.53 35	87.06 36	88.59 27	85.47 39	67.96 52	76.59 47	74.05 28	74.69 35	81.98 29	72.98 40	86.14 33	85.47 32	89.68 36	90.42 41

QAPM		78.47 50	80.22 48	76.43 51	85.03 50	86.75 42	80.62 57	66.00 63	73.77 56	65.35 61	65.54 67	78.02 43	72.69 41	83.71 47	83.36 47	88.87 50	90.41 42

abl_6				79.05 37	87.27 35	88.85 21	83.62 49	68.25 48	81.68 35	72.94 34	73.79 40	84.45 22	72.55 42			89.66 37	90.64 38

MVS_111021_LR		78.13 52	79.85 50	76.13 52	81.12 65	81.50 73	80.28 58	65.25 67	76.09 48	71.32 45	76.49 31	72.87 59	72.21 43	82.79 57	81.29 57	86.59 112	87.91 53

MAR-MVS		79.21 45	80.32 47	77.92 45	87.46 33	88.15 32	83.95 47	67.48 55	74.28 53	68.25 51	64.70 70	77.04 45	72.17 44	85.42 36	85.00 37	88.22 55	87.62 56
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
ACMP	73.23 7	79.79 39	80.53 44	78.94 38	85.61 46	85.68 47	85.61 37	69.59 40	77.33 45	71.00 46	74.45 37	69.16 74	71.88 45	83.15 53	83.37 46	89.92 30	90.57 40
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
TAPA-MVS	71.42 9	77.69 53	80.05 49	74.94 57	80.68 68	84.52 56	81.36 53	63.14 81	84.77 22	64.82 64	68.72 54	75.91 50	71.86 46	81.62 61	79.55 83	87.80 67	85.24 76
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
CDPH-MVS		82.64 28	85.03 29	79.86 33	89.41 25	88.31 30	88.32 24	71.84 28	80.11 39	67.47 55	82.09 20	81.44 34	71.85 47	85.89 34	86.15 27	90.24 26	91.25 33

PCF-MVS	73.28 6	79.42 43	80.41 46	78.26 42	84.88 54	88.17 31	86.08 33	69.85 37	75.23 51	68.43 50	68.03 59	78.38 41	71.76 48	81.26 70	80.65 71	88.56 54	91.18 34
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
X-MVS		83.23 27	85.20 28	80.92 28	89.71 22	88.68 23	88.21 26	73.60 19	82.57 32	71.81 40	77.07 28	81.92 30	71.72 49	86.98 24	86.86 15	90.47 17	92.36 24

HQP-MVS		81.19 35	83.27 32	78.76 40	87.40 34	85.45 50	86.95 29	70.47 34	81.31 36	66.91 58	79.24 25	76.63 46	71.67 50	84.43 42	83.78 43	89.19 44	92.05 29

TSAR-MVS + COLMAP		78.34 51	81.64 38	74.48 61	80.13 72	85.01 54	81.73 52	65.93 65	84.75 23	61.68 71	85.79 14	66.27 85	71.39 51	82.91 55	80.78 62	86.01 128	85.98 64

LGP-MVS_train		79.83 38	81.22 41	78.22 44	86.28 42	85.36 52	86.76 30	69.59 40	77.34 44	65.14 62	75.68 32	70.79 65	71.37 52	84.60 40	84.01 41	90.18 27	90.74 37

Fast-Effi-MVS+		73.11 69	73.66 72	72.48 67	77.72 96	80.88 82	78.55 85	58.83 158	65.19 75	60.36 75	59.98 83	62.42 96	71.22 53	81.66 60	80.61 73	88.20 56	84.88 84

LS3D		74.08 64	73.39 74	74.88 58	85.05 49	82.62 69	79.71 62	68.66 46	72.82 57	58.80 81	57.61 100	61.31 98	71.07 54	80.32 86	78.87 89	86.00 130	80.18 137

Effi-MVS+		75.28 61	76.20 66	74.20 62	81.15 64	83.24 64	81.11 54	63.13 82	66.37 67	60.27 76	64.30 71	68.88 78	70.93 55	81.56 63	81.69 54	88.61 52	87.35 57

OpenMVS	70.44 10	76.15 58	76.82 65	75.37 55	85.01 51	84.79 55	78.99 70	62.07 107	71.27 59	67.88 53	57.91 99	72.36 60	70.15 56	82.23 59	81.41 56	88.12 59	87.78 55

DELS-MVS		79.15 47	81.07 42	76.91 49	83.54 55	87.31 36	84.45 45	64.92 70	69.98 60	69.34 49	71.62 46	76.26 47	69.84 57	86.57 28	85.90 29	89.39 40	89.88 43
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
canonicalmvs		79.16 46	82.37 37	75.41 54	82.33 61	86.38 45	80.80 56	63.18 80	82.90 30	67.34 56	72.79 42	76.07 48	69.62 58	83.46 52	84.41 40	89.20 43	90.60 39

PatchMatch-RL		67.78 131	66.65 153	69.10 105	73.01 157	72.69 174	68.49 165	61.85 110	62.93 90	60.20 77	56.83 106	50.42 184	69.52 59	75.62 155	74.46 166	81.51 172	73.62 185

DI_MVS_plusplus_trai		75.13 62	76.12 67	73.96 63	78.18 82	81.55 72	80.97 55	62.54 101	68.59 64	65.13 63	61.43 76	74.81 53	69.32 60	81.01 74	79.59 81	87.64 69	85.89 65

MVS_Test		75.37 60	77.13 63	73.31 65	79.07 77	81.32 75	79.98 59	60.12 139	69.72 63	64.11 66	70.53 48	73.22 57	68.90 61	80.14 90	79.48 85	87.67 68	85.50 71

Effi-MVS+-dtu		71.82 74	71.86 82	71.78 68	78.77 78	80.47 91	78.55 85	61.67 113	60.68 104	55.49 107	58.48 92	65.48 87	68.85 62	76.92 142	75.55 157	87.35 73	85.46 72

ACMH	65.37 14	70.71 81	70.00 92	71.54 69	82.51 60	82.47 70	77.78 96	68.13 49	56.19 154	46.06 162	54.30 134	51.20 180	68.68 63	80.66 76	80.72 64	86.07 123	84.45 88
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
ACMH+	66.54 13	71.36 77	70.09 91	72.85 66	82.59 59	81.13 77	78.56 84	68.04 50	61.55 99	52.52 126	51.50 174	54.14 143	68.56 64	78.85 106	79.50 84	86.82 98	83.94 92

CLD-MVS		79.35 44	81.23 40	77.16 48	85.01 51	86.92 40	85.87 35	60.89 120	80.07 41	75.35 26	72.96 41	73.21 58	68.43 65	85.41 37	84.63 39	87.41 72	85.44 73
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
HyFIR lowres test		69.47 108	68.94 116	70.09 96	76.77 105	82.93 67	76.63 105	60.17 134	59.00 116	54.03 114	40.54 205	65.23 88	67.89 66	76.54 149	78.30 98	85.03 149	80.07 138

PVSNet_Blended_VisFu		76.57 55	77.90 54	75.02 56	80.56 69	86.58 43	79.24 65	66.18 60	64.81 78	68.18 52	65.61 65	71.45 62	67.05 67	84.16 43	81.80 53	88.90 48	90.92 36

PVSNet_BlendedMVS		76.21 56	77.52 58	74.69 59	79.46 74	83.79 60	77.50 99	64.34 74	69.88 61	71.88 38	68.54 57	70.42 67	67.05 67	83.48 50	79.63 79	87.89 64	86.87 60

PVSNet_Blended		76.21 56	77.52 58	74.69 59	79.46 74	83.79 60	77.50 99	64.34 74	69.88 61	71.88 38	68.54 57	70.42 67	67.05 67	83.48 50	79.63 79	87.89 64	86.87 60

v7		70.33 88	69.87 96	70.88 70	74.79 132	81.04 78	79.22 66	60.57 124	57.70 129	56.65 102	54.23 139	55.29 135	66.95 70	78.28 112	77.47 113	87.12 84	85.05 80

v10		70.22 90	69.76 100	70.74 76	74.79 132	80.30 94	79.22 66	59.81 142	57.71 128	56.58 103	54.22 141	55.31 133	66.95 70	78.28 112	77.47 113	87.12 84	85.07 79

v1192		69.50 107	68.83 119	70.29 92	74.49 145	80.92 81	78.55 85	60.54 125	55.04 166	54.21 112	52.79 163	52.33 168	66.92 72	77.88 117	77.35 118	87.04 87	85.51 70

CHOSEN 1792x2688		69.20 112	69.26 112	69.13 104	76.86 104	78.93 104	77.27 101	60.12 139	61.86 96	54.42 111	42.54 199	61.61 97	66.91 73	78.55 109	78.14 102	79.23 184	83.23 102

v1921920		69.03 113	68.32 132	69.86 98	74.03 150	80.37 92	77.55 97	60.25 133	54.62 169	53.59 118	52.36 170	51.50 179	66.75 74	77.17 138	76.69 142	86.96 90	85.56 67

v144192		69.34 110	68.68 126	70.12 95	74.06 149	80.54 87	78.08 95	60.54 125	54.99 168	54.13 113	52.92 160	52.80 164	66.73 75	77.13 139	76.72 137	87.15 77	85.63 66

v1240		68.64 118	67.89 139	69.51 102	73.89 152	80.26 95	76.73 104	59.97 141	53.43 178	53.08 121	51.82 173	50.84 182	66.62 76	76.79 144	76.77 125	86.78 104	85.34 74

v1144		69.93 99	69.36 111	70.61 81	74.89 125	80.93 79	79.11 68	60.64 122	55.97 157	55.31 109	53.85 146	54.14 143	66.54 77	78.10 114	77.44 115	87.14 80	85.09 78

v11		69.37 109	68.65 127	70.20 93	74.87 128	76.97 140	78.29 92	58.55 162	56.38 151	56.04 105	54.02 143	54.98 137	66.47 78	78.30 111	76.91 123	86.97 89	83.02 103

v2v482		70.05 94	69.46 105	70.74 76	74.62 144	80.32 93	79.00 69	60.62 123	57.41 130	56.89 95	55.43 114	55.14 136	66.39 79	77.25 137	77.14 120	86.90 91	83.57 100

v17		70.03 95	69.43 110	70.72 78	74.75 137	77.09 132	78.78 80	58.85 154	59.53 114	58.72 82	54.87 126	57.39 111	66.38 80	77.60 124	76.75 130	86.83 97	82.80 105

v16		70.07 93	69.46 105	70.79 74	74.74 138	77.08 133	78.79 78	58.86 152	59.75 111	59.15 79	54.87 126	57.33 112	66.38 80	77.61 123	76.77 125	86.81 103	82.79 107

v18		70.10 92	69.52 103	70.77 75	74.66 143	77.06 134	78.84 73	58.84 157	60.01 110	59.23 78	55.06 119	57.47 110	66.34 82	77.50 129	76.75 130	86.71 105	82.77 109

v1neww		70.34 86	69.93 94	70.82 72	74.68 140	80.61 85	78.80 76	60.17 134	58.74 119	58.10 86	55.00 121	57.28 115	66.33 83	77.53 125	76.74 132	86.82 98	83.61 95

v7new		70.34 86	69.93 94	70.82 72	74.68 140	80.61 85	78.80 76	60.17 134	58.74 119	58.10 86	55.00 121	57.28 115	66.33 83	77.53 125	76.74 132	86.82 98	83.61 95

v6		70.35 85	69.94 93	70.83 71	74.68 140	80.62 84	78.81 75	60.16 137	58.81 117	58.17 85	55.01 120	57.31 114	66.32 85	77.53 125	76.73 136	86.82 98	83.62 94

v8		70.23 89	69.86 98	70.67 80	74.69 139	79.82 96	78.79 78	59.18 148	58.80 118	58.20 84	55.00 121	57.33 112	66.31 86	77.51 128	76.71 140	86.82 98	83.88 93

v12		69.54 104	68.79 121	70.41 89	74.88 126	77.03 137	78.54 88	58.85 154	56.71 140	56.87 96	54.13 142	56.23 128	66.15 87	77.89 116	76.74 132	87.17 76	82.80 105

V14		69.59 102	68.86 118	70.45 88	74.83 130	77.04 135	78.70 82	58.83 158	56.95 136	57.08 93	54.41 133	56.34 122	66.15 87	77.77 119	76.76 127	87.08 86	82.74 110

V9		69.58 103	68.83 119	70.46 86	74.85 129	77.04 135	78.65 83	58.85 154	56.83 139	57.12 92	54.26 137	56.31 123	66.14 89	77.83 118	76.76 127	87.13 81	82.79 107

v15		69.61 101	68.88 117	70.46 86	74.81 131	77.03 137	78.75 81	58.83 158	57.06 132	57.18 91	54.55 132	56.37 121	66.13 90	77.70 120	76.76 127	87.03 88	82.69 112

v13		69.52 106	68.76 123	70.41 89	74.88 126	77.02 139	78.52 89	58.86 152	56.61 148	56.91 94	54.00 144	56.17 129	66.11 91	77.93 115	76.74 132	87.21 75	82.83 104

IterMVS-LS		71.69 75	72.82 78	70.37 91	77.54 98	76.34 149	75.13 114	60.46 127	61.53 100	57.57 89	64.89 68	67.33 82	66.04 92	77.09 141	77.37 117	85.48 143	85.18 77
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
v1141		69.96 98	69.44 108	70.58 84	74.78 134	80.50 89	78.85 71	60.30 129	56.95 136	56.74 99	54.68 130	56.26 127	65.93 93	77.38 131	76.72 137	86.88 94	83.57 100

divwei89l23v2f112		69.97 96	69.44 108	70.58 84	74.78 134	80.50 89	78.85 71	60.30 129	56.97 135	56.75 98	54.67 131	56.27 126	65.92 94	77.37 132	76.72 137	86.88 94	83.58 99

v1		69.97 96	69.45 107	70.59 82	74.78 134	80.51 88	78.84 73	60.30 129	56.98 133	56.81 97	54.69 129	56.29 125	65.91 95	77.37 132	76.71 140	86.89 93	83.59 97

MSDG		71.52 76	69.87 96	73.44 64	82.21 62	79.35 100	79.52 63	64.59 72	66.15 69	61.87 70	53.21 155	56.09 130	65.85 96	78.94 105	78.50 91	86.60 111	76.85 165

DWT-MVSNet_training		67.24 143	65.96 160	68.74 107	76.15 111	74.36 170	74.37 123	56.66 171	61.82 97	60.51 74	58.23 98	49.76 188	65.07 97	70.04 194	70.39 181	79.70 181	77.11 162

diffmvs		73.13 68	75.65 68	70.19 94	74.07 148	77.17 131	78.24 93	57.45 166	72.44 58	64.02 67	69.05 52	75.92 49	64.86 98	75.18 158	75.27 160	82.47 169	84.53 86

V42		68.76 117	69.63 101	67.74 117	64.93 199	78.01 117	78.30 91	56.48 172	58.65 121	56.30 104	54.26 137	57.03 118	64.85 99	77.47 130	77.01 122	85.60 141	84.96 82

EPNet		79.08 48	80.62 43	77.28 47	88.90 28	83.17 66	83.65 48	72.41 25	74.41 52	67.15 57	76.78 29	74.37 54	64.43 100	83.70 48	83.69 44	87.15 77	88.19 51
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
RPSCF		67.64 136	71.25 84	63.43 163	61.86 205	70.73 180	67.26 173	50.86 194	74.20 54	58.91 80	67.49 61	69.33 72	64.10 101	71.41 179	68.45 195	77.61 188	77.17 160

LTVRE_ROB	59.44 16	61.82 187	62.64 188	60.87 177	72.83 162	77.19 130	64.37 190	58.97 149	33.56 225	28.00 208	52.59 168	42.21 209	63.93 102	74.52 160	76.28 144	77.15 191	82.13 114
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
MVSTER		72.06 73	74.24 71	69.51 102	70.39 178	75.97 153	76.91 103	57.36 168	64.64 80	61.39 73	68.86 53	63.76 91	63.46 103	81.44 64	79.70 78	87.56 70	85.31 75

tpmp4_e23		68.32 121	67.08 148	69.76 100	77.86 86	75.22 163	78.37 90	56.17 175	66.06 71	64.27 65	57.15 104	54.89 138	63.40 104	70.97 186	68.29 196	78.46 186	77.00 164

PMMVS		65.06 157	69.17 114	60.26 180	55.25 222	63.43 205	66.71 178	43.01 220	62.41 91	50.64 135	69.44 51	67.04 83	63.29 105	74.36 162	73.54 169	82.68 168	73.99 182

EPP-MVSNet		74.00 65	77.41 60	70.02 97	80.53 70	83.91 59	74.99 116	62.68 97	65.06 76	49.77 142	68.68 55	72.09 61	63.06 106	82.49 58	80.73 63	89.12 46	88.91 48

CHOSEN 280x420		58.70 196	61.88 194	54.98 200	55.45 221	50.55 225	64.92 187	40.36 222	55.21 163	38.13 190	48.31 184	63.76 91	63.03 107	73.73 166	68.58 193	68.00 218	73.04 186

TinyColmap		62.84 173	61.03 197	64.96 151	69.61 183	71.69 177	68.48 166	59.76 143	55.41 161	47.69 151	47.33 191	34.20 218	62.76 108	74.52 160	72.59 174	81.44 173	71.47 193

Fast-Effi-MVS+-dtu		68.34 120	69.47 104	67.01 135	75.15 121	77.97 123	77.12 102	55.40 176	57.87 123	46.68 159	56.17 110	60.39 99	62.36 109	76.32 150	76.25 146	85.35 145	81.34 124

USDC		67.36 141	67.90 138	66.74 141	71.72 168	75.23 161	71.58 153	60.28 132	67.45 66	50.54 137	60.93 77	45.20 205	62.08 110	76.56 148	74.50 165	84.25 160	75.38 174

MS-PatchMatch		70.17 91	70.49 89	69.79 99	80.98 67	77.97 123	77.51 98	58.95 150	62.33 92	55.22 110	53.14 156	65.90 86	62.03 111	79.08 104	77.11 121	84.08 161	77.91 155

v148		67.85 129	67.53 141	68.23 112	73.25 156	77.57 129	74.26 130	57.36 168	55.70 159	57.45 90	53.53 148	55.42 132	61.96 112	75.23 157	73.92 167	85.08 148	81.32 125

pmmvs4		67.89 128	67.39 145	68.48 111	71.60 172	73.57 172	74.45 119	60.98 119	64.65 79	57.97 88	54.95 124	51.73 177	61.88 113	73.78 165	75.11 162	83.99 163	77.91 155

CostFormer		68.92 114	69.58 102	68.15 113	75.98 115	76.17 152	78.22 94	51.86 189	65.80 72	61.56 72	63.57 72	62.83 94	61.85 114	70.40 193	68.67 191	79.42 182	79.62 144

tpm cat1		65.41 151	63.81 181	67.28 130	75.61 119	72.88 173	75.32 108	52.85 183	62.97 89	63.66 68	53.24 154	53.29 160	61.83 115	65.54 205	64.14 208	74.43 202	74.60 177

CANet_DTU		73.29 67	76.96 64	69.00 106	77.04 103	82.06 71	79.49 64	56.30 173	67.85 65	53.29 120	71.12 47	70.37 69	61.81 116	81.59 62	80.96 60	86.09 122	84.73 85

GA-MVS		68.14 122	69.17 114	66.93 137	73.77 153	78.50 114	74.45 119	58.28 163	55.11 165	48.44 146	60.08 81	53.99 146	61.50 117	78.43 110	77.57 111	85.13 147	80.54 131

anonymousdsp		65.28 152	67.98 137	62.13 168	58.73 213	73.98 171	67.10 175	50.69 196	48.41 199	47.66 152	54.27 135	52.75 165	61.45 118	76.71 146	80.20 76	87.13 81	89.53 46

v7n		67.05 145	66.94 149	67.17 131	72.35 163	78.97 101	73.26 143	58.88 151	51.16 189	50.90 133	48.21 185	50.11 186	60.96 119	77.70 120	77.38 116	86.68 109	85.05 80

CR-MVSNet		64.83 158	65.54 166	64.01 158	70.64 177	69.41 183	65.97 182	52.74 184	57.81 125	52.65 123	54.27 135	56.31 123	60.92 120	72.20 175	73.09 171	81.12 175	75.69 171

PatchT		61.97 183	64.04 179	59.55 185	60.49 207	67.40 191	56.54 208	48.65 203	56.69 141	52.65 123	51.10 177	52.14 173	60.92 120	72.20 175	73.09 171	78.03 187	75.69 171

dps		64.00 168	62.99 184	65.18 147	73.29 155	72.07 176	68.98 164	53.07 182	57.74 127	58.41 83	55.55 113	47.74 197	60.89 122	69.53 196	67.14 200	76.44 194	71.19 194

IterMVS		66.36 147	68.30 133	64.10 155	69.48 185	74.61 168	73.41 141	50.79 195	57.30 131	48.28 147	60.64 78	59.92 102	60.85 123	74.14 163	72.66 173	81.80 171	78.82 151
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
TDRefinement		66.09 149	65.03 173	67.31 128	69.73 182	76.75 142	75.33 107	64.55 73	60.28 108	49.72 143	45.63 194	42.83 208	60.46 124	75.75 152	75.95 152	84.08 161	78.04 154

V4		65.23 153	66.23 156	64.06 156	61.94 203	76.42 146	72.05 151	54.31 177	49.91 195	50.06 138	47.42 189	52.40 167	60.24 125	75.71 153	76.22 147	85.78 136	85.56 67

v52		65.23 153	66.24 155	64.06 156	61.94 203	76.42 146	72.06 150	54.30 178	49.94 193	50.04 139	47.41 190	52.42 166	60.23 126	75.71 153	76.22 147	85.78 136	85.56 67

PatchmatchNet		64.21 167	64.65 175	63.69 159	71.29 176	68.66 187	69.63 159	51.70 191	63.04 88	53.77 117	59.83 85	58.34 107	60.23 126	68.54 200	66.06 203	75.56 197	68.08 201
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
gm-plane-assit		57.00 199	57.62 205	56.28 196	76.10 112	62.43 212	47.62 221	46.57 209	33.84 224	23.24 215	37.52 206	40.19 213	59.61 128	79.81 93	77.55 112	84.55 159	72.03 190

IB-MVS	66.94 12	71.21 78	71.66 83	70.68 79	79.18 76	82.83 68	72.61 145	61.77 111	59.66 112	63.44 69	53.26 153	59.65 103	59.16 129	76.78 145	82.11 51	87.90 63	87.33 58
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
UniMVSNet_NR-MVSNet		70.59 82	72.19 80	68.72 108	77.72 96	80.72 83	73.81 134	69.65 39	61.99 94	43.23 173	60.54 79	57.50 109	58.57 130	79.56 98	81.07 59	89.34 41	83.97 90

DU-MVS		69.63 100	70.91 86	68.13 114	75.99 113	79.54 97	73.81 134	69.20 44	61.20 102	43.23 173	58.52 90	53.50 150	58.57 130	79.22 102	80.45 74	87.97 61	83.97 90

COLMAP_ROB	62.73 15	67.66 134	66.76 152	68.70 109	80.49 71	77.98 121	75.29 109	62.95 84	63.62 85	49.96 140	47.32 192	50.72 183	58.57 130	76.87 143	75.50 158	84.94 152	75.33 175
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
PM-MVS		60.48 191	60.94 198	59.94 181	58.85 212	66.83 194	64.27 191	51.39 192	55.03 167	48.03 148	50.00 182	40.79 212	58.26 133	69.20 198	67.13 201	78.84 185	77.60 157

SixPastTwentyTwo		61.84 185	62.45 190	61.12 175	69.20 186	72.20 175	62.03 198	57.40 167	46.54 205	38.03 191	57.14 105	41.72 210	58.12 134	69.67 195	71.58 177	81.94 170	78.30 153

CMPMVS	47.78 17	62.49 177	62.52 189	62.46 166	70.01 180	70.66 181	62.97 195	51.84 190	51.98 184	56.71 101	42.87 197	53.62 147	57.80 135	72.23 173	70.37 182	75.45 199	75.91 168
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
GBi-Net		70.78 79	73.37 75	67.76 115	72.95 158	78.00 118	75.15 111	62.72 92	64.13 82	51.44 128	58.37 93	69.02 75	57.59 136	81.33 67	80.72 64	86.70 106	82.02 115

test1		70.78 79	73.37 75	67.76 115	72.95 158	78.00 118	75.15 111	62.72 92	64.13 82	51.44 128	58.37 93	69.02 75	57.59 136	81.33 67	80.72 64	86.70 106	82.02 115

FMVSNet2		70.39 84	72.67 79	67.72 118	72.95 158	78.00 118	75.15 111	62.69 96	63.29 87	51.25 132	55.64 111	68.49 81	57.59 136	80.91 75	80.35 75	86.70 106	82.02 115

FMVSNet3		70.49 83	72.90 77	67.67 119	72.88 161	77.98 121	74.96 117	62.72 92	64.13 82	51.44 128	58.37 93	69.02 75	57.43 139	79.43 100	79.57 82	86.59 112	81.81 122

MDTV_nov1_ep13		64.37 162	65.24 168	63.37 164	68.94 187	70.81 179	72.40 148	50.29 198	60.10 109	53.91 116	60.07 82	59.15 105	57.21 140	69.43 197	67.30 198	77.47 189	69.78 197

v748		65.12 155	65.24 168	64.98 150	69.77 181	76.45 145	69.47 161	57.06 170	49.93 194	50.70 134	47.87 188	49.50 190	57.14 141	73.64 167	75.18 161	85.75 138	84.14 89

FMVSNet1		68.84 115	70.47 90	66.94 136	71.35 175	77.68 126	74.71 118	62.35 106	56.93 138	49.94 141	50.01 180	64.59 89	57.07 142	81.33 67	80.72 64	86.25 115	82.00 118

pmmvs-eth3d		63.52 170	62.44 191	64.77 152	66.82 193	70.12 182	69.41 162	59.48 145	54.34 173	52.71 122	46.24 193	44.35 207	56.93 143	72.37 170	73.77 168	83.30 165	75.91 168

FC-MVSNet-train		72.60 72	75.07 70	69.71 101	81.10 66	78.79 110	73.74 136	65.23 68	66.10 70	53.34 119	70.36 49	63.40 93	56.92 144	81.44 64	80.96 60	87.93 62	84.46 87

tfpn111		68.38 119	69.23 113	67.39 124	77.83 88	78.93 104	74.28 124	62.81 85	56.64 142	46.70 155	56.24 108	53.47 153	56.59 145	80.41 77	78.43 92	86.11 119	80.53 132

conf0.01		67.72 132	67.99 136	67.39 124	77.82 93	78.94 102	74.28 124	62.81 85	56.64 142	46.70 155	53.33 151	48.59 193	56.59 145	80.34 84	78.43 92	86.16 118	79.67 143

conf0.002		67.52 140	67.64 140	67.39 124	77.80 95	78.94 102	74.28 124	62.81 85	56.64 142	46.70 155	53.65 147	46.28 201	56.59 145	80.33 85	78.37 97	86.17 117	79.23 147

conf200view11		68.11 123	68.72 124	67.39 124	77.83 88	78.93 104	74.28 124	62.81 85	56.64 142	46.70 155	52.65 165	53.47 153	56.59 145	80.41 77	78.43 92	86.11 119	80.53 132

tfpn200view9		68.11 123	68.72 124	67.40 123	77.83 88	78.93 104	74.28 124	62.81 85	56.64 142	46.82 153	52.65 165	53.47 153	56.59 145	80.41 77	78.43 92	86.11 119	80.52 134

thres400		67.95 127	68.62 128	67.17 131	77.90 83	78.59 113	74.27 129	62.72 92	56.34 153	45.77 164	53.00 158	53.35 158	56.46 150	80.21 89	78.43 92	85.91 134	80.43 135

thres200		67.98 126	68.55 129	67.30 129	77.89 85	78.86 108	74.18 131	62.75 90	56.35 152	46.48 160	52.98 159	53.54 149	56.46 150	80.41 77	77.97 103	86.05 125	79.78 142

MVS-HIRNet		54.41 204	52.10 211	57.11 194	58.99 211	56.10 217	49.68 218	49.10 200	46.18 206	52.15 127	33.18 216	46.11 203	56.10 152	63.19 211	59.70 219	76.64 193	60.25 215

Baseline_NR-MVSNet		67.53 139	68.77 122	66.09 144	75.99 113	74.75 167	72.43 147	68.41 47	61.33 101	38.33 189	51.31 175	54.13 145	56.03 153	79.22 102	78.19 100	85.37 144	82.45 113

thres100view900		67.60 138	68.02 135	67.12 133	77.83 88	77.75 125	73.90 132	62.52 102	56.64 142	46.82 153	52.65 165	53.47 153	55.92 154	78.77 107	77.62 110	85.72 139	79.23 147

test-mter		60.84 190	64.62 176	56.42 195	55.99 220	64.18 200	65.39 184	34.23 228	54.39 172	46.21 161	57.40 103	59.49 104	55.86 155	71.02 185	69.65 184	80.87 177	76.20 167

tpmrst		62.00 182	62.35 192	61.58 172	71.62 171	64.14 201	69.07 163	48.22 207	62.21 93	53.93 115	58.26 97	55.30 134	55.81 156	63.22 210	62.62 211	70.85 212	70.70 195

TranMVSNet+NR-MVSNet		69.25 111	70.81 87	67.43 122	77.23 102	79.46 99	73.48 140	69.66 38	60.43 107	39.56 185	58.82 89	53.48 152	55.74 157	79.59 96	81.21 58	88.89 49	82.70 111

thres600view7		67.68 133	68.43 130	66.80 138	77.90 83	78.86 108	73.84 133	62.75 90	56.07 155	44.70 170	52.85 162	52.81 163	55.58 158	80.41 77	77.77 106	86.05 125	80.28 136

Vis-MVSNet		72.77 71	77.20 62	67.59 121	74.19 147	84.01 58	76.61 106	61.69 112	60.62 106	50.61 136	70.25 50	71.31 64	55.57 159	83.85 46	82.28 49	86.90 91	88.08 52
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
EG-PatchMatch MVS		67.24 143	66.94 149	67.60 120	78.73 79	81.35 74	73.28 142	59.49 144	46.89 204	51.42 131	43.65 196	53.49 151	55.50 160	81.38 66	80.66 70	87.15 77	81.17 126

view600		67.63 137	68.36 131	66.77 139	77.84 87	78.66 111	73.74 136	62.62 99	56.04 156	44.98 167	52.86 161	52.83 162	55.48 161	80.36 83	77.75 107	85.95 133	80.02 139

test-LLR		64.42 161	64.36 177	64.49 154	75.02 123	63.93 202	66.61 179	61.96 108	54.41 170	47.77 149	57.46 101	60.25 100	55.20 162	70.80 187	69.33 186	80.40 178	74.38 179

TESTMET0.1,1		61.10 189	64.36 177	57.29 192	57.53 215	63.93 202	66.61 179	36.22 226	54.41 170	47.77 149	57.46 101	60.25 100	55.20 162	70.80 187	69.33 186	80.40 178	74.38 179

UA-Net		74.47 63	77.80 55	70.59 82	85.33 47	85.40 51	73.54 138	65.98 64	60.65 105	56.00 106	72.11 43	79.15 39	54.63 164	83.13 54	82.25 50	88.04 60	81.92 121

IS_MVSNet		73.33 66	77.34 61	68.65 110	81.29 63	83.47 62	74.45 119	63.58 78	65.75 73	48.49 145	67.11 64	70.61 66	54.63 164	84.51 41	83.58 45	89.48 39	86.34 63

view800		67.35 142	68.22 134	66.35 143	77.83 88	78.62 112	72.97 144	62.58 100	55.71 158	44.13 171	52.69 164	52.24 172	54.58 166	80.27 87	78.19 100	86.01 128	79.79 141

tfpn		66.58 146	67.18 146	65.88 145	77.82 93	78.45 115	72.07 149	62.52 102	55.35 162	43.21 175	52.54 169	46.12 202	53.68 167	80.02 91	78.23 99	85.99 131	79.55 145

tpm		62.41 178	63.15 183	61.55 173	72.24 164	63.79 204	71.31 154	46.12 211	57.82 124	55.33 108	59.90 84	54.74 139	53.63 168	67.24 203	64.29 206	70.65 213	74.25 181

conf0.05thres1000		66.26 148	66.77 151	65.66 146	77.45 100	78.10 116	71.85 152	62.44 105	51.47 188	43.00 176	47.92 187	51.66 178	53.40 169	79.71 94	77.97 103	85.82 135	80.56 130

MDTV_nov1_ep13_2view		60.16 192	60.51 199	59.75 182	65.39 196	69.05 186	68.00 167	48.29 205	51.99 183	45.95 163	48.01 186	49.64 189	53.39 170	68.83 199	66.52 202	77.47 189	69.55 198

UniMVSNet (Re)		69.53 105	71.90 81	66.76 140	76.42 106	80.93 79	72.59 146	68.03 51	61.75 98	41.68 181	58.34 96	57.23 117	53.27 171	79.53 99	80.62 72	88.57 53	84.90 83

RPMNet		61.71 188	62.88 185	60.34 179	69.51 184	69.41 183	63.48 193	49.23 199	57.81 125	45.64 165	50.51 178	50.12 185	53.13 172	68.17 202	68.49 194	81.07 176	75.62 173

tfpnnormal		64.27 164	63.64 182	65.02 149	75.84 116	75.61 155	71.24 155	62.52 102	47.79 201	42.97 177	42.65 198	44.49 206	52.66 173	78.77 107	76.86 124	84.88 153	79.29 146

MDA-MVSNet-bldmvs		53.37 208	53.01 210	53.79 205	43.67 230	67.95 190	59.69 204	57.92 164	43.69 208	32.41 202	41.47 200	27.89 228	52.38 174	56.97 224	65.99 204	76.68 192	67.13 202

NR-MVSNet		68.79 116	70.56 88	66.71 142	77.48 99	79.54 97	73.52 139	69.20 44	61.20 102	39.76 184	58.52 90	50.11 186	51.37 175	80.26 88	80.71 68	88.97 47	83.59 97

EPMVS		60.00 193	61.97 193	57.71 191	68.46 188	63.17 208	64.54 189	48.23 206	63.30 86	44.72 169	60.19 80	56.05 131	50.85 176	65.27 207	62.02 213	69.44 215	63.81 208

CVMVSNet		62.55 175	65.89 162	58.64 188	66.95 191	69.15 185	66.49 181	56.29 174	52.46 182	32.70 201	59.27 87	58.21 108	50.09 177	71.77 178	71.39 178	79.31 183	78.99 150

thresconf0.02		64.77 159	65.90 161	63.44 162	76.37 107	75.17 166	69.51 160	61.28 114	56.98 133	39.01 187	56.24 108	48.68 192	49.78 178	77.13 139	75.61 155	84.71 156	71.53 192

pmmvs5		62.37 181	64.04 179	60.42 178	65.03 197	71.67 178	67.17 174	52.70 186	50.30 190	44.80 168	54.23 139	51.19 181	49.37 179	72.88 169	73.48 170	83.45 164	74.55 178

UGNet		72.78 70	77.67 56	67.07 134	71.65 170	83.24 64	75.20 110	63.62 77	64.93 77	56.72 100	71.82 45	73.30 56	49.02 180	81.02 73	80.70 69	86.22 116	88.67 50
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
pm-mvs1		65.62 150	67.42 143	63.53 161	73.66 154	76.39 148	69.66 158	60.87 121	49.73 196	43.97 172	51.24 176	57.00 119	48.16 181	79.89 92	77.84 105	84.85 155	79.82 140

gg-mvs-nofinetune		62.55 175	65.05 172	59.62 184	78.72 80	77.61 127	70.83 156	53.63 179	39.71 216	22.04 221	36.36 209	64.32 90	47.53 182	81.16 71	79.03 88	85.00 150	77.17 160

tfpnview11		64.33 163	66.17 157	62.18 167	76.25 108	75.23 161	67.45 170	61.16 115	55.50 160	36.38 194	55.35 115	51.89 174	46.96 183	77.28 136	76.10 151	84.86 154	71.85 191

tfpn_n400		64.23 165	66.05 158	62.12 169	76.20 109	75.24 159	67.43 171	61.15 116	54.04 175	36.38 194	55.35 115	51.89 174	46.94 184	77.31 134	76.15 149	84.59 157	72.36 188

tfpnconf		64.23 165	66.05 158	62.12 169	76.20 109	75.24 159	67.43 171	61.15 116	54.04 175	36.38 194	55.35 115	51.89 174	46.94 184	77.31 134	76.15 149	84.59 157	72.36 188

pmmvs6		62.41 178	62.88 185	61.87 171	71.38 174	75.18 165	67.76 169	59.45 146	41.64 212	42.52 180	37.33 207	52.91 161	46.87 186	77.67 122	76.26 145	83.23 166	79.18 149

tfpn_ndepth		65.09 156	67.12 147	62.73 165	75.75 118	76.23 150	68.00 167	60.36 128	58.16 122	40.27 183	54.89 125	54.22 142	46.80 187	76.69 147	75.66 154	85.19 146	73.98 183

LP		53.62 207	53.43 207	53.83 204	58.51 214	62.59 211	57.31 207	46.04 212	47.86 200	42.69 179	36.08 211	36.86 216	46.53 188	64.38 208	64.25 207	71.92 209	62.00 213

ADS-MVSNet		55.94 202	58.01 202	53.54 206	62.48 202	58.48 214	59.12 206	46.20 210	59.65 113	42.88 178	52.34 171	53.31 159	46.31 189	62.00 214	60.02 218	64.23 224	60.24 216

pmmvs3		47.65 212	49.08 216	45.99 214	44.61 227	54.79 220	50.04 216	31.95 231	33.91 223	29.90 203	30.37 217	33.53 219	46.31 189	63.50 209	63.67 209	73.14 207	63.77 209

TransMVSNet (Re)		64.74 160	65.66 165	63.66 160	77.40 101	75.33 158	69.86 157	62.67 98	47.63 202	41.21 182	50.01 180	52.33 168	45.31 191	79.57 97	77.69 109	85.49 142	77.07 163

CDS-MVSNet		67.65 135	69.83 99	65.09 148	75.39 120	76.55 144	74.42 122	63.75 76	53.55 177	49.37 144	59.41 86	62.45 95	44.44 192	79.71 94	79.82 77	83.17 167	77.36 159
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
TAMVS		59.58 194	62.81 187	55.81 197	66.03 195	65.64 199	63.86 192	48.74 202	49.95 192	37.07 193	54.77 128	58.54 106	44.44 192	72.29 172	71.79 175	74.70 201	66.66 203

EPNet_dtu		68.08 125	71.00 85	64.67 153	79.64 73	68.62 188	75.05 115	63.30 79	66.36 68	45.27 166	67.40 62	66.84 84	43.64 194	75.37 156	74.98 164	81.15 174	77.44 158
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
tfpn1000		63.81 169	66.31 154	60.90 176	75.76 117	75.74 154	65.14 186	60.14 138	56.47 149	35.99 197	55.11 118	52.30 170	43.42 195	76.21 151	75.34 159	84.97 151	73.01 187

FMVSNet5		57.24 198	60.02 200	53.99 203	56.45 217	62.74 209	65.27 185	47.03 208	55.14 164	39.55 186	40.88 202	53.42 157	41.83 196	72.35 171	71.10 180	73.79 204	64.50 207

ambc			53.42 208		64.99 198	63.36 206	49.96 217		47.07 203	37.12 192	28.97 219	16.36 234	41.82 197	75.10 159	67.34 197	71.55 211	75.72 170

FPMVS		51.87 209	50.00 214	54.07 202	66.83 192	57.25 215	60.25 203	50.91 193	50.25 191	34.36 199	36.04 212	32.02 220	41.49 198	58.98 222	56.07 222	70.56 214	59.36 217

CP-MVSNet		62.68 174	65.49 167	59.40 186	71.84 166	75.34 157	62.87 196	67.04 57	52.64 180	27.19 209	53.38 150	48.15 195	41.40 199	71.26 180	75.68 153	86.07 123	82.00 118

PS-CasMVS		62.38 180	65.06 171	59.25 187	71.73 167	75.21 164	62.77 197	66.99 58	51.94 186	26.96 210	52.00 172	47.52 198	41.06 200	71.16 183	75.60 156	85.97 132	81.97 120

PEN-MVS		62.96 172	65.77 164	59.70 183	73.98 151	75.45 156	63.39 194	67.61 54	52.49 181	25.49 211	53.39 149	49.12 191	40.85 201	71.94 177	77.26 119	86.86 96	80.72 129

MIMVSNet		58.52 197	61.34 196	55.22 199	60.76 206	67.01 193	66.81 176	49.02 201	56.43 150	38.90 188	40.59 204	54.54 141	40.57 202	73.16 168	71.65 176	75.30 200	66.00 204

Vis-MVSNet (Re-imp)		67.83 130	73.52 73	61.19 174	78.37 81	76.72 143	66.80 177	62.96 83	65.50 74	34.17 200	67.19 63	69.68 71	39.20 203	79.39 101	79.44 86	85.68 140	76.73 166

DTE-MVSNet		61.85 184	64.96 174	58.22 189	74.32 146	74.39 169	61.01 200	67.85 53	51.76 187	21.91 222	53.28 152	48.17 194	37.74 204	72.22 174	76.44 143	86.52 114	78.49 152

EU-MVSNet		54.63 203	58.69 201	49.90 210	56.99 216	62.70 210	56.41 209	50.64 197	45.95 207	23.14 216	50.42 179	46.51 200	36.63 205	65.51 206	64.85 205	75.57 196	74.91 176

Anonymous20231211		51.46 210	50.59 212	52.46 207	67.30 189	66.70 195	55.00 210	59.22 147	29.96 227	17.62 227	19.11 229	28.74 227	35.72 206	66.42 204	69.52 185	79.92 180	73.71 184

Anonymous20231206		56.36 201	57.80 204	54.67 201	70.08 179	66.39 196	60.46 202	57.54 165	49.50 198	29.30 205	33.86 215	46.64 199	35.18 207	70.44 191	68.88 190	75.47 198	68.88 200

WR-MVS		63.03 171	67.40 144	57.92 190	75.14 122	77.60 128	60.56 201	66.10 61	54.11 174	23.88 212	53.94 145	53.58 148	34.50 208	73.93 164	77.71 108	87.35 73	80.94 127

WR-MVS_H		61.83 186	65.87 163	57.12 193	71.72 168	76.87 141	61.45 199	66.19 59	51.97 185	22.92 219	53.13 157	52.30 170	33.80 209	71.03 184	75.00 163	86.65 110	80.78 128

PMVS	39.38 18	46.06 218	43.30 223	49.28 211	62.93 200	38.75 231	41.88 224	53.50 180	33.33 226	35.46 198	28.90 220	31.01 223	33.04 210	58.61 223	54.63 225	68.86 216	57.88 220
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
test0.0.03 1		58.80 195	61.58 195	55.56 198	75.02 123	68.45 189	59.58 205	61.96 108	52.74 179	29.57 204	49.75 183	54.56 140	31.46 211	71.19 181	69.77 183	75.75 195	64.57 206

testpf		47.41 213	48.47 219	46.18 213	66.30 194	50.67 224	48.15 220	42.60 221	37.10 220	28.75 206	40.97 201	39.01 215	30.82 212	52.95 227	53.74 226	60.46 225	64.87 205

N_pmnet		47.35 214	50.13 213	44.11 217	59.98 208	51.64 223	51.86 214	44.80 216	49.58 197	20.76 223	40.65 203	40.05 214	29.64 213	59.84 220	55.15 223	57.63 226	54.00 224

test2356		47.20 215	48.62 218	45.54 215	56.38 218	54.89 219	50.62 215	45.08 215	38.65 217	23.40 214	36.23 210	31.10 222	29.31 214	62.76 212	62.49 212	68.48 217	54.23 223

1111		43.08 220	44.02 222	41.98 219	59.22 209	49.27 227	41.48 225	45.63 213	35.01 221	23.06 217	28.60 221	30.15 224	27.22 215	60.42 218	57.97 220	55.27 229	46.74 226

.test1245		30.81 227	29.14 229	32.77 226	59.22 209	49.27 227	41.48 225	45.63 213	35.01 221	23.06 217	28.60 221	30.15 224	27.22 215	60.42 218	0.10 233	0.01 237	0.43 235

DeepMVS_CX						18.74 237	18.55 234	8.02 233	26.96 230	7.33 234	23.81 228	13.05 236	25.99 217	25.17 232		22.45 236	36.25 231

MIMVSNet1		49.27 211	53.25 209	44.62 216	44.61 227	61.52 213	53.61 212	52.18 187	41.62 213	18.68 224	28.14 223	41.58 211	25.50 218	68.46 201	69.04 188	73.15 206	62.37 212

Gipuma		36.38 224	35.80 227	37.07 223	45.76 226	33.90 232	29.81 231	48.47 204	39.91 215	18.02 226	8.00 235	8.14 237	25.14 219	59.29 221	61.02 216	55.19 230	40.31 228
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
testgi		54.39 205	57.86 203	50.35 209	71.59 173	67.24 192	54.95 211	53.25 181	43.36 209	23.78 213	44.64 195	47.87 196	24.96 220	70.45 190	68.66 192	73.60 205	62.78 211

new_pmnet		38.40 223	42.64 224	33.44 225	37.54 233	45.00 229	36.60 229	32.72 230	40.27 214	12.72 231	29.89 218	28.90 226	24.78 221	53.17 226	52.90 227	56.31 227	48.34 225

testus		45.61 219	49.06 217	41.59 220	56.13 219	55.28 218	43.51 223	39.64 224	37.74 218	18.23 225	35.52 214	31.28 221	24.69 222	62.46 213	62.90 210	67.33 219	58.26 219

FC-MVSNet-test		56.90 200	65.20 170	47.21 212	66.98 190	63.20 207	49.11 219	58.60 161	59.38 115	11.50 232	65.60 66	56.68 120	24.66 223	71.17 182	71.36 179	72.38 208	69.02 199

test20.03		53.93 206	56.28 206	51.19 208	72.19 165	65.83 197	53.20 213	61.08 118	42.74 210	22.08 220	37.07 208	45.76 204	24.29 224	70.44 191	69.04 188	74.31 203	63.05 210

testmv		42.58 221	44.36 220	40.49 221	54.63 223	52.76 221	41.21 227	44.37 217	28.83 228	12.87 229	27.16 224	25.03 229	23.01 225	60.83 216	61.13 214	66.88 220	54.81 221

test1235678		42.57 222	44.36 220	40.49 221	54.63 223	52.75 222	41.21 227	44.37 217	28.82 229	12.87 229	27.15 225	25.01 230	23.01 225	60.83 216	61.13 214	66.88 220	54.81 221

EMVS		20.98 230	17.15 232	25.44 229	39.51 232	19.37 236	12.66 235	39.59 225	19.10 234	6.62 237	9.27 233	4.40 239	22.43 227	17.99 234	24.40 232	31.81 233	25.53 233

E-PMN		21.77 229	18.24 231	25.89 228	40.22 231	19.58 235	12.46 236	39.87 223	18.68 235	6.71 236	9.57 232	4.31 240	22.36 228	19.89 233	27.28 231	33.73 232	28.34 232

new-patchmatchnet		46.97 216	49.47 215	44.05 218	62.82 201	56.55 216	45.35 222	52.01 188	42.47 211	17.04 228	35.73 213	35.21 217	21.84 229	61.27 215	54.83 224	65.26 223	60.26 214

no-one		36.35 225	37.59 226	34.91 224	46.13 225	49.89 226	27.99 232	43.56 219	20.91 233	7.03 235	14.64 231	15.50 235	18.92 230	42.95 228	60.20 217	65.84 222	59.03 218

MVE	19.12 19	20.47 231	23.27 230	17.20 232	12.66 236	25.41 234	10.52 237	34.14 229	14.79 236	6.53 238	8.79 234	4.68 238	16.64 231	29.49 231	41.63 228	22.73 235	38.11 229
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
test12356		35.10 226	38.50 225	31.13 227	44.14 229	43.70 230	32.27 230	34.42 227	26.51 231	9.47 233	25.22 227	20.34 231	10.86 232	53.47 225	56.15 221	55.59 228	44.11 227

PMMVS2		25.60 228	29.75 228	20.76 231	28.00 234	30.93 233	23.10 233	29.18 232	23.14 232	1.46 239	18.23 230	16.54 233	5.08 233	40.22 229	41.40 229	37.76 231	37.79 230

tmp_tt				14.50 233	14.68 235	7.17 238	10.46 238	2.21 234	37.73 219	28.71 207	25.26 226	16.98 232	4.37 234	31.49 230	29.77 230	26.56 234

GG-mvs-BLEND		46.86 217	67.51 142	22.75 230	0.05 237	76.21 151	64.69 188	0.04 235	61.90 95	0.09 240	55.57 112	71.32 63	0.08 235	70.54 189	67.19 199	71.58 210	69.86 196

test123		0.09 232	0.14 234	0.02 234	0.00 239	0.02 239	0.02 241	0.01 236	0.09 238	0.00 242	0.30 236	0.00 242	0.08 235	0.03 236	0.09 235	0.01 237	0.45 234

testmvs		0.09 232	0.15 233	0.02 234	0.01 238	0.02 239	0.05 240	0.01 236	0.11 237	0.01 241	0.26 237	0.01 241	0.06 237	0.10 235	0.10 233	0.01 237	0.43 235

sosnet-low-res		0.00 234	0.00 235	0.00 236	0.00 239	0.00 241	0.00 242	0.00 238	0.00 239	0.00 242	0.00 238	0.00 242	0.00 238	0.00 237	0.00 236	0.00 240	0.00 237

sosnet		0.00 234	0.00 235	0.00 236	0.00 239	0.00 241	0.00 242	0.00 238	0.00 239	0.00 242	0.00 238	0.00 242	0.00 238	0.00 237	0.00 236	0.00 240	0.00 237

MTAPA										83.48 1		86.45 12

MTMP										82.66 3		84.91 20

Patchmatch-RL test							2.85 239

XVS					86.63 39	88.68 23	85.00 41			71.81 40		81.92 30				90.47 17

X-MVStestdata					86.63 39	88.68 23	85.00 41			71.81 40		81.92 30				90.47 17

mPP-MVS					89.90 20							81.29 35

NP-MVS									80.10 40

Patchmtry						65.80 198	65.97 182	52.74 184		52.65 123

High-resolution multi-view benchmark