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.

CNVR-MVS		94.53 4	94.85 5	94.15 5	98.03 2	98.59 3	95.56 5	92.91 1	94.86 10	88.46 12	91.32 18	90.83 8	94.03 2	95.20 4	94.16 5	95.89 30	99.01 15

DPM-MVS		92.86 13	93.19 16	92.47 12	95.78 32	97.40 19	97.39 1	92.56 2	92.88 23	81.84 36	81.31 37	92.95 2	91.21 14	96.54 1	97.33 1	96.01 28	93.94 106

MCST-MVS		94.10 6	94.77 6	93.31 7	98.31 1	98.34 4	95.43 6	92.54 3	94.41 14	83.05 28	91.38 16	90.97 7	92.24 10	95.05 6	94.02 6	98.31 1	99.20 10

DPE-MVS		95.10 1	95.53 1	94.60 2	97.77 5	98.64 2	96.60 2	92.45 4	96.34 4	91.41 4	96.70 2	92.26 4	93.56 3	93.68 16	91.73 28	95.79 34	99.37 6

NCCC		93.59 8	94.00 11	93.10 8	97.90 4	97.93 10	95.40 7	92.39 5	94.47 13	84.94 19	91.21 19	89.32 12	92.53 7	93.90 15	92.98 12	95.44 38	98.22 29

DVP-MVS		95.06 2	95.37 3	94.70 1	97.59 7	98.89 1	95.37 8	92.04 6	96.85 2	94.00 1	92.81 11	93.02 1	92.93 4	94.22 13	92.15 19	96.30 21	99.61 2

APDe-MVS		94.31 5	94.30 8	94.33 4	97.57 8	98.06 8	95.79 3	91.98 7	95.50 7	92.19 2	95.25 3	87.97 16	92.93 4	93.01 22	91.02 37	95.52 36	99.29 8

APD-MVS		93.47 9	93.44 14	93.50 6	97.06 11	97.09 24	95.27 9	91.47 8	95.71 6	89.57 9	93.66 6	86.28 22	92.81 6	92.06 31	90.70 39	94.83 53	98.60 21
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
AdaColmap		88.46 37	85.75 58	91.62 18	96.25 25	95.35 57	90.71 41	91.08 9	90.22 37	86.17 16	74.33 51	73.67 78	92.00 12	86.31 98	85.82 94	93.52 84	94.53 93

HPM-MVS++		94.04 7	94.96 4	92.96 9	97.93 3	97.71 14	94.65 11	91.01 10	95.91 5	87.43 14	93.52 8	92.63 3	92.29 9	94.22 13	92.34 16	94.47 56	98.37 26

MSP-MVS		95.00 3	95.47 2	94.45 3	96.78 15	98.11 6	95.72 4	90.91 11	96.68 3	91.57 3	96.98 1	89.47 11	94.76 1	95.24 3	92.15 19	96.98 7	99.64 1

SD-MVS		93.36 11	94.33 7	92.22 13	94.68 41	97.89 12	94.56 12	90.89 12	94.80 11	90.04 8	93.53 7	90.14 9	89.78 20	92.74 24	92.17 17	93.35 99	99.07 13
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
DeepC-MVS_fast	86.59 2	91.69 18	91.39 24	92.05 16	97.43 9	96.92 29	94.05 17	90.23 13	93.31 21	83.19 26	77.91 44	84.23 31	92.42 8	94.62 9	94.83 3	95.00 48	97.88 33
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
SMA-MVS		93.47 9	94.29 9	92.52 11	97.72 6	97.77 13	94.46 14	90.19 14	94.96 9	87.15 15	90.15 22	90.99 6	91.49 13	94.31 11	93.33 10	94.10 61	98.53 24

TSAR-MVS + MP.		93.07 12	93.53 13	92.53 10	94.23 44	97.54 18	94.75 10	89.87 15	95.26 8	89.20 11	93.16 9	88.19 15	92.15 11	91.79 35	89.65 53	94.99 49	99.16 11
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
PCF-MVS	82.38 4	85.52 54	84.41 63	86.81 48	91.51 59	96.23 42	90.27 44	89.81 16	77.87 88	70.67 79	69.20 66	77.86 53	85.55 51	85.92 103	86.38 85	93.03 110	97.43 43
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
SteuartSystems-ACMMP		92.31 14	93.31 15	91.15 20	96.88 13	97.36 20	93.95 18	89.44 17	92.62 24	83.20 25	94.34 5	85.55 24	88.95 27	93.07 21	91.90 24	94.51 55	98.30 27
Skip Steuart: Steuart Systems R&D Blog.
DeepC-MVS	84.14 3	88.80 33	88.03 42	89.71 28	94.83 39	96.56 33	92.57 29	89.38 18	89.25 44	79.59 40	70.02 64	77.05 63	88.24 36	92.44 27	92.79 13	93.65 78	98.10 30
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
ACMMP_NAP		92.16 15	92.91 19	91.28 19	96.95 12	97.36 20	93.66 19	89.23 19	93.33 18	83.71 23	90.53 20	86.84 19	90.39 16	93.30 20	91.56 30	93.74 71	97.43 43

train_agg		91.99 17	93.71 12	89.98 25	96.42 24	97.03 26	94.31 16	89.05 20	93.33 18	77.75 43	95.06 4	88.27 14	88.38 34	92.02 32	91.41 32	94.00 64	98.84 18

DeepPCF-MVS	86.71 1	91.00 22	94.05 10	87.43 42	95.58 35	98.17 5	86.22 71	88.59 21	97.01 1	76.77 49	85.11 31	88.90 13	87.29 39	95.02 7	94.69 4	90.15 170	99.48 5

zzz-MVS		91.59 19	91.12 25	92.13 14	96.76 16	96.68 32	93.39 21	88.00 22	93.63 17	90.76 7	83.97 33	85.33 26	89.89 19	91.60 37	89.65 53	94.00 64	96.97 57

HFP-MVS		92.02 16	92.13 21	91.89 17	97.16 10	96.46 37	93.57 20	87.60 23	93.79 16	88.17 13	93.15 10	83.94 35	91.19 15	90.81 45	89.83 48	93.66 75	96.94 59

SR-MVS					96.04 28			87.51 24				87.60 17

MP-MVS		90.81 24	91.45 22	90.06 24	96.59 19	96.33 40	92.46 31	87.19 25	90.27 36	82.54 32	91.38 16	84.88 28	88.27 35	90.58 47	89.30 59	93.30 101	97.44 41
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
ACMMPR		91.15 21	91.44 23	90.81 21	96.61 18	96.25 41	93.09 22	87.08 26	93.32 20	84.78 20	92.08 14	82.10 41	89.71 21	90.24 50	89.82 49	93.61 80	96.30 71

CSCG		89.81 30	89.69 32	89.96 26	96.55 20	97.90 11	92.89 25	87.06 27	88.74 47	86.17 16	78.24 42	86.53 21	84.75 60	87.82 82	90.59 41	92.32 127	98.01 32

X-MVS		89.73 31	90.65 28	88.66 34	96.44 23	95.93 44	92.26 33	86.98 28	90.73 34	76.32 50	89.56 24	82.05 42	86.51 46	89.98 53	89.60 55	93.43 94	96.72 66

TSAR-MVS + ACMM		90.98 23	93.18 17	88.42 36	95.69 33	96.73 31	94.52 13	86.97 29	92.99 22	76.32 50	92.31 13	86.64 20	84.40 65	92.97 23	92.02 21	92.62 122	98.59 22

CP-MVS		90.57 25	90.68 27	90.44 22	96.13 26	95.90 48	92.77 27	86.86 30	92.12 27	84.19 21	89.18 25	82.37 39	89.43 25	89.65 59	88.43 66	93.27 102	97.13 51

MSLP-MVS++		90.33 26	88.82 37	92.10 15	96.52 22	95.93 44	94.35 15	86.26 31	88.37 49	89.24 10	75.94 49	82.60 38	89.71 21	89.45 62	92.17 17	96.51 15	97.24 48

MSDG		78.11 106	73.17 137	83.86 66	91.78 58	86.83 127	85.25 82	86.02 32	72.84 114	69.69 84	51.43 137	54.00 141	77.61 97	81.95 135	82.27 131	92.83 118	82.91 179

CDPH-MVS		88.76 34	90.43 30	86.81 48	96.04 28	96.53 36	92.95 24	85.95 33	90.36 35	67.93 88	85.80 30	80.69 47	83.82 68	90.81 45	91.85 27	94.18 59	96.99 56

LS3D		78.72 98	75.79 121	82.15 74	91.91 55	89.39 115	83.66 95	85.88 34	76.81 96	59.22 115	57.67 109	58.53 132	83.72 69	82.07 132	81.63 140	88.50 182	84.39 172

ACMMP		88.48 36	88.71 38	88.22 38	94.61 42	95.53 52	90.64 43	85.60 35	90.97 31	78.62 42	89.88 23	74.20 75	86.29 47	88.16 79	86.37 86	93.57 81	95.86 75
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
CNLPA		84.72 61	82.14 83	87.73 40	92.85 50	93.83 72	84.70 87	85.07 36	90.90 32	83.16 27	56.28 118	71.53 85	88.14 37	84.19 112	84.00 116	92.48 124	94.26 100

abl_6				89.54 29	95.55 36	97.59 16	89.01 52	85.00 37	94.67 12	83.04 29	84.70 32	91.47 5	89.46 23			95.20 44	98.63 20

CPTT-MVS		88.17 40	87.84 43	88.55 35	93.33 47	93.75 73	92.33 32	84.75 38	89.87 40	81.72 37	83.93 34	81.12 45	88.45 31	85.42 107	84.07 112	90.72 162	96.72 66

PHI-MVS		89.88 29	92.75 20	86.52 52	94.97 38	97.57 17	89.99 47	84.56 39	92.52 25	69.72 83	90.35 21	87.11 18	84.89 57	91.82 34	92.37 15	95.02 47	97.51 39

DELS-MVS		87.75 42	86.92 48	88.71 33	94.69 40	97.34 23	92.78 26	84.50 40	77.87 88	81.94 34	67.17 72	75.49 70	82.84 74	95.38 2	95.93 2	95.55 35	99.27 9
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
PLC	81.02 6	84.81 59	81.81 88	88.31 37	93.77 46	90.35 105	88.80 53	84.47 41	86.76 57	82.17 33	66.56 75	71.01 90	88.41 33	85.48 105	84.28 110	92.26 129	88.21 159
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
HQP-MVS		86.17 51	87.35 45	84.80 60	91.41 60	92.37 91	91.05 40	84.35 42	88.52 48	64.21 93	87.05 28	68.91 97	84.80 59	89.12 65	88.16 70	92.96 113	97.31 45

TSAR-MVS + GP.		91.29 20	93.11 18	89.18 31	87.81 83	96.21 43	92.51 30	83.83 43	94.24 15	83.77 22	91.87 15	89.62 10	90.07 17	90.40 49	90.31 43	97.09 6	99.10 12

OPM-MVS		81.34 79	78.18 104	85.02 58	91.27 61	91.78 96	90.66 42	83.62 44	62.39 145	65.91 90	63.35 91	64.33 114	85.03 55	87.77 83	85.88 93	93.66 75	91.75 133

OMC-MVS		86.38 50	86.21 55	86.57 51	92.30 53	94.35 67	87.60 60	83.51 45	92.32 26	77.37 47	72.27 56	77.83 55	86.59 45	87.62 84	85.95 91	92.08 131	93.11 119

CANet		89.98 27	90.42 31	89.47 30	94.13 45	98.05 9	91.76 36	83.27 46	90.87 33	81.90 35	72.32 55	84.82 29	88.42 32	94.52 10	93.78 8	97.34 4	98.58 23

MVS_111021_HR		87.82 41	88.84 36	86.62 50	94.42 43	97.36 20	88.21 56	83.26 47	83.42 65	72.52 70	82.63 35	76.93 64	84.95 56	91.93 33	91.15 35	96.39 19	98.49 25

EPNet		89.30 32	90.89 26	87.44 41	95.67 34	96.81 30	91.13 39	83.12 48	91.14 30	76.31 54	87.60 27	80.40 49	84.45 63	92.13 30	91.12 36	93.96 66	97.01 55
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
3Dnovator	80.58 8	88.20 39	86.53 50	90.15 23	96.86 14	96.46 37	91.97 34	83.06 49	85.16 62	83.66 24	62.28 96	82.15 40	88.98 26	90.99 43	92.65 14	96.38 20	96.03 73

TAPA-MVS	80.99 7	84.83 58	84.42 62	85.31 56	91.89 56	93.73 75	88.53 55	82.80 50	89.99 39	69.78 82	71.53 60	75.03 71	85.47 53	86.26 99	84.54 107	93.39 97	89.90 146
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
CLD-MVS		85.43 55	84.24 66	86.83 47	87.69 85	93.16 81	90.01 46	82.72 51	87.17 54	79.28 41	71.43 61	65.81 109	86.02 48	87.33 86	86.96 79	95.25 43	97.83 35
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
QAPM		87.06 45	86.46 51	87.75 39	96.63 17	97.09 24	91.71 37	82.62 52	80.58 78	71.28 76	66.04 79	84.24 30	87.01 42	89.93 54	89.91 47	97.26 5	97.44 41

MVS_111021_LR		87.58 44	88.67 39	86.31 53	92.58 51	95.89 50	86.20 72	82.49 53	89.08 46	77.47 46	86.20 29	74.22 74	85.49 52	90.03 52	88.52 64	93.66 75	96.74 65

PGM-MVS		89.97 28	90.64 29	89.18 31	96.53 21	95.90 48	93.06 23	82.48 54	90.04 38	80.37 38	92.75 12	80.96 46	88.93 28	89.88 55	89.08 61	93.69 74	95.86 75

3Dnovator+	81.14 5	88.59 35	87.49 44	89.88 27	95.83 31	96.45 39	91.94 35	82.41 55	87.09 56	85.94 18	62.80 93	85.37 25	89.46 23	91.51 38	91.89 26	93.72 72	97.30 46

TSAR-MVS + COLMAP		84.93 57	85.79 57	83.92 65	90.90 62	93.57 77	89.25 51	82.00 56	91.29 29	61.66 101	88.25 26	59.46 128	86.71 44	89.79 56	87.09 77	93.01 111	91.09 137

PVSNet_BlendedMVS		86.98 46	87.05 46	86.90 45	93.03 48	96.98 27	86.57 68	81.82 57	89.78 41	82.78 30	71.54 58	66.07 107	80.73 85	93.46 18	91.97 22	96.45 17	99.53 3

PVSNet_Blended		86.98 46	87.05 46	86.90 45	93.03 48	96.98 27	86.57 68	81.82 57	89.78 41	82.78 30	71.54 58	66.07 107	80.73 85	93.46 18	91.97 22	96.45 17	99.53 3

OpenMVS	77.91 11	85.09 56	83.42 70	87.03 44	96.12 27	96.55 35	89.36 49	81.59 59	79.19 84	75.20 56	55.84 122	79.04 51	84.45 63	88.47 73	89.35 58	95.48 37	95.48 83

thres100view900		79.83 89	77.79 108	82.21 73	88.42 75	93.54 78	87.07 61	81.11 60	70.15 123	61.01 107	56.65 112	51.22 143	81.78 78	89.77 57	85.95 91	93.84 68	97.26 47

MVS_0304		88.43 38	89.46 33	87.21 43	91.85 57	97.60 15	92.62 28	81.10 61	87.16 55	73.80 60	72.19 57	83.36 37	87.03 41	94.64 8	93.67 9	96.88 9	97.64 38

baseline1		82.63 69	82.02 84	83.34 69	88.30 78	91.89 95	88.03 59	80.86 62	75.05 102	65.96 89	64.27 87	72.20 83	80.01 90	91.32 41	89.56 56	96.90 8	89.85 147

ACMM	78.09 10	80.91 82	78.39 102	83.86 66	89.61 69	87.71 122	85.16 84	80.67 63	79.04 85	74.18 58	63.82 90	60.84 123	82.59 75	84.33 111	83.59 119	90.96 156	89.39 152
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
casdiffmvs		83.84 63	82.65 79	85.22 57	87.25 89	94.62 65	86.01 75	79.62 64	79.48 81	77.59 45	61.92 99	64.34 113	85.57 50	90.55 48	90.51 42	95.26 41	97.14 50

thres400		78.39 103	76.39 118	80.73 87	88.02 82	92.94 82	84.77 86	78.88 65	65.20 138	59.70 113	55.20 125	50.85 146	79.45 93	88.81 68	84.81 101	93.57 81	96.91 61

EPNet_dtu		78.49 102	81.96 86	74.45 122	92.57 52	88.74 117	82.98 97	78.83 66	83.28 66	44.64 177	77.40 46	67.73 101	53.98 182	85.44 106	84.91 100	93.71 73	86.22 167
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
diffmvs		83.69 65	83.17 75	84.31 62	85.45 107	93.92 68	86.89 63	78.62 67	82.71 71	75.95 55	66.78 74	63.90 116	83.84 67	87.90 81	89.16 60	95.10 46	97.82 36

FC-MVSNet-train		79.54 92	78.20 103	81.09 83	86.55 96	88.63 118	79.96 115	78.53 68	70.90 121	68.24 86	65.87 80	56.45 138	80.29 89	86.20 101	84.08 111	92.97 112	95.31 86

ACMP	79.58 9	82.23 73	81.82 87	82.71 72	88.15 80	90.95 103	85.23 83	78.52 69	81.70 73	72.52 70	78.41 40	60.63 124	80.48 87	82.88 123	83.44 120	91.37 148	94.70 91
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
LGP-MVS_train		82.12 75	82.57 80	81.59 78	89.26 70	90.23 108	88.76 54	78.05 70	81.26 75	61.64 102	79.52 39	62.11 119	79.59 92	85.20 108	84.68 105	92.27 128	95.02 89

Anonymous202405211			75.59 123		85.13 111	91.06 101	84.62 88	77.96 71	69.47 126		40.79 178	63.84 117	84.57 62	83.55 117	84.69 104	89.69 173	95.75 80

tfpn200view9		79.05 96	77.21 110	81.18 82	88.42 75	92.55 89	85.12 85	77.94 72	70.15 123	61.01 107	56.65 112	51.22 143	81.11 79	88.23 76	84.80 102	93.50 89	96.90 62

thres600view7		77.66 109	75.67 122	79.98 90	87.71 84	92.56 87	83.79 94	77.94 72	64.41 140	58.69 117	54.32 130	50.54 147	78.23 96	88.23 76	83.06 123	93.52 84	96.55 70

thres200		78.69 99	76.71 113	80.99 86	88.35 77	92.56 87	86.03 74	77.94 72	66.27 130	60.66 109	56.08 119	51.11 145	79.45 93	88.23 76	85.54 98	93.52 84	97.20 49

Anonymous20231211		78.61 100	75.57 124	82.15 74	84.43 116	90.26 106	84.08 93	77.68 75	71.09 119	72.90 63	39.24 182	66.21 106	84.23 66	82.15 130	84.04 113	89.61 175	96.03 73

ACMH	71.22 14	72.65 135	70.13 150	75.59 113	86.19 102	86.14 142	75.76 158	77.63 76	54.79 167	46.16 167	53.28 133	47.28 157	77.24 99	78.91 158	81.18 152	90.57 164	89.33 153
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
IB-MVS	74.10 12	78.52 101	78.51 101	78.52 98	90.15 64	95.39 55	71.95 173	77.53 77	74.95 104	77.25 48	58.93 106	55.92 139	58.37 173	79.01 157	87.89 71	95.88 31	97.47 40
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
PatchMatch-RL		78.75 97	76.47 117	81.41 80	88.53 74	91.10 100	78.09 133	77.51 78	77.33 92	71.98 72	64.38 86	48.10 155	82.55 76	84.06 114	82.35 129	89.78 172	87.97 161

PVSNet_Blended_VisFu		82.55 70	83.70 69	81.21 81	89.66 66	95.15 61	82.41 102	77.36 79	72.53 116	73.64 61	61.15 102	77.19 62	70.35 143	91.31 42	89.72 52	93.84 68	98.85 17

DI_MVS_plusplus_trai		83.32 67	82.53 81	84.25 63	86.26 101	93.66 76	90.23 45	77.16 80	77.05 95	74.06 59	53.74 131	74.33 73	83.61 70	91.40 40	89.82 49	94.17 60	97.73 37

MVSTER		87.68 43	89.12 34	86.01 54	88.11 81	90.05 110	89.28 50	77.05 81	91.37 28	79.97 39	76.70 47	85.25 27	84.89 57	93.53 17	91.41 32	96.73 11	95.55 82

ETV-MVS		86.73 48	88.60 40	84.55 61	86.73 92	95.19 60	84.56 89	76.83 82	87.70 51	74.40 57	78.22 43	77.23 61	88.73 29	92.30 29	90.69 40	96.06 24	98.83 19

MAR-MVS		85.65 53	86.30 53	84.88 59	95.51 37	95.89 50	86.50 70	76.71 83	89.23 45	68.59 85	70.93 62	74.49 72	88.55 30	89.40 63	90.30 44	93.42 95	93.88 110
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
UGNet		80.71 87	83.09 76	77.93 104	87.02 90	92.71 83	80.28 114	76.53 84	73.83 110	71.35 74	70.07 63	73.71 77	58.93 171	87.39 85	86.97 78	93.48 91	96.94 59
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
COLMAP_ROB	66.31 15	69.91 158	66.61 168	73.76 127	86.44 99	82.76 171	76.59 149	76.46 85	63.82 141	50.92 144	45.60 154	49.13 150	65.87 156	74.96 180	74.45 187	86.30 191	75.57 193
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
EIA-MVS		84.75 60	86.43 52	82.79 71	86.88 91	95.36 56	82.84 101	76.39 86	87.61 53	71.03 77	74.33 51	71.12 89	85.16 54	89.69 58	88.70 63	94.40 57	98.23 28

ACMH+	72.14 13	72.38 137	69.34 157	75.93 112	85.21 110	84.89 158	76.96 147	76.04 87	59.76 150	51.63 136	50.37 141	48.69 152	76.90 102	76.06 174	78.69 167	88.85 180	86.90 165

canonicalmvs		85.93 52	86.26 54	85.54 55	88.94 71	95.44 54	89.56 48	76.01 88	87.83 50	77.70 44	76.43 48	68.66 99	87.80 38	87.02 87	91.51 31	93.25 103	96.95 58

UniMVSNet_NR-MVSNet		73.11 132	72.59 138	73.71 129	76.90 155	86.58 133	77.01 144	75.82 89	65.59 134	48.82 154	50.97 139	48.42 153	71.61 130	79.19 155	83.03 124	92.11 130	94.37 96

TranMVSNet+NR-MVSNet		71.12 149	70.24 149	72.15 144	76.01 163	84.80 160	76.55 150	75.65 90	61.99 146	45.29 172	48.42 150	43.07 173	67.55 149	78.28 162	82.83 125	91.85 138	92.29 125

UA-Net		78.30 104	80.92 90	75.25 116	87.42 87	92.48 90	79.54 119	75.49 91	60.47 149	60.52 110	68.44 71	84.08 33	57.54 175	88.54 72	88.45 65	90.96 156	83.97 174

CHOSEN 1792x2688		80.23 88	79.16 99	81.48 79	91.97 54	96.56 33	86.18 73	75.40 92	76.17 98	61.32 104	37.43 188	61.08 122	76.52 104	92.35 28	91.64 29	97.46 3	98.86 16

tfpnnormal		69.29 162	65.58 171	73.62 130	79.87 139	84.82 159	76.97 146	75.12 93	45.29 193	49.03 152	35.57 191	37.20 195	68.02 147	82.70 125	81.24 150	92.69 119	92.20 127

CS-MVS		86.48 49	89.02 35	83.52 68	87.37 88	95.52 53	84.21 91	75.09 94	87.63 52	71.30 75	80.94 38	78.24 52	87.23 40	92.72 25	90.05 45	95.95 29	98.03 31

Baseline_NR-MVSNet		70.61 153	68.87 160	72.65 138	75.95 164	80.49 184	75.92 156	74.75 95	65.10 139	48.78 156	41.28 177	44.28 165	68.45 146	78.67 159	79.64 163	92.04 132	92.62 123

IS_MVSNet		80.92 81	84.14 67	77.16 108	87.43 86	93.90 70	80.44 109	74.64 96	75.05 102	61.10 106	65.59 81	76.89 65	67.39 151	90.88 44	90.05 45	91.95 135	96.62 69

EPP-MVSNet		80.82 83	82.79 77	78.52 98	86.31 100	92.37 91	79.83 116	74.51 97	73.79 111	64.46 92	67.01 73	80.63 48	74.33 113	85.63 104	84.35 109	91.68 141	95.79 78

Vis-MVSNet (Re-imp)		78.28 105	82.68 78	73.16 134	86.64 95	92.68 85	78.07 134	74.48 98	74.05 107	53.47 126	64.22 88	76.52 66	54.28 178	88.96 67	88.29 68	92.03 133	94.00 103

DU-MVS		72.19 138	71.35 146	73.17 133	75.95 164	86.02 144	77.01 144	74.42 99	65.39 136	48.82 154	49.10 146	42.81 174	71.61 130	78.67 159	83.10 122	91.22 151	94.37 96

NR-MVSNet		71.47 147	71.11 147	71.90 148	77.73 151	86.02 144	76.88 148	74.42 99	65.39 136	46.09 169	49.10 146	39.87 187	64.27 159	81.40 139	82.24 132	91.99 134	93.75 115

TransMVSNet (Re)		66.87 170	64.30 179	69.88 163	78.32 145	81.35 179	73.88 165	74.34 101	43.19 197	45.20 173	40.12 179	42.37 179	55.97 177	80.85 144	79.15 164	91.56 144	83.06 178

baseline2		81.21 80	83.36 73	78.70 95	83.22 121	92.71 83	80.32 113	74.25 102	80.39 79	63.94 95	68.89 67	68.44 100	74.67 110	89.61 60	86.68 83	95.83 33	96.81 64

thisisatest0530		81.67 77	84.27 65	78.63 96	85.53 105	93.88 71	81.77 104	73.84 103	81.35 74	63.85 96	68.79 68	77.64 57	73.02 124	88.73 71	85.73 95	93.76 70	93.80 114

tttt0517		81.51 78	84.12 68	78.47 102	85.33 109	93.74 74	81.42 108	73.84 103	81.21 76	63.59 97	68.73 69	77.46 60	73.02 124	88.47 73	85.73 95	93.63 79	93.49 118

TDRefinement		67.82 166	64.91 177	71.22 157	82.08 125	81.45 176	77.42 141	73.79 105	59.62 152	48.35 158	42.35 173	42.40 178	60.87 169	74.69 181	74.64 186	84.83 195	79.20 188

baseline		83.83 64	84.38 64	83.18 70	86.65 94	94.59 66	85.79 78	73.78 106	85.83 60	72.94 62	69.28 65	70.80 92	83.45 71	86.80 90	87.59 73	96.47 16	95.77 79

ET-MVSNet_ETH3D		82.37 71	85.68 59	78.51 100	62.90 201	94.66 63	87.06 62	73.57 107	83.13 67	61.52 103	78.37 41	76.01 68	89.99 18	84.14 113	89.03 62	96.03 27	94.42 95

UniMVSNet (Re)		72.12 140	72.28 141	71.93 146	76.77 156	87.38 124	75.73 159	73.51 108	65.76 132	50.24 148	48.65 149	46.49 158	63.85 161	80.10 150	82.47 127	91.49 146	95.13 88

MVS_Test		84.60 62	85.13 61	83.99 64	88.17 79	95.27 58	88.21 56	73.15 109	84.31 64	70.55 80	68.67 70	68.78 98	86.99 43	91.71 36	91.90 24	96.84 10	95.27 87

v148		70.34 154	68.46 163	72.54 141	76.04 162	86.38 134	74.83 161	72.73 110	55.88 163	55.26 123	43.32 168	43.49 168	64.52 158	76.93 172	80.11 161	91.85 138	93.11 119

CDS-MVSNet		76.57 118	76.78 112	76.32 111	80.94 133	89.75 112	82.94 100	72.64 111	59.01 155	62.95 99	58.60 107	62.67 118	66.91 153	86.26 99	87.20 74	91.57 143	93.97 105
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
GBi-Net		80.72 84	80.49 91	81.00 84	78.18 146	86.19 139	86.73 64	72.57 112	83.02 68	72.63 67	56.55 114	73.48 79	80.99 80	86.57 92	86.83 80	94.89 50	90.77 139

test1		80.72 84	80.49 91	81.00 84	78.18 146	86.19 139	86.73 64	72.57 112	83.02 68	72.63 67	56.55 114	73.48 79	80.99 80	86.57 92	86.83 80	94.89 50	90.77 139

FMVSNet3		81.93 76	81.98 85	81.88 76	79.49 142	87.02 125	88.15 58	72.57 112	83.02 68	72.63 67	56.55 114	73.48 79	82.34 77	91.49 39	91.20 34	96.07 22	91.13 136

Vis-MVSNet		77.24 112	79.99 96	74.02 124	84.62 114	93.92 68	80.33 112	72.55 115	62.58 144	55.25 124	64.45 85	69.49 96	57.00 176	88.78 69	88.21 69	94.36 58	92.54 124
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
RPSCF		74.27 124	73.24 136	75.48 115	81.01 132	80.18 186	76.24 152	72.37 116	74.84 105	68.24 86	72.47 54	67.39 102	73.89 114	71.05 190	69.38 196	81.14 201	77.37 190

DCV-MVSNet		79.76 91	79.17 98	80.44 89	84.65 113	84.51 163	84.20 92	72.36 117	75.17 101	70.81 78	66.21 78	66.56 104	80.99 80	82.89 122	84.56 106	89.65 174	94.30 99

DWT-MVSNet_training		82.66 68	83.34 74	81.87 77	88.71 73	92.63 86	82.07 103	72.21 118	86.37 58	72.64 65	64.51 84	71.44 87	80.35 88	84.43 110	87.73 72	95.27 39	96.25 72

FMVSNet2		79.24 94	78.14 105	80.53 88	78.18 146	86.19 139	86.73 64	71.91 119	72.97 113	70.48 81	50.63 140	66.56 104	80.99 80	90.10 51	89.77 51	94.89 50	90.77 139

v2v482		71.73 144	69.80 152	73.99 125	75.88 168	86.66 131	79.58 118	71.90 120	57.58 158	50.41 147	45.35 155	43.24 172	73.05 122	79.69 152	82.18 133	93.08 109	93.87 111

USDC		73.43 130	72.31 140	74.73 119	80.86 134	86.21 137	80.42 110	71.83 121	71.69 118	46.94 161	59.60 105	42.58 176	76.47 105	82.66 126	81.22 151	91.88 137	82.24 184

GA-MVS		73.62 126	74.52 131	72.58 140	79.93 138	89.29 116	78.02 135	71.67 122	60.79 148	42.68 181	54.41 129	49.07 151	70.07 144	89.39 64	86.55 84	93.13 108	92.12 129

pmmvs4		73.38 131	71.53 145	75.55 114	75.95 164	85.24 154	77.25 143	71.59 123	71.03 120	63.10 98	49.09 148	44.22 166	73.73 117	82.04 133	80.18 160	91.68 141	88.89 157

pm-mvs1		69.62 160	68.07 166	71.44 153	77.21 153	85.32 153	76.11 155	71.05 124	46.55 192	51.17 139	41.83 175	48.20 154	61.81 167	84.00 115	81.14 155	91.28 150	89.42 150

CVMVSNet		68.95 164	70.79 148	66.79 174	79.69 141	83.75 168	72.05 172	70.90 125	56.20 161	36.30 191	54.94 128	59.22 129	54.03 181	78.33 161	78.65 168	87.77 187	84.44 171

PMMVS		82.26 72	85.48 60	78.51 100	85.92 104	91.92 94	78.30 132	70.77 126	86.30 59	61.11 105	82.46 36	70.88 91	84.70 61	88.05 80	84.78 103	90.24 169	93.98 104

CMPMVS	50.59 17	66.74 171	62.72 187	71.42 154	85.40 108	89.72 113	72.69 170	70.72 127	51.24 178	51.75 135	38.91 183	44.40 163	63.74 162	70.84 191	71.52 190	84.19 196	72.45 198
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
Effi-MVS+		79.80 90	80.04 93	79.52 93	85.53 105	93.31 80	85.28 81	70.68 128	74.15 106	58.79 116	62.03 98	60.51 125	83.37 72	88.41 75	86.09 90	93.49 90	95.80 77

test-LLR		79.52 93	83.42 70	74.97 117	81.79 126	91.26 98	76.17 153	70.57 129	77.71 90	52.14 133	66.26 76	77.47 58	73.10 120	87.02 87	87.16 75	96.05 25	97.02 53

test0.0.03 1		71.70 145	74.68 129	68.23 167	81.79 126	83.81 167	68.64 177	70.57 129	68.81 127	43.47 178	62.77 94	60.09 127	51.77 188	82.48 128	81.67 139	93.16 106	83.13 177

UniMVSNet_ETH3D		69.49 161	65.86 170	73.72 128	76.51 158	85.88 146	78.65 127	70.52 131	48.08 189	55.71 122	37.64 185	40.56 183	71.38 133	75.05 179	81.49 143	89.57 177	92.29 125

FMVSNet1		74.26 125	71.95 142	76.95 109	74.28 178	83.94 166	83.61 96	69.99 132	57.08 159	65.08 91	42.39 171	57.41 135	76.98 100	86.57 92	86.83 80	91.77 140	89.42 150

V42		71.58 146	70.11 151	73.30 132	75.66 171	86.68 130	79.17 125	69.92 133	59.29 154	52.80 130	44.36 159	45.66 160	68.83 145	79.48 154	81.49 143	93.44 93	93.82 113

HyFIR lowres test		78.08 107	76.81 111	79.56 91	90.77 63	94.64 64	82.97 98	69.85 134	69.81 125	59.53 114	33.52 193	64.66 110	78.97 95	88.77 70	88.38 67	95.27 39	97.86 34

TinyColmap		67.16 168	63.51 183	71.42 154	77.94 149	79.54 189	72.80 169	69.78 135	56.58 160	45.52 170	44.53 158	33.53 200	74.45 112	76.91 173	77.06 178	88.03 186	76.41 191

MS-PatchMatch		77.47 110	76.48 116	78.63 96	89.89 65	90.42 104	85.42 80	69.53 136	70.79 122	60.43 111	50.05 142	70.62 94	70.66 140	86.71 91	82.54 126	95.86 32	84.23 173

FC-MVSNet-test		67.04 169	72.47 139	60.70 191	76.92 154	81.41 177	61.52 191	69.45 137	65.58 135	26.74 204	61.79 100	60.40 126	41.17 196	77.60 168	77.78 174	88.41 183	82.70 181

v1144		70.93 151	69.42 156	72.70 137	75.48 172	86.26 135	79.22 124	69.39 138	55.61 164	48.05 159	43.47 166	42.55 177	71.51 132	82.11 131	81.74 136	92.56 123	94.17 102

WR-MVS		64.98 174	66.59 169	63.09 184	74.34 177	82.68 172	64.98 188	69.17 139	54.42 170	36.18 192	44.32 160	44.35 164	44.65 191	73.60 182	77.83 173	89.21 179	88.96 156

CANet_DTU		83.33 66	86.59 49	79.53 92	88.88 72	94.87 62	86.63 67	68.85 140	85.45 61	50.54 146	77.86 45	69.94 95	85.62 49	92.63 26	90.88 38	96.63 12	94.46 94

v1192		70.32 155	68.77 161	72.12 145	74.76 174	85.62 148	78.73 126	68.53 141	55.08 166	46.34 166	42.39 171	40.67 182	71.90 129	82.27 129	81.53 142	92.43 126	93.86 112

IterMVS-LS		76.80 117	76.33 119	77.35 107	84.07 119	84.11 164	81.54 106	68.52 142	66.17 131	61.74 100	57.84 108	64.31 115	74.88 109	83.48 119	86.21 88	93.34 100	92.16 128
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
MDA-MVSNet-bldmvs		54.99 194	52.66 197	57.71 192	52.74 206	74.87 197	55.61 199	68.41 143	43.65 196	32.54 196	37.93 184	22.11 209	54.11 179	48.85 203	67.34 197	82.85 198	73.88 197

DTE-MVSNet		63.26 187	63.41 185	63.08 185	72.59 183	78.56 190	65.03 187	68.28 144	50.53 183	32.38 197	44.03 162	37.79 193	49.48 189	70.83 192	76.73 180	90.73 161	85.42 169

PEN-MVS		64.35 180	64.29 180	64.42 181	72.67 182	79.83 187	66.97 180	68.24 145	51.21 179	35.29 194	44.09 161	38.51 190	52.36 186	71.06 189	77.65 175	90.99 154	87.68 162

v144192		70.10 156	68.55 162	71.90 148	74.55 175	85.67 147	77.81 136	68.22 146	54.65 168	46.91 162	42.76 169	41.27 181	70.95 137	80.48 147	81.11 156	92.96 113	93.90 109

TAMVS		72.06 143	71.76 144	72.41 142	76.68 157	88.12 120	74.82 162	68.09 147	53.52 172	56.91 119	52.94 135	56.93 137	66.91 153	81.37 140	82.44 128	91.07 153	86.99 164

thisisatest0515		70.62 152	71.94 143	69.07 164	76.48 159	85.59 149	68.03 178	68.02 148	59.70 151	52.94 128	52.19 136	50.36 148	58.10 174	83.15 120	81.63 140	90.87 159	90.99 138

Fast-Effi-MVS+		77.37 111	76.68 114	78.17 103	82.84 123	89.94 111	81.47 107	68.01 149	72.99 112	60.26 112	55.07 126	53.20 142	82.99 73	86.47 97	86.12 89	93.46 92	92.98 122

pmmvs5		70.01 157	69.31 158	70.82 159	75.80 170	86.26 135	72.94 168	67.91 150	53.84 171	47.22 160	47.31 151	41.47 180	67.61 148	83.93 116	81.93 135	93.42 95	90.42 143

v1921920		69.85 159	68.38 164	71.58 151	74.35 176	85.39 152	77.78 137	67.88 151	54.64 169	45.39 171	42.11 174	39.97 186	71.10 135	81.68 138	81.17 154	92.96 113	93.69 117

TESTMET0.1,1		79.15 95	83.42 70	74.18 123	79.81 140	91.26 98	76.17 153	67.83 152	77.71 90	52.14 133	66.26 76	77.47 58	73.10 120	87.02 87	87.16 75	96.05 25	97.02 53

Effi-MVS+-dtu		74.57 123	74.60 130	74.53 121	81.38 130	86.74 129	80.39 111	67.70 153	67.36 129	53.06 127	59.86 104	57.50 134	75.84 107	80.19 149	78.62 169	88.79 181	91.95 132

Anonymous20231206		62.05 190	61.83 190	62.30 188	72.09 186	77.84 192	63.10 190	67.62 154	50.20 184	36.68 188	29.59 200	37.05 197	43.90 193	77.33 171	77.31 176	90.41 165	83.49 175

CP-MVSNet		64.84 176	64.97 175	64.69 179	72.09 186	81.04 182	66.66 181	67.53 155	52.45 175	37.40 187	44.00 164	38.37 191	53.54 183	72.26 186	76.93 179	90.94 158	89.75 148

v8		71.42 148	69.69 153	73.43 131	76.45 160	85.12 157	79.53 120	67.47 156	59.34 153	52.90 129	44.60 157	45.82 159	71.05 136	79.56 153	81.45 145	93.17 105	91.96 131

PS-CasMVS		64.22 183	64.19 181	64.25 182	71.86 188	80.67 183	66.42 183	67.43 157	50.64 181	36.48 189	42.60 170	37.46 194	52.56 185	71.98 187	76.69 181	90.76 160	89.29 154

v1240		69.28 163	67.82 167	71.00 158	74.09 179	85.13 156	76.54 151	67.28 158	53.17 173	44.70 175	41.55 176	39.38 188	70.51 142	81.29 141	81.18 152	92.88 117	93.02 121

WR-MVS_H		64.14 184	65.36 173	62.71 186	72.47 184	82.33 174	65.13 185	66.99 159	51.81 177	36.47 190	43.33 167	42.77 175	43.99 192	72.41 185	75.99 183	91.20 152	88.86 158

N_pmnet		60.52 191	58.83 194	62.50 187	68.97 194	75.61 196	59.72 196	66.47 160	51.90 176	41.26 183	35.42 192	35.63 198	52.25 187	67.07 197	70.08 194	86.35 190	76.10 192

v10		70.97 150	69.44 154	72.75 136	75.90 167	84.58 162	79.43 121	66.45 161	58.07 157	49.93 149	43.87 165	43.68 167	71.91 128	82.04 133	81.70 137	92.89 116	92.11 130

pmmvs-eth3d		64.24 181	61.96 189	66.90 173	66.35 196	76.04 195	66.09 184	66.31 162	52.59 174	50.94 143	37.61 186	32.79 202	62.43 166	75.78 176	75.48 184	89.27 178	83.39 176

testgi		63.11 188	64.88 178	61.05 189	75.83 169	78.51 191	60.42 193	66.20 163	48.77 187	34.56 195	56.96 111	40.35 184	40.95 197	77.46 170	77.22 177	88.37 185	74.86 196

test20.03		57.93 193	59.22 193	56.44 193	71.84 189	73.78 198	53.55 201	65.96 164	43.02 198	28.46 201	37.50 187	38.17 192	30.41 202	75.25 177	74.42 188	88.41 183	72.37 199

pmmvs6		64.24 181	61.77 191	67.12 172	72.39 185	81.39 178	71.33 174	65.95 165	36.05 202	48.48 157	30.55 195	43.45 170	58.75 172	77.88 167	76.36 182	85.83 192	86.70 166

test-mter		77.90 108	82.44 82	72.60 139	78.52 144	90.24 107	73.85 166	65.31 166	76.37 97	51.29 137	65.58 82	75.94 69	71.36 134	85.98 102	86.26 87	95.26 41	96.71 68

v7n		66.43 172	65.51 172	67.51 170	71.63 190	83.10 169	70.89 176	65.02 167	50.13 185	44.68 176	39.59 180	38.77 189	62.57 165	77.59 169	78.91 165	90.29 168	90.44 142

LTVRE_ROB	63.07 16	64.49 179	63.16 186	66.04 175	77.47 152	82.64 173	70.98 175	65.02 167	34.01 206	29.61 200	49.12 145	35.58 199	70.57 141	75.10 178	78.45 170	82.60 199	87.24 163
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
SixPastTwentyTwo		63.75 185	63.42 184	64.13 183	72.91 181	80.34 185	61.29 192	63.90 169	49.58 186	40.42 184	54.99 127	37.13 196	60.90 168	68.46 194	70.80 191	85.37 194	82.65 183

FMVSNet5		72.83 133	73.89 134	71.59 150	67.42 195	76.28 193	75.88 157	63.74 170	77.27 93	54.59 125	53.32 132	71.48 86	73.85 115	81.95 135	81.69 138	94.06 63	75.20 194

EG-PatchMatch MVS		66.23 173	65.20 174	67.43 171	77.74 150	86.20 138	72.51 171	63.68 171	43.95 195	43.44 179	36.22 190	45.43 162	54.04 180	81.00 142	80.95 157	93.15 107	82.67 182

anonymousdsp		67.61 167	68.94 159	66.04 175	71.44 191	83.97 165	66.45 182	63.53 172	50.54 182	42.42 182	49.39 144	45.63 161	62.84 164	77.99 164	81.34 149	89.59 176	93.75 115

MIMVSNet1		52.76 196	53.95 195	51.38 198	41.96 209	70.79 201	53.56 200	63.03 173	39.36 200	27.83 203	22.73 204	33.07 201	34.47 201	70.49 193	72.69 189	87.41 188	68.51 201

PM-MVS		63.52 186	62.51 188	64.70 178	64.79 200	76.08 194	65.07 186	62.08 174	58.13 156	46.56 165	44.98 156	31.31 203	62.89 163	72.58 184	69.93 195	86.81 189	84.55 170

new-patchmatchnet		53.91 195	52.69 196	55.33 196	64.83 199	70.90 200	52.24 202	61.75 175	41.09 199	30.82 198	29.90 198	28.22 205	36.69 199	61.52 198	65.08 198	85.64 193	72.14 200

MDTV_nov1_ep13		77.20 113	80.04 93	73.90 126	82.22 124	90.14 109	79.25 123	61.52 176	78.63 87	56.98 118	65.52 83	72.80 82	73.05 122	80.93 143	83.20 121	90.36 166	89.05 155

EU-MVSNet		58.73 192	60.92 192	56.17 194	66.17 197	72.39 199	58.85 197	61.24 177	48.47 188	27.91 202	46.70 153	40.06 185	39.07 198	68.27 195	70.34 193	83.77 197	80.23 186

IterMVS-SCA-FT		72.18 139	73.96 133	70.11 162	80.15 137	81.11 181	77.42 141	61.09 178	63.67 142	46.73 164	55.77 123	59.15 130	63.95 160	82.83 124	83.70 118	91.31 149	91.49 135

IterMVS		72.43 136	74.05 132	70.55 160	80.34 136	81.17 180	77.44 140	61.00 179	63.57 143	46.82 163	55.88 120	59.09 131	65.03 157	83.15 120	83.83 117	92.67 121	91.65 134
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
CHOSEN 280x420		82.15 74	85.87 56	77.80 105	86.54 97	93.42 79	81.74 105	59.96 180	78.99 86	63.99 94	74.50 50	83.95 34	80.99 80	89.53 61	85.01 99	93.56 83	95.71 81

CostFormer		80.72 84	81.81 88	79.44 94	86.50 98	91.65 97	84.31 90	59.84 181	80.86 77	72.69 64	62.46 95	73.74 76	79.93 91	82.58 127	84.50 108	93.37 98	96.90 62

tpm cat1		76.93 115	76.19 120	77.79 106	85.08 112	88.58 119	82.96 99	59.33 182	75.72 100	72.64 65	51.25 138	64.41 112	75.74 108	77.90 165	80.10 162	90.97 155	95.35 84

FPMVS		50.25 199	45.67 201	55.58 195	70.48 193	60.12 205	59.78 195	59.33 182	46.66 191	37.94 185	30.22 197	27.51 206	35.94 200	50.98 202	47.90 202	70.02 204	56.31 203

dps		75.76 121	75.02 126	76.63 110	84.51 115	88.12 120	77.51 139	58.33 184	75.91 99	71.98 72	57.37 110	57.85 133	76.81 103	77.89 166	78.40 171	90.63 163	89.63 149

Fast-Effi-MVS+-dtu		73.56 127	75.32 125	71.50 152	80.35 135	86.83 127	79.72 117	58.07 185	67.64 128	44.83 174	60.28 103	54.07 140	73.59 119	81.90 137	82.30 130	92.46 125	94.18 101

PMVS	36.83 18	40.62 200	36.39 202	45.56 200	58.40 202	33.20 209	32.62 209	56.02 186	28.25 207	37.92 186	22.29 205	26.15 208	25.29 204	48.49 204	43.82 205	63.13 207	52.53 206
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
tpm		73.50 128	74.85 127	71.93 146	83.19 122	86.84 126	78.61 129	55.91 187	65.64 133	48.90 153	56.30 117	61.09 121	72.31 126	79.10 156	80.61 158	92.68 120	94.35 98

PatchmatchNet		76.85 116	80.03 95	73.15 135	84.08 118	91.04 102	77.76 138	55.85 188	79.43 82	52.74 131	62.08 97	76.02 67	74.56 111	79.92 151	81.41 147	93.92 67	90.29 144
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
EPMVS		77.16 114	79.08 100	74.92 118	86.73 92	91.98 93	78.62 128	55.44 189	79.43 82	56.59 120	61.24 101	70.73 93	76.97 101	80.59 146	81.43 146	95.15 45	88.17 160

tpmrst		76.27 120	77.65 109	74.66 120	86.13 103	89.53 114	79.31 122	54.91 190	77.19 94	56.27 121	55.87 121	64.58 111	77.25 98	80.85 144	80.21 159	94.07 62	95.32 85

SCA		76.41 119	79.90 97	72.35 143	84.26 117	85.24 154	75.57 160	54.56 191	79.95 80	52.72 132	64.22 88	77.84 54	73.73 117	80.48 147	81.37 148	93.25 103	90.20 145

MDTV_nov1_ep13_2view		64.72 177	64.94 176	64.46 180	71.14 192	81.94 175	67.53 179	54.54 192	55.92 162	43.29 180	44.02 163	43.27 171	59.87 170	71.85 188	74.77 185	90.36 166	82.82 180

CR-MVSNet		74.84 122	77.91 106	71.26 156	81.77 128	85.52 150	78.32 130	54.14 193	74.05 107	51.09 140	50.00 143	71.38 88	70.77 138	86.48 95	84.03 114	91.46 147	93.92 107

Patchmtry						87.41 123	78.32 130	54.14 193		51.09 140

MIMVSNet		68.66 165	69.43 155	67.76 169	64.92 198	84.68 161	74.16 164	54.10 195	60.85 147	51.27 138	39.47 181	49.48 149	67.48 150	84.86 109	85.57 97	94.63 54	81.10 185

gm-plane-assit		64.86 175	68.15 165	61.02 190	76.44 161	68.29 202	41.60 207	53.37 196	34.68 205	26.19 206	33.22 194	57.09 136	71.97 127	95.12 5	93.97 7	96.54 14	94.66 92

MVS-HIRNet		64.63 178	64.03 182	65.33 177	75.01 173	82.84 170	58.54 198	52.10 197	55.42 165	49.29 150	29.83 199	43.48 169	66.97 152	78.28 162	78.81 166	90.07 171	79.52 187

PatchT		72.66 134	76.58 115	68.09 168	79.02 143	86.09 143	59.81 194	51.78 198	72.00 117	51.09 140	46.84 152	66.70 103	70.77 138	86.48 95	84.03 114	96.07 22	93.92 107

ADS-MVSNet		72.11 141	73.72 135	70.24 161	81.24 131	86.59 132	74.75 163	50.56 199	72.58 115	49.17 151	55.40 124	61.46 120	73.80 116	76.01 175	78.14 172	91.93 136	85.86 168

RPMNet		73.46 129	77.85 107	68.34 166	81.71 129	85.52 150	73.83 167	50.54 200	74.05 107	46.10 168	53.03 134	71.91 84	66.31 155	83.55 117	82.18 133	91.55 145	94.71 90

gg-mvs-nofinetune		72.10 142	74.79 128	68.97 165	83.31 120	95.22 59	85.66 79	48.77 201	35.68 203	22.17 208	30.49 196	77.73 56	76.37 106	94.30 12	93.03 11	97.55 2	97.05 52

Gipuma		35.20 201	33.96 203	36.65 202	43.30 208	32.51 210	26.96 211	48.31 202	38.87 201	20.08 209	8.08 207	7.41 213	26.44 203	53.60 200	58.43 201	54.81 208	38.79 208
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
new_pmnet		50.32 198	51.36 199	49.11 199	49.19 207	64.89 203	48.66 205	47.99 203	47.55 190	26.27 205	29.51 201	28.66 204	44.89 190	61.12 199	62.74 200	77.66 202	65.03 202

pmmvs3		52.59 197	52.43 198	52.78 197	54.53 205	64.49 204	50.07 203	46.89 204	35.31 204	30.19 199	27.27 202	26.96 207	53.02 184	67.28 196	70.54 192	81.96 200	75.20 194

EMVS		20.61 205	16.32 207	25.62 206	36.41 210	18.93 214	11.51 213	43.75 205	15.65 209	6.53 212	7.56 210	4.68 214	22.03 205	14.56 209	23.10 208	33.51 211	29.77 210

E-PMN		21.42 203	17.56 206	25.94 205	36.25 211	19.02 213	11.56 212	43.72 206	15.25 210	6.99 211	8.04 208	4.53 215	21.77 206	16.13 208	26.16 207	35.34 210	33.77 209

DeepMVS_CX						48.96 207	43.77 206	40.58 207	50.93 180	24.67 207	36.95 189	20.18 210	41.60 194	38.92 206		52.37 209	53.31 205

PMMVS2		32.52 202	33.92 204	30.88 204	34.15 212	44.70 208	27.79 210	39.69 208	22.21 208	4.31 213	15.73 206	14.13 211	12.45 209	40.11 205	47.00 203	66.88 205	53.54 204

MVE	25.07 19	21.25 204	23.51 205	18.62 207	15.07 213	29.77 212	10.67 214	34.60 209	12.51 211	9.46 210	7.84 209	3.82 216	14.38 208	27.45 207	42.42 206	27.56 212	40.74 207
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
tmp_tt				39.78 201	56.31 203	31.71 211	35.84 208	15.08 210	82.57 72	50.83 145	63.07 92	47.51 156	15.28 207	52.23 201	44.24 204	65.35 206

GG-mvs-BLEND		62.08 189	88.31 41	31.46 203	0.16 214	98.10 7	91.57 38	0.09 211	85.07 63	0.21 214	73.90 53	83.74 36	0.19 212	88.98 66	89.39 57	96.58 13	99.02 14

testmvs		0.76 206	1.23 208	0.21 208	0.05 215	0.21 215	0.38 216	0.09 211	0.94 212	0.05 215	2.13 212	0.08 217	0.60 211	0.82 210	0.77 209	0.11 213	3.62 212

test123		0.67 207	1.11 209	0.16 209	0.01 216	0.14 216	0.20 217	0.04 213	0.77 213	0.02 216	2.15 211	0.02 218	0.61 210	0.23 211	0.72 210	0.07 214	3.76 211

sosnet-low-res		0.00 208	0.00 210	0.00 210	0.00 217	0.00 217	0.00 218	0.00 214	0.00 214	0.00 217	0.00 213	0.00 219	0.00 213	0.00 212	0.00 211	0.00 215	0.00 213

sosnet		0.00 208	0.00 210	0.00 210	0.00 217	0.00 217	0.00 218	0.00 214	0.00 214	0.00 217	0.00 213	0.00 219	0.00 213	0.00 212	0.00 211	0.00 215	0.00 213

our_test_3					73.80 180	79.57 188	64.47 189

test_part1																	99.34 7

ambc			50.35 200		55.61 204	59.93 206	48.73 204		44.08 194	35.81 193	24.01 203	10.64 212	41.57 195	72.83 183	63.35 199	74.99 203	77.61 189

MTAPA										91.14 5		85.84 23

MTMP										90.95 6		84.13 32

Patchmatch-RL test							8.17 215

XVS					89.65 67	95.93 44	85.97 76			76.32 50		82.05 42				93.51 87

X-MVStestdata					89.65 67	95.93 44	85.97 76			76.32 50		82.05 42				93.51 87

mPP-MVS					95.90 30							80.22 50

NP-MVS									89.55 43

High-resolution multi-view benchmark