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.

ESAPD		88.46 1	91.07 1	85.41 1	91.73 3	92.08 1	91.91 2	76.73 1	90.14 4	80.33 8	92.75 1	90.44 1	80.73 3	88.97 5	87.63 9	91.01 6	95.48 2

APDe-MVS		88.00 2	90.50 2	85.08 2	90.95 6	91.58 4	92.03 1	75.53 8	91.15 1	80.10 9	92.27 3	88.34 7	80.80 2	88.00 11	86.99 15	91.09 4	95.16 3

HPM-MVS++		87.09 5	88.92 9	84.95 3	92.61 1	87.91 35	90.23 10	76.06 3	88.85 8	81.20 4	87.33 9	87.93 8	79.47 6	88.59 6	88.23 5	90.15 29	93.60 16

APD-MVS		86.84 8	88.91 10	84.41 4	90.66 9	90.10 8	90.78 3	75.64 5	87.38 13	78.72 14	90.68 6	86.82 11	80.15 4	87.13 21	86.45 24	90.51 16	93.83 10
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
SMA-MVS		87.48 3	90.13 4	84.39 5	91.76 2	90.70 5	90.63 4	75.36 9	90.51 3	79.89 10	85.65 15	88.82 4	77.90 14	90.00 1	89.77 1	90.82 7	95.49 1

zzz-MVS		85.71 13	86.88 19	84.34 6	90.54 13	87.11 39	89.77 13	74.17 14	88.54 9	83.08 2	78.60 28	86.10 14	78.11 11	87.80 14	87.46 11	90.35 25	92.56 22

CNVR-MVS		86.36 10	88.19 13	84.23 7	91.33 5	89.84 10	90.34 7	75.56 6	87.36 14	78.97 13	81.19 24	86.76 12	78.74 7	89.30 3	88.58 2	90.45 22	94.33 6

HSP-MVS		87.45 4	90.22 3	84.22 8	90.00 19	91.80 3	90.59 5	75.80 4	89.93 5	78.35 16	92.54 2	89.18 3	80.89 1	87.99 12	86.29 26	89.70 36	93.85 9

TSAR-MVS + MP.		86.88 7	89.23 6	84.14 9	89.78 22	88.67 27	90.59 5	73.46 22	88.99 7	80.52 7	91.26 4	88.65 5	79.91 5	86.96 26	86.22 27	90.59 14	93.83 10

HFP-MVS		86.15 11	87.95 14	84.06 10	90.80 7	89.20 19	89.62 15	74.26 12	87.52 11	80.63 6	86.82 12	84.19 24	78.22 10	87.58 15	87.19 13	90.81 8	93.13 20

SD-MVS		86.96 6	89.45 5	84.05 11	90.13 16	89.23 18	89.77 13	74.59 10	89.17 6	80.70 5	89.93 7	89.67 2	78.47 8	87.57 16	86.79 18	90.67 13	93.76 12

NCCC		85.34 16	86.59 21	83.88 12	91.48 4	88.88 21	89.79 12	75.54 7	86.67 17	77.94 19	76.55 31	84.99 20	78.07 12	88.04 9	87.68 8	90.46 21	93.31 17

ACMMP_Plus		86.52 9	89.01 7	83.62 13	90.28 15	90.09 9	90.32 8	74.05 16	88.32 10	79.74 11	87.04 11	85.59 18	76.97 25	89.35 2	88.44 4	90.35 25	94.27 7

MCST-MVS		85.13 19	86.62 20	83.39 14	90.55 12	89.82 12	89.29 17	73.89 19	84.38 27	76.03 24	79.01 27	85.90 16	78.47 8	87.81 13	86.11 29	92.11 1	93.29 18

DeepC-MVS	78.47 2	84.81 22	86.03 25	83.37 15	89.29 27	90.38 7	88.61 22	76.50 2	86.25 19	77.22 20	75.12 35	80.28 39	77.59 18	88.39 7	88.17 6	91.02 5	93.66 14
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
MP-MVS		85.50 15	87.40 17	83.28 16	90.65 10	89.51 15	89.16 19	74.11 15	83.70 29	78.06 18	85.54 16	84.89 22	77.31 20	87.40 18	87.14 14	90.41 23	93.65 15

SteuartSystems-ACMMP		85.99 12	88.31 12	83.27 17	90.73 8	89.84 10	90.27 9	74.31 11	84.56 26	75.88 25	87.32 10	85.04 19	77.31 20	89.01 4	88.46 3	91.14 3	93.96 8
Skip Steuart: Steuart Systems R&D Blog.
ACMMPR		85.52 14	87.53 16	83.17 18	90.13 16	89.27 16	89.30 16	73.97 17	86.89 16	77.14 21	86.09 13	83.18 27	77.74 16	87.42 17	87.20 12	90.77 9	92.63 21

DeepC-MVS_fast	78.24 3	84.27 25	85.50 27	82.85 19	90.46 14	89.24 17	87.83 28	74.24 13	84.88 22	76.23 23	75.26 34	81.05 37	77.62 17	88.02 10	87.62 10	90.69 12	92.41 24
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
CP-MVS		84.74 23	86.43 23	82.77 20	89.48 25	88.13 34	88.64 21	73.93 18	84.92 21	76.77 22	81.94 22	83.50 25	77.29 22	86.92 27	86.49 23	90.49 17	93.14 19

CSCG		85.28 18	87.68 15	82.49 21	89.95 20	91.99 2	88.82 20	71.20 32	86.41 18	79.63 12	79.26 25	88.36 6	73.94 35	86.64 28	86.67 21	91.40 2	94.41 4

PGM-MVS		84.42 24	86.29 24	82.23 22	90.04 18	88.82 23	89.23 18	71.74 30	82.82 32	74.61 28	84.41 19	82.09 29	77.03 24	87.13 21	86.73 20	90.73 11	92.06 28

train_agg		84.86 21	87.21 18	82.11 23	90.59 11	85.47 50	89.81 11	73.55 21	83.95 28	73.30 33	89.84 8	87.23 10	75.61 28	86.47 30	85.46 34	89.78 32	92.06 28

3Dnovator+	75.73 4	82.40 30	82.76 35	81.97 24	88.02 32	89.67 13	86.60 32	71.48 31	81.28 38	78.18 17	64.78 70	77.96 45	77.13 23	87.32 19	86.83 17	90.41 23	91.48 32

MSLP-MVS++		82.09 32	82.66 36	81.42 25	87.03 38	87.22 38	85.82 37	70.04 37	80.30 39	78.66 15	68.67 57	81.04 38	77.81 15	85.19 39	84.88 39	89.19 45	91.31 33

ACMMP		83.42 27	85.27 28	81.26 26	88.47 31	88.49 30	88.31 26	72.09 27	83.42 30	72.77 36	82.65 20	78.22 43	75.18 29	86.24 33	85.76 31	90.74 10	92.13 27
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
DeepPCF-MVS	79.04 1	85.30 17	88.93 8	81.06 27	88.77 30	90.48 6	85.46 41	73.08 23	90.97 2	73.77 32	84.81 18	85.95 15	77.43 19	88.22 8	87.73 7	87.85 67	94.34 5

AdaColmap		79.74 42	78.62 53	81.05 28	89.23 28	86.06 47	84.95 44	71.96 28	79.39 43	75.51 26	63.16 74	68.84 80	76.51 26	83.55 50	82.85 49	88.13 59	86.46 63

X-MVS		83.23 28	85.20 29	80.92 29	89.71 23	88.68 24	88.21 27	73.60 20	82.57 33	71.81 41	77.07 29	81.92 31	71.72 50	86.98 25	86.86 16	90.47 18	92.36 25

TSAR-MVS + ACMM		85.10 20	88.81 11	80.77 30	89.55 24	88.53 29	88.59 23	72.55 25	87.39 12	71.90 38	90.95 5	87.55 9	74.57 30	87.08 23	86.54 22	87.47 72	93.67 13

TSAR-MVS + GP.		83.69 26	86.58 22	80.32 31	85.14 49	86.96 40	84.91 45	70.25 36	84.71 25	73.91 31	85.16 17	85.63 17	77.92 13	85.44 36	85.71 32	89.77 33	92.45 23

CPTT-MVS		81.77 33	83.10 34	80.21 32	85.93 45	86.45 45	87.72 29	70.98 33	82.54 34	71.53 44	74.23 40	81.49 34	76.31 27	82.85 57	81.87 53	88.79 52	92.26 26

OPM-MVS		79.68 43	79.28 52	80.15 33	87.99 33	86.77 42	88.52 24	72.72 24	64.55 82	67.65 55	67.87 61	74.33 56	74.31 33	86.37 32	85.25 36	89.73 35	89.81 45

CDPH-MVS		82.64 29	85.03 30	79.86 34	89.41 26	88.31 31	88.32 25	71.84 29	80.11 40	67.47 56	82.09 21	81.44 35	71.85 48	85.89 35	86.15 28	90.24 27	91.25 34

ACMM	72.26 8	78.86 50	78.13 54	79.71 35	86.89 39	83.40 64	86.02 35	70.50 34	75.28 50	71.49 45	63.01 75	69.26 74	73.57 37	84.11 45	83.98 43	89.76 34	87.84 55
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
CANet		81.62 35	83.41 32	79.53 36	87.06 37	88.59 28	85.47 40	67.96 53	76.59 48	74.05 29	74.69 36	81.98 30	72.98 41	86.14 34	85.47 33	89.68 37	90.42 42

MVS_0304		81.73 34	83.86 31	79.26 37	86.22 44	89.18 20	86.41 33	67.15 57	75.28 50	70.75 48	74.59 37	83.49 26	74.42 32	87.05 24	86.34 25	90.58 15	91.08 36

abl_6				79.05 38	87.27 36	88.85 22	83.62 50	68.25 49	81.68 36	72.94 35	73.79 41	84.45 23	72.55 43			89.66 38	90.64 39

ACMP	73.23 7	79.79 40	80.53 45	78.94 39	85.61 47	85.68 48	85.61 38	69.59 41	77.33 46	71.00 47	74.45 38	69.16 75	71.88 46	83.15 54	83.37 47	89.92 31	90.57 41
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
3Dnovator	73.76 5	79.75 41	80.52 46	78.84 40	84.94 54	87.35 36	84.43 47	65.54 67	78.29 44	73.97 30	63.00 76	75.62 52	74.07 34	85.00 40	85.34 35	90.11 30	89.04 48

HQP-MVS		81.19 36	83.27 33	78.76 41	87.40 35	85.45 51	86.95 30	70.47 35	81.31 37	66.91 59	79.24 26	76.63 47	71.67 51	84.43 43	83.78 44	89.19 45	92.05 30

MVS_111021_HR		80.13 38	81.46 40	78.58 42	85.77 46	85.17 54	83.45 51	69.28 44	74.08 56	70.31 49	74.31 39	75.26 53	73.13 39	86.46 31	85.15 37	89.53 39	89.81 45

PCF-MVS	73.28 6	79.42 44	80.41 47	78.26 43	84.88 55	88.17 32	86.08 34	69.85 38	75.23 52	68.43 51	68.03 60	78.38 42	71.76 49	81.26 71	80.65 72	88.56 55	91.18 35
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
PHI-MVS		82.36 31	85.89 26	78.24 44	86.40 42	89.52 14	85.52 39	69.52 43	82.38 35	65.67 61	81.35 23	82.36 28	73.07 40	87.31 20	86.76 19	89.24 43	91.56 31

LGP-MVS_train		79.83 39	81.22 42	78.22 45	86.28 43	85.36 53	86.76 31	69.59 41	77.34 45	65.14 63	75.68 33	70.79 66	71.37 53	84.60 41	84.01 42	90.18 28	90.74 38

MAR-MVS		79.21 46	80.32 48	77.92 46	87.46 34	88.15 33	83.95 48	67.48 56	74.28 54	68.25 52	64.70 71	77.04 46	72.17 45	85.42 37	85.00 38	88.22 56	87.62 57
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
OMC-MVS		80.26 37	82.59 37	77.54 47	83.04 57	85.54 49	83.25 52	65.05 70	87.32 15	72.42 37	72.04 45	78.97 41	73.30 38	83.86 46	81.60 56	88.15 58	88.83 50

EPNet		79.08 49	80.62 44	77.28 48	88.90 29	83.17 67	83.65 49	72.41 26	74.41 53	67.15 58	76.78 30	74.37 55	64.43 101	83.70 49	83.69 45	87.15 78	88.19 52
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
CLD-MVS		79.35 45	81.23 41	77.16 49	85.01 52	86.92 41	85.87 36	60.89 121	80.07 42	75.35 27	72.96 42	73.21 59	68.43 66	85.41 38	84.63 40	87.41 73	85.44 74
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
DELS-MVS		79.15 48	81.07 43	76.91 50	83.54 56	87.31 37	84.45 46	64.92 71	69.98 61	69.34 50	71.62 47	76.26 48	69.84 58	86.57 29	85.90 30	89.39 41	89.88 44
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
CNLPA		77.20 55	77.54 58	76.80 51	82.63 59	84.31 58	79.77 62	64.64 72	85.17 20	73.18 34	56.37 108	69.81 71	74.53 31	81.12 73	78.69 91	86.04 128	87.29 60

QAPM		78.47 51	80.22 49	76.43 52	85.03 51	86.75 43	80.62 58	66.00 64	73.77 57	65.35 62	65.54 68	78.02 44	72.69 42	83.71 48	83.36 48	88.87 51	90.41 43

MVS_111021_LR		78.13 53	79.85 51	76.13 53	81.12 66	81.50 74	80.28 59	65.25 68	76.09 49	71.32 46	76.49 32	72.87 60	72.21 44	82.79 58	81.29 58	86.59 113	87.91 54

PLC	68.99 11	75.68 60	75.31 70	76.12 54	82.94 58	81.26 77	79.94 61	66.10 62	77.15 47	66.86 60	59.13 89	68.53 81	73.73 36	80.38 83	79.04 88	87.13 82	81.68 124
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
canonicalmvs		79.16 47	82.37 38	75.41 55	82.33 62	86.38 46	80.80 57	63.18 81	82.90 31	67.34 57	72.79 43	76.07 49	69.62 59	83.46 53	84.41 41	89.20 44	90.60 40

OpenMVS	70.44 10	76.15 59	76.82 66	75.37 56	85.01 52	84.79 56	78.99 71	62.07 108	71.27 60	67.88 54	57.91 100	72.36 61	70.15 57	82.23 60	81.41 57	88.12 60	87.78 56

PVSNet_Blended_VisFu		76.57 56	77.90 55	75.02 57	80.56 70	86.58 44	79.24 66	66.18 61	64.81 79	68.18 53	65.61 66	71.45 63	67.05 68	84.16 44	81.80 54	88.90 49	90.92 37

TAPA-MVS	71.42 9	77.69 54	80.05 50	74.94 58	80.68 69	84.52 57	81.36 54	63.14 82	84.77 23	64.82 65	68.72 55	75.91 51	71.86 47	81.62 62	79.55 84	87.80 68	85.24 77
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
LS3D		74.08 65	73.39 75	74.88 59	85.05 50	82.62 70	79.71 63	68.66 47	72.82 58	58.80 82	57.61 101	61.31 99	71.07 55	80.32 87	78.87 90	86.00 131	80.18 138

PVSNet_BlendedMVS		76.21 57	77.52 59	74.69 60	79.46 75	83.79 61	77.50 100	64.34 75	69.88 62	71.88 39	68.54 58	70.42 68	67.05 68	83.48 51	79.63 80	87.89 65	86.87 61

PVSNet_Blended		76.21 57	77.52 59	74.69 60	79.46 75	83.79 61	77.50 100	64.34 75	69.88 62	71.88 39	68.54 58	70.42 68	67.05 68	83.48 51	79.63 80	87.89 65	86.87 61

TSAR-MVS + COLMAP		78.34 52	81.64 39	74.48 62	80.13 73	85.01 55	81.73 53	65.93 66	84.75 24	61.68 72	85.79 14	66.27 86	71.39 52	82.91 56	80.78 63	86.01 129	85.98 65

Effi-MVS+		75.28 62	76.20 67	74.20 63	81.15 65	83.24 65	81.11 55	63.13 83	66.37 68	60.27 77	64.30 72	68.88 79	70.93 56	81.56 64	81.69 55	88.61 53	87.35 58

DI_MVS_plusplus_trai		75.13 63	76.12 68	73.96 64	78.18 83	81.55 73	80.97 56	62.54 102	68.59 65	65.13 64	61.43 77	74.81 54	69.32 61	81.01 75	79.59 82	87.64 70	85.89 66

MSDG		71.52 77	69.87 97	73.44 65	82.21 63	79.35 101	79.52 64	64.59 73	66.15 70	61.87 71	53.21 156	56.09 131	65.85 97	78.94 106	78.50 92	86.60 112	76.85 166

MVS_Test		75.37 61	77.13 64	73.31 66	79.07 78	81.32 76	79.98 60	60.12 140	69.72 64	64.11 67	70.53 49	73.22 58	68.90 62	80.14 91	79.48 86	87.67 69	85.50 72

ACMH+	66.54 13	71.36 78	70.09 92	72.85 67	82.59 60	81.13 78	78.56 85	68.04 51	61.55 100	52.52 127	51.50 175	54.14 144	68.56 65	78.85 107	79.50 85	86.82 99	83.94 93

Fast-Effi-MVS+		73.11 70	73.66 73	72.48 68	77.72 97	80.88 83	78.55 86	58.83 159	65.19 76	60.36 76	59.98 84	62.42 97	71.22 54	81.66 61	80.61 74	88.20 57	84.88 85

Effi-MVS+-dtu		71.82 75	71.86 83	71.78 69	78.77 79	80.47 92	78.55 86	61.67 114	60.68 105	55.49 108	58.48 93	65.48 88	68.85 63	76.92 143	75.55 158	87.35 74	85.46 73

ACMH	65.37 14	70.71 82	70.00 93	71.54 70	82.51 61	82.47 71	77.78 97	68.13 50	56.19 155	46.06 163	54.30 135	51.20 181	68.68 64	80.66 77	80.72 65	86.07 124	84.45 89
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
v7		70.33 89	69.87 97	70.88 71	74.79 133	81.04 79	79.22 67	60.57 125	57.70 130	56.65 103	54.23 140	55.29 136	66.95 71	78.28 113	77.47 114	87.12 85	85.05 81

v6		70.35 86	69.94 94	70.83 72	74.68 141	80.62 85	78.81 76	60.16 138	58.81 118	58.17 86	55.01 121	57.31 115	66.32 86	77.53 126	76.73 137	86.82 99	83.62 95

v1neww		70.34 87	69.93 95	70.82 73	74.68 141	80.61 86	78.80 77	60.17 135	58.74 120	58.10 87	55.00 122	57.28 116	66.33 84	77.53 126	76.74 133	86.82 99	83.61 96

v7new		70.34 87	69.93 95	70.82 73	74.68 141	80.61 86	78.80 77	60.17 135	58.74 120	58.10 87	55.00 122	57.28 116	66.33 84	77.53 126	76.74 133	86.82 99	83.61 96

v16		70.07 94	69.46 106	70.79 75	74.74 139	77.08 134	78.79 79	58.86 153	59.75 112	59.15 80	54.87 127	57.33 113	66.38 81	77.61 124	76.77 126	86.81 104	82.79 108

v18		70.10 93	69.52 104	70.77 76	74.66 144	77.06 135	78.84 74	58.84 158	60.01 111	59.23 79	55.06 120	57.47 111	66.34 83	77.50 130	76.75 131	86.71 106	82.77 110

v10		70.22 91	69.76 101	70.74 77	74.79 133	80.30 95	79.22 67	59.81 143	57.71 129	56.58 104	54.22 142	55.31 134	66.95 71	78.28 113	77.47 114	87.12 85	85.07 80

v2v482		70.05 95	69.46 106	70.74 77	74.62 145	80.32 94	79.00 70	60.62 124	57.41 131	56.89 96	55.43 115	55.14 137	66.39 80	77.25 138	77.14 121	86.90 92	83.57 101

v17		70.03 96	69.43 111	70.72 79	74.75 138	77.09 133	78.78 81	58.85 155	59.53 115	58.72 83	54.87 127	57.39 112	66.38 81	77.60 125	76.75 131	86.83 98	82.80 106

IB-MVS	66.94 12	71.21 79	71.66 84	70.68 80	79.18 77	82.83 69	72.61 146	61.77 112	59.66 113	63.44 70	53.26 154	59.65 104	59.16 130	76.78 146	82.11 52	87.90 64	87.33 59
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
v8		70.23 90	69.86 99	70.67 81	74.69 140	79.82 97	78.79 79	59.18 149	58.80 119	58.20 85	55.00 122	57.33 113	66.31 87	77.51 129	76.71 141	86.82 99	83.88 94

v1144		69.93 100	69.36 112	70.61 82	74.89 126	80.93 80	79.11 69	60.64 123	55.97 158	55.31 110	53.85 147	54.14 144	66.54 78	78.10 115	77.44 116	87.14 81	85.09 79

UA-Net		74.47 64	77.80 56	70.59 83	85.33 48	85.40 52	73.54 139	65.98 65	60.65 106	56.00 107	72.11 44	79.15 40	54.63 165	83.13 55	82.25 51	88.04 61	81.92 122

v1		69.97 97	69.45 108	70.59 83	74.78 135	80.51 89	78.84 74	60.30 130	56.98 134	56.81 98	54.69 130	56.29 126	65.91 96	77.37 133	76.71 141	86.89 94	83.59 98

v1141		69.96 99	69.44 109	70.58 85	74.78 135	80.50 90	78.85 72	60.30 130	56.95 137	56.74 100	54.68 131	56.26 128	65.93 94	77.38 132	76.72 138	86.88 95	83.57 101

divwei89l23v2f112		69.97 97	69.44 109	70.58 85	74.78 135	80.50 90	78.85 72	60.30 130	56.97 136	56.75 99	54.67 132	56.27 127	65.92 95	77.37 133	76.72 138	86.88 95	83.58 100

v15		69.61 102	68.88 118	70.46 87	74.81 132	77.03 138	78.75 82	58.83 159	57.06 133	57.18 92	54.55 133	56.37 122	66.13 91	77.70 121	76.76 128	87.03 89	82.69 113

V9		69.58 104	68.83 120	70.46 87	74.85 130	77.04 136	78.65 84	58.85 155	56.83 140	57.12 93	54.26 138	56.31 124	66.14 90	77.83 119	76.76 128	87.13 82	82.79 108

V14		69.59 103	68.86 119	70.45 89	74.83 131	77.04 136	78.70 83	58.83 159	56.95 137	57.08 94	54.41 134	56.34 123	66.15 88	77.77 120	76.76 128	87.08 87	82.74 111

v13		69.52 107	68.76 124	70.41 90	74.88 127	77.02 140	78.52 90	58.86 153	56.61 149	56.91 95	54.00 145	56.17 130	66.11 92	77.93 116	76.74 133	87.21 76	82.83 105

v12		69.54 105	68.79 122	70.41 90	74.88 127	77.03 138	78.54 89	58.85 155	56.71 141	56.87 97	54.13 143	56.23 129	66.15 88	77.89 117	76.74 133	87.17 77	82.80 106

IterMVS-LS		71.69 76	72.82 79	70.37 92	77.54 99	76.34 150	75.13 115	60.46 128	61.53 101	57.57 90	64.89 69	67.33 83	66.04 93	77.09 142	77.37 118	85.48 144	85.18 78
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
v1192		69.50 108	68.83 120	70.29 93	74.49 146	80.92 82	78.55 86	60.54 126	55.04 167	54.21 113	52.79 164	52.33 169	66.92 73	77.88 118	77.35 119	87.04 88	85.51 71

v11		69.37 110	68.65 128	70.20 94	74.87 129	76.97 141	78.29 93	58.55 163	56.38 152	56.04 106	54.02 144	54.98 138	66.47 79	78.30 112	76.91 124	86.97 90	83.02 104

diffmvs		73.13 69	75.65 69	70.19 95	74.07 149	77.17 132	78.24 94	57.45 167	72.44 59	64.02 68	69.05 53	75.92 50	64.86 99	75.18 159	75.27 161	82.47 170	84.53 87

v144192		69.34 111	68.68 127	70.12 96	74.06 150	80.54 88	78.08 96	60.54 126	54.99 169	54.13 114	52.92 161	52.80 165	66.73 76	77.13 140	76.72 138	87.15 78	85.63 67

HyFIR lowres test		69.47 109	68.94 117	70.09 97	76.77 106	82.93 68	76.63 106	60.17 135	59.00 117	54.03 115	40.54 206	65.23 89	67.89 67	76.54 150	78.30 99	85.03 150	80.07 139

EPP-MVSNet		74.00 66	77.41 61	70.02 98	80.53 71	83.91 60	74.99 117	62.68 98	65.06 77	49.77 143	68.68 56	72.09 62	63.06 107	82.49 59	80.73 64	89.12 47	88.91 49

v1921920		69.03 114	68.32 133	69.86 99	74.03 151	80.37 93	77.55 98	60.25 134	54.62 170	53.59 119	52.36 171	51.50 180	66.75 75	77.17 139	76.69 143	86.96 91	85.56 68

MS-PatchMatch		70.17 92	70.49 90	69.79 100	80.98 68	77.97 124	77.51 99	58.95 151	62.33 93	55.22 111	53.14 157	65.90 87	62.03 112	79.08 105	77.11 122	84.08 162	77.91 156

tpmp4_e23		68.32 122	67.08 149	69.76 101	77.86 87	75.22 164	78.37 91	56.17 176	66.06 72	64.27 66	57.15 105	54.89 139	63.40 105	70.97 187	68.29 197	78.46 187	77.00 165

FC-MVSNet-train		72.60 73	75.07 71	69.71 102	81.10 67	78.79 111	73.74 137	65.23 69	66.10 71	53.34 120	70.36 50	63.40 94	56.92 145	81.44 65	80.96 61	87.93 63	84.46 88

v1240		68.64 119	67.89 140	69.51 103	73.89 153	80.26 96	76.73 105	59.97 142	53.43 179	53.08 122	51.82 174	50.84 183	66.62 77	76.79 145	76.77 126	86.78 105	85.34 75

MVSTER		72.06 74	74.24 72	69.51 103	70.39 179	75.97 154	76.91 104	57.36 169	64.64 81	61.39 74	68.86 54	63.76 92	63.46 104	81.44 65	79.70 79	87.56 71	85.31 76

CHOSEN 1792x2688		69.20 113	69.26 113	69.13 105	76.86 105	78.93 105	77.27 102	60.12 140	61.86 97	54.42 112	42.54 200	61.61 98	66.91 74	78.55 110	78.14 103	79.23 185	83.23 103

PatchMatch-RL		67.78 132	66.65 154	69.10 106	73.01 158	72.69 175	68.49 166	61.85 111	62.93 91	60.20 78	56.83 107	50.42 185	69.52 60	75.62 156	74.46 167	81.51 173	73.62 186

CANet_DTU		73.29 68	76.96 65	69.00 107	77.04 104	82.06 72	79.49 65	56.30 174	67.85 66	53.29 121	71.12 48	70.37 70	61.81 117	81.59 63	80.96 61	86.09 123	84.73 86

DWT-MVSNet_training		67.24 144	65.96 161	68.74 108	76.15 112	74.36 171	74.37 124	56.66 172	61.82 98	60.51 75	58.23 99	49.76 189	65.07 98	70.04 195	70.39 182	79.70 182	77.11 163

UniMVSNet_NR-MVSNet		70.59 83	72.19 81	68.72 109	77.72 97	80.72 84	73.81 135	69.65 40	61.99 95	43.23 174	60.54 80	57.50 110	58.57 131	79.56 99	81.07 60	89.34 42	83.97 91

COLMAP_ROB	62.73 15	67.66 135	66.76 153	68.70 110	80.49 72	77.98 122	75.29 110	62.95 85	63.62 86	49.96 141	47.32 193	50.72 184	58.57 131	76.87 144	75.50 159	84.94 153	75.33 176
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
IS_MVSNet		73.33 67	77.34 62	68.65 111	81.29 64	83.47 63	74.45 120	63.58 79	65.75 74	48.49 146	67.11 65	70.61 67	54.63 165	84.51 42	83.58 46	89.48 40	86.34 64

pmmvs4		67.89 129	67.39 146	68.48 112	71.60 173	73.57 173	74.45 120	60.98 120	64.65 80	57.97 89	54.95 125	51.73 178	61.88 114	73.78 166	75.11 163	83.99 164	77.91 156

v148		67.85 130	67.53 142	68.23 113	73.25 157	77.57 130	74.26 131	57.36 169	55.70 160	57.45 91	53.53 149	55.42 133	61.96 113	75.23 158	73.92 168	85.08 149	81.32 126

CostFormer		68.92 115	69.58 103	68.15 114	75.98 116	76.17 153	78.22 95	51.86 190	65.80 73	61.56 73	63.57 73	62.83 95	61.85 115	70.40 194	68.67 192	79.42 183	79.62 145

DU-MVS		69.63 101	70.91 87	68.13 115	75.99 114	79.54 98	73.81 135	69.20 45	61.20 103	43.23 174	58.52 91	53.50 151	58.57 131	79.22 103	80.45 75	87.97 62	83.97 91

GBi-Net		70.78 80	73.37 76	67.76 116	72.95 159	78.00 119	75.15 112	62.72 93	64.13 83	51.44 129	58.37 94	69.02 76	57.59 137	81.33 68	80.72 65	86.70 107	82.02 116

test1		70.78 80	73.37 76	67.76 116	72.95 159	78.00 119	75.15 112	62.72 93	64.13 83	51.44 129	58.37 94	69.02 76	57.59 137	81.33 68	80.72 65	86.70 107	82.02 116

V42		68.76 118	69.63 102	67.74 118	64.93 200	78.01 118	78.30 92	56.48 173	58.65 122	56.30 105	54.26 138	57.03 119	64.85 100	77.47 131	77.01 123	85.60 142	84.96 83

FMVSNet2		70.39 85	72.67 80	67.72 119	72.95 159	78.00 119	75.15 112	62.69 97	63.29 88	51.25 133	55.64 112	68.49 82	57.59 137	80.91 76	80.35 76	86.70 107	82.02 116

FMVSNet3		70.49 84	72.90 78	67.67 120	72.88 162	77.98 122	74.96 118	62.72 93	64.13 83	51.44 129	58.37 94	69.02 76	57.43 140	79.43 101	79.57 83	86.59 113	81.81 123

EG-PatchMatch MVS		67.24 144	66.94 150	67.60 121	78.73 80	81.35 75	73.28 143	59.49 145	46.89 205	51.42 132	43.65 197	53.49 152	55.50 161	81.38 67	80.66 71	87.15 78	81.17 127

Vis-MVSNet		72.77 72	77.20 63	67.59 122	74.19 148	84.01 59	76.61 107	61.69 113	60.62 107	50.61 137	70.25 51	71.31 65	55.57 160	83.85 47	82.28 50	86.90 92	88.08 53
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
TranMVSNet+NR-MVSNet		69.25 112	70.81 88	67.43 123	77.23 103	79.46 100	73.48 141	69.66 39	60.43 108	39.56 186	58.82 90	53.48 153	55.74 158	79.59 97	81.21 59	88.89 50	82.70 112

tfpn200view9		68.11 124	68.72 125	67.40 124	77.83 89	78.93 105	74.28 125	62.81 86	56.64 143	46.82 154	52.65 166	53.47 154	56.59 146	80.41 78	78.43 93	86.11 120	80.52 135

tfpn111		68.38 120	69.23 114	67.39 125	77.83 89	78.93 105	74.28 125	62.81 86	56.64 143	46.70 156	56.24 109	53.47 154	56.59 146	80.41 78	78.43 93	86.11 120	80.53 133

conf0.01		67.72 133	67.99 137	67.39 125	77.82 94	78.94 103	74.28 125	62.81 86	56.64 143	46.70 156	53.33 152	48.59 194	56.59 146	80.34 85	78.43 93	86.16 119	79.67 144

conf0.002		67.52 141	67.64 141	67.39 125	77.80 96	78.94 103	74.28 125	62.81 86	56.64 143	46.70 156	53.65 148	46.28 202	56.59 146	80.33 86	78.37 98	86.17 118	79.23 148

conf200view11		68.11 124	68.72 125	67.39 125	77.83 89	78.93 105	74.28 125	62.81 86	56.64 143	46.70 156	52.65 166	53.47 154	56.59 146	80.41 78	78.43 93	86.11 120	80.53 133

TDRefinement		66.09 150	65.03 174	67.31 129	69.73 183	76.75 143	75.33 108	64.55 74	60.28 109	49.72 144	45.63 195	42.83 209	60.46 125	75.75 153	75.95 153	84.08 162	78.04 155

thres200		67.98 127	68.55 130	67.30 130	77.89 86	78.86 109	74.18 132	62.75 91	56.35 153	46.48 161	52.98 160	53.54 150	56.46 151	80.41 78	77.97 104	86.05 126	79.78 143

tpm cat1		65.41 152	63.81 182	67.28 131	75.61 120	72.88 174	75.32 109	52.85 184	62.97 90	63.66 69	53.24 155	53.29 161	61.83 116	65.54 206	64.14 209	74.43 203	74.60 178

v7n		67.05 146	66.94 150	67.17 132	72.35 164	78.97 102	73.26 144	58.88 152	51.16 190	50.90 134	48.21 186	50.11 187	60.96 120	77.70 121	77.38 117	86.68 110	85.05 81

thres400		67.95 128	68.62 129	67.17 132	77.90 84	78.59 114	74.27 130	62.72 93	56.34 154	45.77 165	53.00 159	53.35 159	56.46 151	80.21 90	78.43 93	85.91 135	80.43 136

thres100view900		67.60 139	68.02 136	67.12 134	77.83 89	77.75 126	73.90 133	62.52 103	56.64 143	46.82 154	52.65 166	53.47 154	55.92 155	78.77 108	77.62 111	85.72 140	79.23 148

UGNet		72.78 71	77.67 57	67.07 135	71.65 171	83.24 65	75.20 111	63.62 78	64.93 78	56.72 101	71.82 46	73.30 57	49.02 181	81.02 74	80.70 70	86.22 117	88.67 51
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
Fast-Effi-MVS+-dtu		68.34 121	69.47 105	67.01 136	75.15 122	77.97 124	77.12 103	55.40 177	57.87 124	46.68 160	56.17 111	60.39 100	62.36 110	76.32 151	76.25 147	85.35 146	81.34 125

FMVSNet1		68.84 116	70.47 91	66.94 137	71.35 176	77.68 127	74.71 119	62.35 107	56.93 139	49.94 142	50.01 181	64.59 90	57.07 143	81.33 68	80.72 65	86.25 116	82.00 119

GA-MVS		68.14 123	69.17 115	66.93 138	73.77 154	78.50 115	74.45 120	58.28 164	55.11 166	48.44 147	60.08 82	53.99 147	61.50 118	78.43 111	77.57 112	85.13 148	80.54 132

thres600view7		67.68 134	68.43 131	66.80 139	77.90 84	78.86 109	73.84 134	62.75 91	56.07 156	44.70 171	52.85 163	52.81 164	55.58 159	80.41 78	77.77 107	86.05 126	80.28 137

view600		67.63 138	68.36 132	66.77 140	77.84 88	78.66 112	73.74 137	62.62 100	56.04 157	44.98 168	52.86 162	52.83 163	55.48 162	80.36 84	77.75 108	85.95 134	80.02 140

UniMVSNet (Re)		69.53 106	71.90 82	66.76 141	76.42 107	80.93 80	72.59 147	68.03 52	61.75 99	41.68 182	58.34 97	57.23 118	53.27 172	79.53 100	80.62 73	88.57 54	84.90 84

USDC		67.36 142	67.90 139	66.74 142	71.72 169	75.23 162	71.58 154	60.28 133	67.45 67	50.54 138	60.93 78	45.20 206	62.08 111	76.56 149	74.50 166	84.25 161	75.38 175

NR-MVSNet		68.79 117	70.56 89	66.71 143	77.48 100	79.54 98	73.52 140	69.20 45	61.20 103	39.76 185	58.52 91	50.11 187	51.37 176	80.26 89	80.71 69	88.97 48	83.59 98

view800		67.35 143	68.22 135	66.35 144	77.83 89	78.62 113	72.97 145	62.58 101	55.71 159	44.13 172	52.69 165	52.24 173	54.58 167	80.27 88	78.19 101	86.01 129	79.79 142

Baseline_NR-MVSNet		67.53 140	68.77 123	66.09 145	75.99 114	74.75 168	72.43 148	68.41 48	61.33 102	38.33 190	51.31 176	54.13 146	56.03 154	79.22 103	78.19 101	85.37 145	82.45 114

tfpn		66.58 147	67.18 147	65.88 146	77.82 94	78.45 116	72.07 150	62.52 103	55.35 163	43.21 176	52.54 170	46.12 203	53.68 168	80.02 92	78.23 100	85.99 132	79.55 146

conf0.05thres1000		66.26 149	66.77 152	65.66 147	77.45 101	78.10 117	71.85 153	62.44 106	51.47 189	43.00 177	47.92 188	51.66 179	53.40 170	79.71 95	77.97 104	85.82 136	80.56 131

dps		64.00 169	62.99 185	65.18 148	73.29 156	72.07 177	68.98 165	53.07 183	57.74 128	58.41 84	55.55 114	47.74 198	60.89 123	69.53 197	67.14 201	76.44 195	71.19 195

CDS-MVSNet		67.65 136	69.83 100	65.09 149	75.39 121	76.55 145	74.42 123	63.75 77	53.55 178	49.37 145	59.41 87	62.45 96	44.44 193	79.71 95	79.82 78	83.17 168	77.36 160
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
tfpnnormal		64.27 165	63.64 183	65.02 150	75.84 117	75.61 156	71.24 156	62.52 103	47.79 202	42.97 178	42.65 199	44.49 207	52.66 174	78.77 108	76.86 125	84.88 154	79.29 147

v748		65.12 156	65.24 169	64.98 151	69.77 182	76.45 146	69.47 162	57.06 171	49.93 195	50.70 135	47.87 189	49.50 191	57.14 142	73.64 168	75.18 162	85.75 139	84.14 90

TinyColmap		62.84 174	61.03 198	64.96 152	69.61 184	71.69 178	68.48 167	59.76 144	55.41 162	47.69 152	47.33 192	34.20 219	62.76 109	74.52 161	72.59 175	81.44 174	71.47 194

pmmvs-eth3d		63.52 171	62.44 192	64.77 153	66.82 194	70.12 183	69.41 163	59.48 146	54.34 174	52.71 123	46.24 194	44.35 208	56.93 144	72.37 171	73.77 169	83.30 166	75.91 169

EPNet_dtu		68.08 126	71.00 86	64.67 154	79.64 74	68.62 189	75.05 116	63.30 80	66.36 69	45.27 167	67.40 63	66.84 85	43.64 195	75.37 157	74.98 165	81.15 175	77.44 159
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
test-LLR		64.42 162	64.36 178	64.49 155	75.02 124	63.93 203	66.61 180	61.96 109	54.41 171	47.77 150	57.46 102	60.25 101	55.20 163	70.80 188	69.33 187	80.40 179	74.38 180

IterMVS		66.36 148	68.30 134	64.10 156	69.48 186	74.61 169	73.41 142	50.79 196	57.30 132	48.28 148	60.64 79	59.92 103	60.85 124	74.14 164	72.66 174	81.80 172	78.82 152
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
v52		65.23 154	66.24 156	64.06 157	61.94 204	76.42 147	72.06 151	54.30 179	49.94 194	50.04 140	47.41 191	52.42 167	60.23 127	75.71 154	76.22 148	85.78 137	85.56 68

V4		65.23 154	66.23 157	64.06 157	61.94 204	76.42 147	72.05 152	54.31 178	49.91 196	50.06 139	47.42 190	52.40 168	60.24 126	75.71 154	76.22 148	85.78 137	85.56 68

CR-MVSNet		64.83 159	65.54 167	64.01 159	70.64 178	69.41 184	65.97 183	52.74 185	57.81 126	52.65 124	54.27 136	56.31 124	60.92 121	72.20 176	73.09 172	81.12 176	75.69 172

PatchmatchNet		64.21 168	64.65 176	63.69 160	71.29 177	68.66 188	69.63 160	51.70 192	63.04 89	53.77 118	59.83 86	58.34 108	60.23 127	68.54 201	66.06 204	75.56 198	68.08 202
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
TransMVSNet (Re)		64.74 161	65.66 166	63.66 161	77.40 102	75.33 159	69.86 158	62.67 99	47.63 203	41.21 183	50.01 181	52.33 169	45.31 192	79.57 98	77.69 110	85.49 143	77.07 164

pm-mvs1		65.62 151	67.42 144	63.53 162	73.66 155	76.39 149	69.66 159	60.87 122	49.73 197	43.97 173	51.24 177	57.00 120	48.16 182	79.89 93	77.84 106	84.85 156	79.82 141

thresconf0.02		64.77 160	65.90 162	63.44 163	76.37 108	75.17 167	69.51 161	61.28 115	56.98 134	39.01 188	56.24 109	48.68 193	49.78 179	77.13 140	75.61 156	84.71 157	71.53 193

RPSCF		67.64 137	71.25 85	63.43 164	61.86 206	70.73 181	67.26 174	50.86 195	74.20 55	58.91 81	67.49 62	69.33 73	64.10 102	71.41 180	68.45 196	77.61 189	77.17 161

MDTV_nov1_ep13		64.37 163	65.24 169	63.37 165	68.94 188	70.81 180	72.40 149	50.29 199	60.10 110	53.91 117	60.07 83	59.15 106	57.21 141	69.43 198	67.30 199	77.47 190	69.78 198

tfpn_ndepth		65.09 157	67.12 148	62.73 166	75.75 119	76.23 151	68.00 168	60.36 129	58.16 123	40.27 184	54.89 126	54.22 143	46.80 188	76.69 148	75.66 155	85.19 147	73.98 184

CMPMVS	47.78 17	62.49 178	62.52 190	62.46 167	70.01 181	70.66 182	62.97 196	51.84 191	51.98 185	56.71 102	42.87 198	53.62 148	57.80 136	72.23 174	70.37 183	75.45 200	75.91 169
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
tfpnview11		64.33 164	66.17 158	62.18 168	76.25 109	75.23 162	67.45 171	61.16 116	55.50 161	36.38 195	55.35 116	51.89 175	46.96 184	77.28 137	76.10 152	84.86 155	71.85 192

anonymousdsp		65.28 153	67.98 138	62.13 169	58.73 214	73.98 172	67.10 176	50.69 197	48.41 200	47.66 153	54.27 136	52.75 166	61.45 119	76.71 147	80.20 77	87.13 82	89.53 47

tfpn_n400		64.23 166	66.05 159	62.12 170	76.20 110	75.24 160	67.43 172	61.15 117	54.04 176	36.38 195	55.35 116	51.89 175	46.94 185	77.31 135	76.15 150	84.59 158	72.36 189

tfpnconf		64.23 166	66.05 159	62.12 170	76.20 110	75.24 160	67.43 172	61.15 117	54.04 176	36.38 195	55.35 116	51.89 175	46.94 185	77.31 135	76.15 150	84.59 158	72.36 189

pmmvs6		62.41 179	62.88 186	61.87 172	71.38 175	75.18 166	67.76 170	59.45 147	41.64 213	42.52 181	37.33 208	52.91 162	46.87 187	77.67 123	76.26 146	83.23 167	79.18 150

tpmrst		62.00 183	62.35 193	61.58 173	71.62 172	64.14 202	69.07 164	48.22 208	62.21 94	53.93 116	58.26 98	55.30 135	55.81 157	63.22 211	62.62 212	70.85 213	70.70 196

tpm		62.41 179	63.15 184	61.55 174	72.24 165	63.79 205	71.31 155	46.12 212	57.82 125	55.33 109	59.90 85	54.74 140	53.63 169	67.24 204	64.29 207	70.65 214	74.25 182

Vis-MVSNet (Re-imp)		67.83 131	73.52 74	61.19 175	78.37 82	76.72 144	66.80 178	62.96 84	65.50 75	34.17 201	67.19 64	69.68 72	39.20 204	79.39 102	79.44 87	85.68 141	76.73 167

SixPastTwentyTwo		61.84 186	62.45 191	61.12 176	69.20 187	72.20 176	62.03 199	57.40 168	46.54 206	38.03 192	57.14 106	41.72 211	58.12 135	69.67 196	71.58 178	81.94 171	78.30 154

tfpn1000		63.81 170	66.31 155	60.90 177	75.76 118	75.74 155	65.14 187	60.14 139	56.47 150	35.99 198	55.11 119	52.30 171	43.42 196	76.21 152	75.34 160	84.97 152	73.01 188

LTVRE_ROB	59.44 16	61.82 188	62.64 189	60.87 178	72.83 163	77.19 131	64.37 191	58.97 150	33.56 226	28.00 209	52.59 169	42.21 210	63.93 103	74.52 161	76.28 145	77.15 192	82.13 115
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
pmmvs5		62.37 182	64.04 180	60.42 179	65.03 198	71.67 179	67.17 175	52.70 187	50.30 191	44.80 169	54.23 140	51.19 182	49.37 180	72.88 170	73.48 171	83.45 165	74.55 179

RPMNet		61.71 189	62.88 186	60.34 180	69.51 185	69.41 184	63.48 194	49.23 200	57.81 126	45.64 166	50.51 179	50.12 186	53.13 173	68.17 203	68.49 195	81.07 177	75.62 174

PMMVS		65.06 158	69.17 115	60.26 181	55.25 223	63.43 206	66.71 179	43.01 221	62.41 92	50.64 136	69.44 52	67.04 84	63.29 106	74.36 163	73.54 170	82.68 169	73.99 183

PM-MVS		60.48 192	60.94 199	59.94 182	58.85 213	66.83 195	64.27 192	51.39 193	55.03 168	48.03 149	50.00 183	40.79 213	58.26 134	69.20 199	67.13 202	78.84 186	77.60 158

MDTV_nov1_ep13_2view		60.16 193	60.51 200	59.75 183	65.39 197	69.05 187	68.00 168	48.29 206	51.99 184	45.95 164	48.01 187	49.64 190	53.39 171	68.83 200	66.52 203	77.47 190	69.55 199

PEN-MVS		62.96 173	65.77 165	59.70 184	73.98 152	75.45 157	63.39 195	67.61 55	52.49 182	25.49 212	53.39 150	49.12 192	40.85 202	71.94 178	77.26 120	86.86 97	80.72 130

gg-mvs-nofinetune		62.55 176	65.05 173	59.62 185	78.72 81	77.61 128	70.83 157	53.63 180	39.71 217	22.04 222	36.36 210	64.32 91	47.53 183	81.16 72	79.03 89	85.00 151	77.17 161

PatchT		61.97 184	64.04 180	59.55 186	60.49 208	67.40 192	56.54 209	48.65 204	56.69 142	52.65 124	51.10 178	52.14 174	60.92 121	72.20 176	73.09 172	78.03 188	75.69 172

CP-MVSNet		62.68 175	65.49 168	59.40 187	71.84 167	75.34 158	62.87 197	67.04 58	52.64 181	27.19 210	53.38 151	48.15 196	41.40 200	71.26 181	75.68 154	86.07 124	82.00 119

PS-CasMVS		62.38 181	65.06 172	59.25 188	71.73 168	75.21 165	62.77 198	66.99 59	51.94 187	26.96 211	52.00 173	47.52 199	41.06 201	71.16 184	75.60 157	85.97 133	81.97 121

CVMVSNet		62.55 176	65.89 163	58.64 189	66.95 192	69.15 186	66.49 182	56.29 175	52.46 183	32.70 202	59.27 88	58.21 109	50.09 178	71.77 179	71.39 179	79.31 184	78.99 151

DTE-MVSNet		61.85 185	64.96 175	58.22 190	74.32 147	74.39 170	61.01 201	67.85 54	51.76 188	21.91 223	53.28 153	48.17 195	37.74 205	72.22 175	76.44 144	86.52 115	78.49 153

WR-MVS		63.03 172	67.40 145	57.92 191	75.14 123	77.60 129	60.56 202	66.10 62	54.11 175	23.88 213	53.94 146	53.58 149	34.50 209	73.93 165	77.71 109	87.35 74	80.94 128

EPMVS		60.00 194	61.97 194	57.71 192	68.46 189	63.17 209	64.54 190	48.23 207	63.30 87	44.72 170	60.19 81	56.05 132	50.85 177	65.27 208	62.02 214	69.44 216	63.81 209

TESTMET0.1,1		61.10 190	64.36 178	57.29 193	57.53 216	63.93 203	66.61 180	36.22 227	54.41 171	47.77 150	57.46 102	60.25 101	55.20 163	70.80 188	69.33 187	80.40 179	74.38 180

WR-MVS_H		61.83 187	65.87 164	57.12 194	71.72 169	76.87 142	61.45 200	66.19 60	51.97 186	22.92 220	53.13 158	52.30 171	33.80 210	71.03 185	75.00 164	86.65 111	80.78 129

MVS-HIRNet		54.41 205	52.10 212	57.11 195	58.99 212	56.10 218	49.68 219	49.10 201	46.18 207	52.15 128	33.18 217	46.11 204	56.10 153	63.19 212	59.70 220	76.64 194	60.25 216

test-mter		60.84 191	64.62 177	56.42 196	55.99 221	64.18 201	65.39 185	34.23 229	54.39 173	46.21 162	57.40 104	59.49 105	55.86 156	71.02 186	69.65 185	80.87 178	76.20 168

gm-plane-assit		57.00 200	57.62 206	56.28 197	76.10 113	62.43 213	47.62 222	46.57 210	33.84 225	23.24 216	37.52 207	40.19 214	59.61 129	79.81 94	77.55 113	84.55 160	72.03 191

TAMVS		59.58 195	62.81 188	55.81 198	66.03 196	65.64 200	63.86 193	48.74 203	49.95 193	37.07 194	54.77 129	58.54 107	44.44 193	72.29 173	71.79 176	74.70 202	66.66 204

test0.0.03 1		58.80 196	61.58 196	55.56 199	75.02 124	68.45 190	59.58 206	61.96 109	52.74 180	29.57 205	49.75 184	54.56 141	31.46 212	71.19 182	69.77 184	75.75 196	64.57 207

MIMVSNet		58.52 198	61.34 197	55.22 200	60.76 207	67.01 194	66.81 177	49.02 202	56.43 151	38.90 189	40.59 205	54.54 142	40.57 203	73.16 169	71.65 177	75.30 201	66.00 205

CHOSEN 280x420		58.70 197	61.88 195	54.98 201	55.45 222	50.55 226	64.92 188	40.36 223	55.21 164	38.13 191	48.31 185	63.76 92	63.03 108	73.73 167	68.58 194	68.00 219	73.04 187

Anonymous20231206		56.36 202	57.80 205	54.67 202	70.08 180	66.39 197	60.46 203	57.54 166	49.50 199	29.30 206	33.86 216	46.64 200	35.18 208	70.44 192	68.88 191	75.47 199	68.88 201

FPMVS		51.87 210	50.00 215	54.07 203	66.83 193	57.25 216	60.25 204	50.91 194	50.25 192	34.36 200	36.04 213	32.02 221	41.49 199	58.98 223	56.07 223	70.56 215	59.36 218

FMVSNet5		57.24 199	60.02 201	53.99 204	56.45 218	62.74 210	65.27 186	47.03 209	55.14 165	39.55 187	40.88 203	53.42 158	41.83 197	72.35 172	71.10 181	73.79 205	64.50 208

LP		53.62 208	53.43 208	53.83 205	58.51 215	62.59 212	57.31 208	46.04 213	47.86 201	42.69 180	36.08 212	36.86 217	46.53 189	64.38 209	64.25 208	71.92 210	62.00 214

MDA-MVSNet-bldmvs		53.37 209	53.01 211	53.79 206	43.67 231	67.95 191	59.69 205	57.92 165	43.69 209	32.41 203	41.47 201	27.89 229	52.38 175	56.97 225	65.99 205	76.68 193	67.13 203

ADS-MVSNet		55.94 203	58.01 203	53.54 207	62.48 203	58.48 215	59.12 207	46.20 211	59.65 114	42.88 179	52.34 172	53.31 160	46.31 190	62.00 215	60.02 219	64.23 225	60.24 217

Anonymous20231211		51.46 211	50.59 213	52.46 208	67.30 190	66.70 196	55.00 211	59.22 148	29.96 228	17.62 228	19.11 230	28.74 228	35.72 207	66.42 205	69.52 186	79.92 181	73.71 185

test20.03		53.93 207	56.28 207	51.19 209	72.19 166	65.83 198	53.20 214	61.08 119	42.74 211	22.08 221	37.07 209	45.76 205	24.29 225	70.44 192	69.04 189	74.31 204	63.05 211

testgi		54.39 206	57.86 204	50.35 210	71.59 174	67.24 193	54.95 212	53.25 182	43.36 210	23.78 214	44.64 196	47.87 197	24.96 221	70.45 191	68.66 193	73.60 206	62.78 212

EU-MVSNet		54.63 204	58.69 202	49.90 211	56.99 217	62.70 211	56.41 210	50.64 198	45.95 208	23.14 217	50.42 180	46.51 201	36.63 206	65.51 207	64.85 206	75.57 197	74.91 177

PMVS	39.38 18	46.06 219	43.30 224	49.28 212	62.93 201	38.75 232	41.88 225	53.50 181	33.33 227	35.46 199	28.90 221	31.01 224	33.04 211	58.61 224	54.63 226	68.86 217	57.88 221
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
FC-MVSNet-test		56.90 201	65.20 171	47.21 213	66.98 191	63.20 208	49.11 220	58.60 162	59.38 116	11.50 233	65.60 67	56.68 121	24.66 224	71.17 183	71.36 180	72.38 209	69.02 200

testpf		47.41 214	48.47 220	46.18 214	66.30 195	50.67 225	48.15 221	42.60 222	37.10 221	28.75 207	40.97 202	39.01 216	30.82 213	52.95 228	53.74 227	60.46 226	64.87 206

pmmvs3		47.65 213	49.08 217	45.99 215	44.61 228	54.79 221	50.04 217	31.95 232	33.91 224	29.90 204	30.37 218	33.53 220	46.31 190	63.50 210	63.67 210	73.14 208	63.77 210

test2356		47.20 216	48.62 219	45.54 216	56.38 219	54.89 220	50.62 216	45.08 216	38.65 218	23.40 215	36.23 211	31.10 223	29.31 215	62.76 213	62.49 213	68.48 218	54.23 224

MIMVSNet1		49.27 212	53.25 210	44.62 217	44.61 228	61.52 214	53.61 213	52.18 188	41.62 214	18.68 225	28.14 224	41.58 212	25.50 219	68.46 202	69.04 189	73.15 207	62.37 213

N_pmnet		47.35 215	50.13 214	44.11 218	59.98 209	51.64 224	51.86 215	44.80 217	49.58 198	20.76 224	40.65 204	40.05 215	29.64 214	59.84 221	55.15 224	57.63 227	54.00 225

new-patchmatchnet		46.97 217	49.47 216	44.05 219	62.82 202	56.55 217	45.35 223	52.01 189	42.47 212	17.04 229	35.73 214	35.21 218	21.84 230	61.27 216	54.83 225	65.26 224	60.26 215

1111		43.08 221	44.02 223	41.98 220	59.22 210	49.27 228	41.48 226	45.63 214	35.01 222	23.06 218	28.60 222	30.15 225	27.22 216	60.42 219	57.97 221	55.27 230	46.74 227

testus		45.61 220	49.06 218	41.59 221	56.13 220	55.28 219	43.51 224	39.64 225	37.74 219	18.23 226	35.52 215	31.28 222	24.69 223	62.46 214	62.90 211	67.33 220	58.26 220

testmv		42.58 222	44.36 221	40.49 222	54.63 224	52.76 222	41.21 228	44.37 218	28.83 229	12.87 230	27.16 225	25.03 230	23.01 226	60.83 217	61.13 215	66.88 221	54.81 222

test1235678		42.57 223	44.36 221	40.49 222	54.63 224	52.75 223	41.21 228	44.37 218	28.82 230	12.87 230	27.15 226	25.01 231	23.01 226	60.83 217	61.13 215	66.88 221	54.81 222

Gipuma		36.38 225	35.80 228	37.07 224	45.76 227	33.90 233	29.81 232	48.47 205	39.91 216	18.02 227	8.00 236	8.14 238	25.14 220	59.29 222	61.02 217	55.19 231	40.31 229
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
no-one		36.35 226	37.59 227	34.91 225	46.13 226	49.89 227	27.99 233	43.56 220	20.91 234	7.03 236	14.64 232	15.50 236	18.92 231	42.95 229	60.20 218	65.84 223	59.03 219

new_pmnet		38.40 224	42.64 225	33.44 226	37.54 234	45.00 230	36.60 230	32.72 231	40.27 215	12.72 232	29.89 219	28.90 227	24.78 222	53.17 227	52.90 228	56.31 228	48.34 226

.test1245		30.81 228	29.14 230	32.77 227	59.22 210	49.27 228	41.48 226	45.63 214	35.01 222	23.06 218	28.60 222	30.15 225	27.22 216	60.42 219	0.10 234	0.01 238	0.43 236

test12356		35.10 227	38.50 226	31.13 228	44.14 230	43.70 231	32.27 231	34.42 228	26.51 232	9.47 234	25.22 228	20.34 232	10.86 233	53.47 226	56.15 222	55.59 229	44.11 228

E-PMN		21.77 230	18.24 232	25.89 229	40.22 232	19.58 236	12.46 237	39.87 224	18.68 236	6.71 237	9.57 233	4.31 241	22.36 229	19.89 234	27.28 232	33.73 233	28.34 233

EMVS		20.98 231	17.15 233	25.44 230	39.51 233	19.37 237	12.66 236	39.59 226	19.10 235	6.62 238	9.27 234	4.40 240	22.43 228	17.99 235	24.40 233	31.81 234	25.53 234

GG-mvs-BLEND		46.86 218	67.51 143	22.75 231	0.05 238	76.21 152	64.69 189	0.04 236	61.90 96	0.09 241	55.57 113	71.32 64	0.08 236	70.54 190	67.19 200	71.58 211	69.86 197

PMMVS2		25.60 229	29.75 229	20.76 232	28.00 235	30.93 234	23.10 234	29.18 233	23.14 233	1.46 240	18.23 231	16.54 234	5.08 234	40.22 230	41.40 230	37.76 232	37.79 231

MVE	19.12 19	20.47 232	23.27 231	17.20 233	12.66 237	25.41 235	10.52 238	34.14 230	14.79 237	6.53 239	8.79 235	4.68 239	16.64 232	29.49 232	41.63 229	22.73 236	38.11 230
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
tmp_tt				14.50 234	14.68 236	7.17 239	10.46 239	2.21 235	37.73 220	28.71 208	25.26 227	16.98 233	4.37 235	31.49 231	29.77 231	26.56 235

testmvs		0.09 233	0.15 234	0.02 235	0.01 239	0.02 240	0.05 241	0.01 237	0.11 238	0.01 242	0.26 238	0.01 242	0.06 238	0.10 236	0.10 234	0.01 238	0.43 236

test123		0.09 233	0.14 235	0.02 235	0.00 240	0.02 240	0.02 242	0.01 237	0.09 239	0.00 243	0.30 237	0.00 243	0.08 236	0.03 237	0.09 236	0.01 238	0.45 235

sosnet-low-res		0.00 235	0.00 236	0.00 237	0.00 240	0.00 242	0.00 243	0.00 239	0.00 240	0.00 243	0.00 239	0.00 243	0.00 239	0.00 238	0.00 237	0.00 241	0.00 238

sosnet		0.00 235	0.00 236	0.00 237	0.00 240	0.00 242	0.00 243	0.00 239	0.00 240	0.00 243	0.00 239	0.00 243	0.00 239	0.00 238	0.00 237	0.00 241	0.00 238

ambc			53.42 209		64.99 199	63.36 207	49.96 218		47.07 204	37.12 193	28.97 220	16.36 235	41.82 198	75.10 160	67.34 198	71.55 212	75.72 171

MTAPA										83.48 1		86.45 13

MTMP										82.66 3		84.91 21

Patchmatch-RL test							2.85 240

XVS					86.63 40	88.68 24	85.00 42			71.81 41		81.92 31				90.47 18

X-MVStestdata					86.63 40	88.68 24	85.00 42			71.81 41		81.92 31				90.47 18

mPP-MVS					89.90 21							81.29 36

NP-MVS									80.10 41

Patchmtry						65.80 199	65.97 183	52.74 185		52.65 124

DeepMVS_CX						18.74 238	18.55 235	8.02 234	26.96 231	7.33 235	23.81 229	13.05 237	25.99 218	25.17 233		22.45 237	36.25 232

High-resolution multi-view benchmark