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.

APDe-MVS		77.58 2	82.93 2	71.35 2	77.86 2	80.55 2	83.38 1	57.61 7	85.57 1	61.11 15	86.10 4	82.98 4	64.76 2	78.29 12	76.78 19	83.40 6	90.20 2

ESAPD		78.19 1	83.74 1	71.72 1	79.01 1	81.38 1	83.23 2	58.63 2	83.92 5	62.44 12	87.06 2	85.82 1	64.54 3	79.39 5	77.99 8	82.44 17	90.61 1

SMA-MVS		77.34 3	82.65 3	71.13 3	75.33 10	80.39 3	82.14 3	58.49 3	84.51 4	63.89 5	78.09 11	83.76 2	63.31 7	81.19 1	80.62 1	83.60 4	90.03 3

SteuartSystems-ACMMP		75.23 9	79.60 11	70.13 9	76.81 3	78.92 8	81.74 4	57.99 5	75.30 25	59.83 21	75.69 14	78.45 19	60.48 25	80.58 2	79.77 2	83.94 3	88.52 6
Skip Steuart: Steuart Systems R&D Blog.
CNVR-MVS		75.62 8	79.91 10	70.61 5	75.76 6	78.82 10	81.66 5	57.12 9	79.77 13	63.04 8	70.69 20	81.15 10	62.99 8	80.23 3	79.54 3	83.11 7	89.16 4

ACMMP_Plus		76.15 5	81.17 5	70.30 6	74.09 15	79.47 6	81.59 6	57.09 10	81.38 7	63.89 5	79.02 9	80.48 13	62.24 14	80.05 4	79.12 4	82.94 10	88.64 5

APD-MVS		75.80 7	80.90 7	69.86 12	75.42 9	78.48 12	81.43 7	57.44 8	80.45 11	59.32 22	85.28 5	80.82 12	63.96 5	76.89 25	76.08 24	81.58 33	88.30 8
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
HSP-MVS		76.78 4	82.44 4	70.19 8	75.26 11	80.22 4	80.59 8	57.85 6	84.79 3	60.84 16	88.54 1	83.43 3	66.24 1	78.21 15	76.47 21	80.34 38	85.43 27

TSAR-MVS + MP.		75.22 10	80.06 9	69.56 13	74.61 13	72.74 45	80.59 8	55.70 20	80.80 9	62.65 10	86.25 3	82.92 5	62.07 16	76.89 25	75.66 27	81.77 29	85.19 29

NCCC		74.27 15	77.83 21	70.13 9	75.70 7	77.41 18	80.51 10	57.09 10	78.25 17	62.28 13	65.54 33	78.26 20	62.18 15	79.13 6	78.51 6	83.01 9	87.68 12

HFP-MVS		74.87 11	78.86 16	70.21 7	73.99 16	77.91 14	80.36 11	56.63 12	78.41 16	64.27 3	74.54 16	77.75 23	62.96 9	78.70 9	77.82 10	83.02 8	86.91 16

HPM-MVS++		76.01 6	80.47 8	70.81 4	76.60 4	74.96 31	80.18 12	58.36 4	81.96 6	63.50 7	78.80 10	82.53 7	64.40 4	78.74 8	78.84 5	81.81 27	87.46 13

train_agg		73.89 18	78.25 18	68.80 19	75.25 12	72.27 47	79.75 13	56.05 17	74.87 28	58.97 23	81.83 7	79.76 16	61.05 22	77.39 23	76.01 25	81.71 30	85.61 25

zzz-MVS		74.25 16	77.97 20	69.91 11	73.43 19	74.06 39	79.69 14	56.44 14	80.74 10	64.98 2	68.72 26	79.98 15	62.92 10	78.24 14	77.77 12	81.99 25	86.30 18

OPM-MVS		69.33 33	71.05 41	67.32 23	72.34 24	75.70 28	79.57 15	56.34 16	55.21 64	53.81 49	59.51 53	68.96 47	59.67 29	77.61 21	76.44 22	82.19 22	83.88 36

ACMMPR		73.79 20	78.41 17	68.40 20	72.35 23	77.79 15	79.32 16	56.38 15	77.67 20	58.30 27	74.16 17	76.66 24	61.40 19	78.32 11	77.80 11	82.68 14	86.51 17

PGM-MVS		72.89 22	77.13 23	67.94 21	72.47 22	77.25 19	79.27 17	54.63 24	73.71 30	57.95 29	72.38 18	75.33 29	60.75 23	78.25 13	77.36 15	82.57 16	85.62 24

MP-MVS		74.31 14	78.87 14	68.99 17	73.49 18	78.56 11	79.25 18	56.51 13	75.33 23	60.69 18	75.30 15	79.12 18	61.81 17	77.78 19	77.93 9	82.18 23	88.06 10

CSCG		74.68 12	79.22 12	69.40 14	75.69 8	80.01 5	79.12 19	52.83 36	79.34 14	63.99 4	70.49 21	82.02 8	60.35 27	77.48 22	77.22 16	84.38 1	87.97 11

SD-MVS		74.43 13	78.87 14	69.26 16	74.39 14	73.70 41	79.06 20	55.24 22	81.04 8	62.71 9	80.18 8	82.61 6	61.70 18	75.43 36	73.92 39	82.44 17	85.22 28

MCST-MVS		73.67 21	77.39 22	69.33 15	76.26 5	78.19 13	78.77 21	54.54 25	75.33 23	59.99 20	67.96 28	79.23 17	62.43 13	78.00 16	75.71 26	84.02 2	87.30 14

TSAR-MVS + ACMM		72.56 24	79.07 13	64.96 36	73.24 20	73.16 44	78.50 22	48.80 59	79.34 14	55.32 36	85.04 6	81.49 9	58.57 33	75.06 39	73.75 40	75.35 103	85.61 25

ACMMP		71.57 26	75.84 26	66.59 26	70.30 34	76.85 24	78.46 23	53.95 29	73.52 31	55.56 34	70.13 22	71.36 42	58.55 34	77.00 24	76.23 23	82.71 13	85.81 23
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
DeepC-MVS	66.32 2	73.85 19	78.10 19	68.90 18	67.92 44	79.31 7	78.16 24	59.28 1	78.24 18	61.13 14	67.36 32	76.10 27	63.40 6	79.11 7	78.41 7	83.52 5	88.16 9
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
CDPH-MVS		71.47 27	75.82 27	66.41 27	72.97 21	77.15 20	78.14 25	54.71 23	69.88 43	53.07 51	70.98 19	74.83 31	56.95 44	76.22 29	76.57 20	82.62 15	85.09 30

CP-MVS		72.63 23	76.95 24	67.59 22	70.67 30	75.53 29	77.95 26	56.01 18	75.65 22	58.82 24	69.16 25	76.48 25	60.46 26	77.66 20	77.20 17	81.65 31	86.97 15

HQP-MVS		70.88 29	75.02 29	66.05 30	71.69 26	74.47 36	77.51 27	53.17 33	72.89 32	54.88 40	70.03 23	70.48 44	57.26 40	76.02 31	75.01 31	81.78 28	86.21 19

X-MVS		71.18 28	75.66 28	65.96 31	71.71 25	76.96 21	77.26 28	55.88 19	72.75 33	54.48 44	64.39 37	74.47 32	54.19 56	77.84 18	77.37 14	82.21 21	85.85 22

LGP-MVS_train		68.87 35	72.03 37	65.18 35	69.33 38	74.03 40	76.67 29	53.88 30	68.46 44	52.05 54	63.21 39	63.89 59	56.31 46	75.99 32	74.43 35	82.83 12	84.18 32

DeepC-MVS_fast	65.08 3	72.00 25	76.11 25	67.21 24	68.93 40	77.46 16	76.54 30	54.35 26	74.92 27	58.64 26	65.18 34	74.04 37	62.62 11	77.92 17	77.02 18	82.16 24	86.21 19
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
ACMM	60.30 7	67.58 43	68.82 52	66.13 29	70.59 31	72.01 49	76.54 30	54.26 27	65.64 49	54.78 42	50.35 85	61.72 67	58.74 32	75.79 34	75.03 29	81.88 26	81.17 47
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
CPTT-MVS		68.76 37	73.01 33	63.81 43	65.42 55	73.66 42	76.39 32	52.08 38	72.61 34	50.33 57	60.73 50	72.65 40	59.43 31	73.32 45	72.12 45	79.19 49	85.99 21

ACMP	61.42 5	68.72 38	71.37 39	65.64 33	69.06 39	74.45 37	75.88 33	53.30 32	68.10 45	55.74 33	61.53 49	62.29 64	56.97 43	74.70 40	74.23 37	82.88 11	84.31 31
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
DeepPCF-MVS	66.49 1	74.25 16	80.97 6	66.41 27	67.75 46	78.87 9	75.61 34	54.16 28	84.86 2	58.22 28	77.94 12	81.01 11	62.52 12	78.34 10	77.38 13	80.16 41	88.40 7

AdaColmap		67.89 41	68.85 51	66.77 25	73.73 17	74.30 38	75.28 35	53.58 31	70.24 41	57.59 30	51.19 83	59.19 76	60.74 24	75.33 38	73.72 41	79.69 45	77.96 60

MSLP-MVS++		68.17 39	70.72 44	65.19 34	69.41 37	70.64 51	74.99 36	45.76 67	70.20 42	60.17 19	56.42 62	73.01 38	61.14 20	72.80 47	70.54 51	79.70 43	81.42 46

MVS_0304		69.49 32	73.96 31	64.28 41	67.92 44	76.13 27	74.90 37	47.60 61	63.29 53	54.09 48	67.44 31	76.35 26	59.53 30	75.81 33	75.03 29	81.62 32	83.70 37

PCF-MVS	59.98 8	67.32 44	71.04 42	62.97 45	64.77 57	74.49 35	74.78 38	49.54 53	67.44 46	54.39 47	58.35 57	72.81 39	55.79 52	71.54 52	69.24 59	78.57 51	83.41 38
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
MAR-MVS		68.04 40	70.74 43	64.90 37	71.68 27	76.33 26	74.63 39	50.48 50	63.81 51	55.52 35	54.88 69	69.90 46	57.39 39	75.42 37	74.79 33	79.71 42	80.03 50
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
XVS					70.49 32	76.96 21	74.36 40			54.48 44		74.47 32				82.24 19

X-MVStestdata					70.49 32	76.96 21	74.36 40			54.48 44		74.47 32				82.24 19

PHI-MVS		69.27 34	74.84 30	62.76 46	66.83 48	74.83 32	73.88 42	49.32 55	70.61 40	50.93 55	69.62 24	74.84 30	57.25 41	75.53 35	74.32 36	78.35 55	84.17 33

CANet		68.77 36	73.01 33	63.83 42	68.30 41	75.19 30	73.73 43	47.90 60	63.86 50	54.84 41	67.51 30	74.36 35	57.62 37	74.22 42	73.57 43	80.56 37	82.36 41

3Dnovator+	62.63 4	69.51 31	72.62 35	65.88 32	68.21 43	76.47 25	73.50 44	52.74 37	70.85 39	58.65 25	55.97 64	69.95 45	61.11 21	76.80 27	75.09 28	81.09 36	83.23 40

abl_6				64.36 40	70.08 35	77.45 17	72.88 45	50.15 51	71.31 38	54.77 43	62.79 41	77.99 22	56.80 45			81.50 34	83.91 35

TSAR-MVS + GP.		69.71 30	73.92 32	64.80 38	68.27 42	70.56 52	71.90 46	50.75 46	71.38 37	57.46 31	68.68 27	75.42 28	60.10 28	73.47 44	73.99 38	80.32 39	83.97 34

CLD-MVS		67.02 45	71.57 38	61.71 47	71.01 29	74.81 33	71.62 47	38.91 166	71.86 36	60.70 17	64.97 35	67.88 53	51.88 94	76.77 28	74.98 32	76.11 95	69.75 120
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
3Dnovator	60.86 6	66.99 46	70.32 46	63.11 44	66.63 49	74.52 34	71.56 48	45.76 67	67.37 47	55.00 39	54.31 73	68.19 51	58.49 35	73.97 43	73.63 42	81.22 35	80.23 49

MVS_111021_HR		67.62 42	70.39 45	64.39 39	69.77 36	70.45 53	71.44 49	51.72 42	60.77 58	55.06 38	62.14 46	66.40 56	58.13 36	76.13 30	74.79 33	80.19 40	82.04 44

DELS-MVS		65.87 47	70.30 47	60.71 48	64.05 64	72.68 46	70.90 50	45.43 71	57.49 61	49.05 61	64.43 36	68.66 48	55.11 54	74.31 41	73.02 44	79.70 43	81.51 45
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
OMC-MVS		65.16 50	71.35 40	57.94 61	52.95 159	68.82 57	69.00 51	38.28 173	79.89 12	55.20 37	62.76 42	68.31 50	56.14 49	71.30 54	68.70 64	76.06 97	79.67 51

Effi-MVS+		63.28 53	65.96 58	60.17 50	64.26 61	68.06 61	68.78 52	45.71 69	54.08 68	46.64 67	55.92 65	63.13 62	55.94 50	70.38 61	71.43 47	79.68 46	78.70 55

MSDG		58.46 88	58.97 116	57.85 65	66.27 53	66.23 81	67.72 53	42.33 137	53.43 70	43.68 96	43.39 153	45.35 154	49.75 100	68.66 72	67.77 71	77.38 63	67.96 133

tpmp4_e23		56.84 110	57.14 129	56.49 91	62.45 66	62.05 117	67.57 54	41.56 146	54.17 67	48.57 63	49.18 89	46.54 147	50.44 99	61.93 164	58.82 178	68.34 171	67.28 138

EPNet		65.14 51	69.54 49	60.00 51	66.61 50	67.67 66	67.53 55	55.32 21	62.67 55	46.22 71	67.74 29	65.93 57	48.07 110	72.17 49	72.12 45	76.28 87	78.47 57
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
ACMH+	53.71 12	59.26 68	60.28 86	58.06 58	64.17 63	68.46 59	67.51 56	50.93 45	52.46 77	35.83 130	40.83 179	45.12 157	52.32 90	69.88 64	69.00 62	77.59 61	76.21 78

LS3D		60.20 65	61.70 74	58.45 56	64.18 62	67.77 63	67.19 57	48.84 58	61.67 56	41.27 109	45.89 133	51.81 115	54.18 57	68.78 69	66.50 90	75.03 104	69.48 126

v2v482		58.69 79	60.12 96	57.03 70	57.16 126	66.05 88	67.17 58	43.52 107	46.33 120	45.19 81	49.46 88	51.02 118	52.51 85	67.30 112	66.03 95	76.61 80	74.62 96

v1144		58.88 73	60.16 93	57.39 68	58.03 85	67.26 70	67.14 59	44.46 81	45.17 137	44.33 94	47.81 116	49.92 126	53.20 70	67.77 97	66.62 86	77.15 66	76.58 72

v7		59.19 70	60.62 79	57.53 66	57.96 86	67.19 72	67.09 60	44.28 82	46.84 115	45.45 78	48.19 111	51.06 117	53.62 60	67.84 92	66.59 87	76.79 69	76.60 71

v10		59.17 71	60.60 80	57.50 67	57.95 87	66.73 76	67.09 60	44.11 84	46.85 114	45.42 79	48.18 113	51.07 116	53.63 59	67.84 92	66.59 87	76.79 69	76.92 67

CostFormer		56.57 111	59.13 114	53.60 102	57.52 96	61.12 129	66.94 62	35.95 186	53.44 69	44.68 88	55.87 66	54.44 92	48.21 108	60.37 172	58.33 181	68.27 173	70.33 118

v13		58.44 90	59.72 105	56.94 73	57.55 91	63.51 103	66.86 63	42.81 126	45.90 129	44.98 85	48.17 114	51.87 111	52.68 79	68.20 80	65.78 100	76.78 71	74.63 95

V9		58.45 89	59.75 101	56.92 77	57.51 97	63.49 105	66.86 63	42.73 128	46.07 126	45.05 83	48.45 106	51.99 109	52.66 81	68.04 90	65.75 102	76.72 74	74.50 98

V14		58.44 90	59.75 101	56.90 80	57.48 99	63.46 106	66.85 65	42.68 129	46.16 123	45.03 84	48.57 104	52.04 108	52.65 82	67.93 91	65.72 105	76.69 75	74.40 100

v15		58.43 93	59.75 101	56.88 82	57.45 100	63.44 107	66.84 66	42.65 130	46.24 122	45.07 82	48.68 103	52.07 107	52.63 83	67.84 92	65.70 106	76.65 76	74.31 103

v12		58.44 90	59.74 104	56.92 77	57.54 93	63.50 104	66.84 66	42.77 127	45.96 127	44.95 86	48.31 107	51.94 110	52.67 80	68.14 83	65.75 102	76.75 73	74.55 97

v1141		58.56 82	60.05 98	56.81 86	57.36 106	66.18 83	66.80 68	43.11 118	45.87 131	44.60 90	48.71 101	51.83 113	52.38 87	67.46 106	65.64 110	76.63 77	74.66 92

divwei89l23v2f112		58.56 82	60.05 98	56.81 86	57.36 106	66.18 83	66.80 68	43.11 118	45.89 130	44.60 90	48.71 101	51.84 112	52.38 87	67.45 108	65.65 107	76.63 77	74.66 92

v1		58.56 82	60.06 97	56.83 85	57.36 106	66.19 82	66.80 68	43.10 120	45.87 131	44.68 88	48.73 100	51.83 113	52.38 87	67.45 108	65.65 107	76.63 77	74.66 92

v6		58.89 72	60.54 82	56.96 72	57.34 111	66.13 85	66.71 71	42.84 123	47.85 109	45.80 74	49.04 91	52.95 100	52.79 74	67.53 102	65.59 113	76.26 88	74.73 89

ACMH	52.42 13	58.24 97	59.56 109	56.70 88	66.34 52	69.59 54	66.71 71	49.12 56	46.08 125	28.90 157	42.67 167	41.20 186	52.60 84	71.39 53	70.28 53	76.51 82	75.72 80
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
v1neww		58.88 73	60.54 82	56.94 73	57.33 113	66.13 85	66.70 73	42.84 123	47.84 110	45.74 76	49.02 93	52.93 101	52.78 75	67.53 102	65.59 113	76.26 88	74.73 89

v7new		58.88 73	60.54 82	56.94 73	57.33 113	66.13 85	66.70 73	42.84 123	47.84 110	45.74 76	49.02 93	52.93 101	52.78 75	67.53 102	65.59 113	76.26 88	74.73 89

v8		58.88 73	60.57 81	56.92 77	57.35 109	65.69 90	66.69 75	42.64 131	47.89 108	45.77 75	49.04 91	52.98 99	52.77 77	67.51 105	65.57 117	76.26 88	75.30 85

v17		58.69 79	60.19 92	56.94 73	57.38 104	63.37 109	66.67 76	42.47 135	48.52 105	46.10 73	48.90 98	53.00 98	52.84 71	67.58 99	65.60 112	76.19 92	74.38 101

v16		58.71 78	60.20 89	56.97 71	57.35 109	63.36 110	66.67 76	42.49 133	48.69 101	46.36 69	48.87 99	52.92 103	52.82 73	67.57 100	65.58 116	76.15 94	74.38 101

v18		58.68 81	60.20 89	56.90 80	57.26 122	63.28 111	66.58 78	42.42 136	48.86 95	46.37 68	49.01 95	53.05 97	52.74 78	67.40 110	65.52 118	76.02 99	74.28 104

MVS_Test		62.40 58	66.23 57	57.94 61	59.77 78	64.77 98	66.50 79	41.76 141	57.26 62	49.33 59	62.68 43	67.47 55	53.50 64	68.57 74	66.25 92	76.77 72	76.58 72

PVSNet_Blended_VisFu		63.65 52	66.92 53	59.83 52	60.03 75	73.44 43	66.33 80	48.95 57	52.20 78	50.81 56	56.07 63	60.25 72	53.56 61	73.23 46	70.01 56	79.30 47	83.24 39

V42		56.97 107	60.14 94	53.28 105	48.16 182	62.78 116	66.30 81	37.93 175	47.44 112	42.68 102	48.19 111	52.59 105	51.90 93	67.46 106	65.94 97	72.72 130	76.55 74

v11		58.19 99	59.47 110	56.70 88	57.54 93	63.42 108	66.28 82	42.49 133	45.62 135	44.59 92	48.16 115	50.78 121	52.84 71	67.80 96	65.76 101	76.49 83	74.76 88

v1192		58.51 85	59.66 106	57.17 69	57.82 88	67.72 64	66.21 83	44.83 77	44.15 144	43.49 97	46.68 121	47.94 130	53.55 62	67.39 111	66.51 89	77.13 67	77.20 65

DI_MVS_plusplus_trai		61.88 59	65.17 63	58.06 58	60.05 74	65.26 93	66.03 84	44.22 83	55.75 63	46.73 66	54.64 71	68.12 52	54.13 58	69.13 67	66.66 84	77.18 65	76.61 70

Effi-MVS+-dtu		60.34 64	62.32 73	58.03 60	64.31 59	67.44 69	65.99 85	42.26 138	49.55 86	42.00 106	48.92 97	59.79 74	56.27 47	68.07 86	67.03 77	77.35 64	75.45 83

CNLPA		62.78 56	66.31 56	58.65 55	58.47 83	68.41 60	65.98 86	41.22 151	78.02 19	56.04 32	46.65 122	59.50 75	57.50 38	69.67 65	65.27 122	72.70 136	76.67 69

MVS_111021_LR		63.05 54	66.43 55	59.10 54	61.33 69	63.77 102	65.87 87	43.58 105	60.20 59	53.70 50	62.09 47	62.38 63	55.84 51	70.24 62	68.08 68	74.30 108	78.28 59

v144192		58.23 98	59.40 112	56.87 83	57.56 90	66.89 74	65.70 88	45.01 76	44.06 145	42.88 99	46.61 123	48.09 129	53.49 65	66.94 117	65.90 98	76.61 80	77.29 63

QAPM		65.27 49	69.49 50	60.35 49	65.43 54	72.20 48	65.69 89	47.23 62	63.46 52	49.14 60	53.56 74	71.04 43	57.01 42	72.60 48	71.41 48	77.62 59	82.14 43

Fast-Effi-MVS+-dtu		56.30 112	59.29 113	52.82 111	58.64 82	64.89 96	65.56 90	32.89 204	45.80 133	35.04 132	45.89 133	54.14 93	49.41 101	67.16 114	66.45 91	75.37 102	70.69 115

TSAR-MVS + COLMAP		62.65 57	69.90 48	54.19 100	46.31 191	66.73 76	65.49 91	41.36 149	76.57 21	46.31 70	76.80 13	56.68 84	53.27 69	69.50 66	66.65 85	72.40 142	76.36 77

TAPA-MVS	54.74 10	60.85 62	66.61 54	54.12 101	47.38 187	65.33 91	65.35 92	36.51 184	75.16 26	48.82 62	54.70 70	63.51 60	53.31 68	68.36 75	64.97 127	73.37 121	74.27 105
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
Fast-Effi-MVS+		60.36 63	63.35 70	56.87 83	58.70 80	65.86 89	65.08 93	37.11 179	53.00 75	45.36 80	52.12 79	56.07 89	56.27 47	71.28 55	69.42 58	78.71 50	75.69 81

MS-PatchMatch		58.19 99	60.20 89	55.85 93	65.17 56	64.16 100	64.82 94	41.48 148	50.95 80	42.17 105	45.38 138	56.42 85	48.08 109	68.30 76	66.70 83	73.39 120	69.46 128

diffmvs		59.53 66	64.04 69	54.26 99	55.09 139	59.86 142	64.80 95	39.55 165	58.39 60	46.21 72	60.48 51	67.82 54	49.27 102	63.53 151	63.32 147	70.64 157	74.89 87

HyFIR lowres test		56.87 109	58.60 119	54.84 97	56.62 129	69.27 56	64.77 96	42.21 139	45.66 134	37.50 123	33.08 198	57.47 83	53.33 67	65.46 143	67.94 69	74.60 105	71.35 111

v1921920		57.89 102	59.02 115	56.58 90	57.55 91	66.66 79	64.72 97	44.70 79	43.55 148	42.73 101	46.17 131	46.93 144	53.51 63	66.78 120	65.75 102	76.29 86	77.28 64

canonicalmvs		65.62 48	72.06 36	58.11 57	63.94 65	71.05 50	64.49 98	43.18 116	74.08 29	47.35 64	64.17 38	71.97 41	51.17 97	71.87 50	70.74 49	78.51 53	80.56 48

PLC	52.09 14	59.21 69	62.47 72	55.41 96	53.24 158	64.84 97	64.47 99	40.41 159	65.92 48	44.53 93	46.19 130	55.69 90	55.33 53	68.24 79	65.30 121	74.50 106	71.09 112
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
OpenMVS	57.13 9	62.81 55	65.75 59	59.39 53	66.47 51	69.52 55	64.26 100	43.07 121	61.34 57	50.19 58	47.29 119	64.41 58	54.60 55	70.18 63	68.62 66	77.73 57	78.89 54

CANet_DTU		58.88 73	64.68 66	52.12 115	55.77 132	66.75 75	63.92 101	37.04 180	53.32 71	37.45 124	59.81 52	61.81 66	44.43 126	68.25 77	67.47 75	74.12 111	75.33 84

v1240		57.55 103	58.63 118	56.29 92	57.30 119	66.48 80	63.77 102	44.56 80	42.77 164	42.48 103	45.64 136	46.28 149	53.46 66	66.32 128	65.80 99	76.16 93	77.13 66

tpm cat1		53.30 132	53.41 160	53.17 108	58.16 84	59.15 152	63.73 103	38.27 174	50.73 82	46.98 65	45.57 137	44.00 172	49.20 103	55.90 200	54.02 199	62.65 193	64.50 167

PVSNet_BlendedMVS		61.63 60	64.82 64	57.91 63	57.21 124	67.55 67	63.47 104	46.08 65	54.72 65	52.46 52	58.59 55	60.73 69	51.82 95	70.46 59	65.20 124	76.44 84	76.50 75

PVSNet_Blended		61.63 60	64.82 64	57.91 63	57.21 124	67.55 67	63.47 104	46.08 65	54.72 65	52.46 52	58.59 55	60.73 69	51.82 95	70.46 59	65.20 124	76.44 84	76.50 75

CHOSEN 1792x2688		55.85 114	58.01 123	53.33 104	57.26 122	62.82 115	63.29 106	41.55 147	46.65 117	38.34 118	34.55 196	53.50 94	52.43 86	67.10 115	67.56 74	67.13 178	73.92 107

IB-MVS	54.11 11	58.36 95	60.70 78	55.62 94	58.67 81	68.02 62	61.56 107	43.15 117	46.09 124	44.06 95	44.24 145	50.99 120	48.71 105	66.70 121	70.33 52	77.60 60	78.50 56
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
MVSTER		57.19 104	61.11 77	52.62 112	50.82 175	58.79 155	61.55 108	37.86 176	48.81 97	41.31 108	57.43 61	52.10 106	48.60 106	68.19 82	66.75 82	75.56 101	75.68 82

EG-PatchMatch MVS		56.98 106	58.24 122	55.50 95	64.66 58	68.62 58	61.48 109	43.63 104	38.44 197	41.44 107	38.05 189	46.18 151	43.95 127	71.71 51	70.61 50	77.87 56	74.08 106

v148		55.58 117	57.61 127	53.20 107	54.59 148	61.86 118	61.18 110	38.70 171	44.30 143	42.25 104	47.53 117	50.24 125	48.73 104	65.15 145	62.61 157	73.79 114	71.61 110

GA-MVS		55.67 115	58.33 120	52.58 113	55.23 137	63.09 112	61.08 111	40.15 161	42.95 155	37.02 126	52.61 77	47.68 133	47.51 112	65.92 136	65.35 119	74.49 107	70.68 116

v7n		55.67 115	57.46 128	53.59 103	56.06 130	65.29 92	61.06 112	43.26 115	40.17 184	37.99 120	40.79 180	45.27 156	47.09 114	67.67 98	66.21 93	76.08 96	76.82 68

v52		53.60 128	56.74 132	49.93 125	45.54 194	61.64 121	60.65 113	36.99 181	38.75 193	36.32 128	39.64 184	47.13 140	47.05 115	66.89 118	65.65 107	73.04 125	77.48 61

V4		53.60 128	56.73 133	49.93 125	45.54 194	61.64 121	60.65 113	36.99 181	38.74 195	36.33 127	39.64 184	47.12 141	47.05 115	66.89 118	65.64 110	73.04 125	77.48 61

DWT-MVSNet_training		53.80 127	54.31 152	53.21 106	57.65 89	59.04 153	60.65 113	40.11 162	46.35 119	42.77 100	49.07 90	41.07 187	51.06 98	58.62 182	58.96 177	67.00 181	67.06 139

COLMAP_ROB	46.52 15	51.99 144	54.86 147	48.63 143	49.13 180	61.73 120	60.53 116	36.57 183	53.14 72	32.95 138	37.10 190	38.68 199	40.49 142	65.72 139	63.08 150	72.11 147	64.60 166
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
pmmvs4		54.66 125	56.07 134	53.00 109	54.63 145	57.08 166	60.43 117	44.10 85	51.69 79	40.55 111	46.55 126	44.79 164	45.95 121	62.54 156	63.66 142	72.36 144	66.20 148

Vis-MVSNet		58.48 87	65.70 60	50.06 124	53.40 157	67.20 71	60.24 118	43.32 113	48.83 96	30.23 150	62.38 45	61.61 68	40.35 143	71.03 57	69.77 57	72.82 129	79.11 53
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
IterMVS-LS		58.30 96	61.39 75	54.71 98	59.92 77	58.40 159	59.42 119	43.64 102	48.71 99	40.25 113	57.53 60	58.55 78	52.15 92	65.42 144	65.34 120	72.85 127	75.77 79
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
TDRefinement		49.31 162	52.44 167	45.67 165	30.44 225	59.42 146	59.24 120	39.78 164	48.76 98	31.20 145	35.73 193	29.90 215	42.81 134	64.24 150	62.59 158	70.55 158	66.43 144

IterMVS		53.45 131	57.12 130	49.17 131	49.23 179	60.93 130	59.05 121	34.63 190	44.53 139	33.22 136	51.09 84	51.01 119	48.38 107	62.43 157	60.79 167	70.54 159	69.05 131
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
EPP-MVSNet		59.39 67	65.45 61	52.32 114	60.96 71	67.70 65	58.42 122	44.75 78	49.71 85	27.23 169	59.03 54	62.20 65	43.34 132	70.71 58	69.13 60	79.25 48	79.63 52

MDTV_nov1_ep13		50.32 158	52.43 168	47.86 153	49.87 178	54.70 175	58.10 123	34.29 192	45.59 136	37.71 121	47.44 118	47.42 137	41.86 137	58.07 185	55.21 192	65.34 186	58.56 191

TranMVSNet+NR-MVSNet		55.87 113	60.14 94	50.88 118	59.46 79	63.82 101	57.93 124	52.98 34	48.94 94	20.52 190	52.87 76	47.33 138	36.81 164	69.12 68	69.03 61	77.56 62	69.89 119

NR-MVSNet		55.35 119	59.46 111	50.56 120	61.33 69	62.97 113	57.91 125	51.80 40	48.62 103	20.59 189	51.99 80	44.73 165	34.10 174	68.58 73	68.64 65	77.66 58	70.67 117

UniMVSNet_NR-MVSNet		56.94 108	61.14 76	52.05 116	60.02 76	65.21 95	57.44 126	52.93 35	49.37 89	24.31 180	54.62 72	50.54 122	39.04 147	68.69 70	68.84 63	78.53 52	70.72 113

DU-MVS		55.41 118	59.59 107	50.54 121	54.60 146	62.97 113	57.44 126	51.80 40	48.62 103	24.31 180	51.99 80	47.00 143	39.04 147	68.11 84	67.75 72	76.03 98	70.72 113

IS_MVSNet		57.95 101	64.26 68	50.60 119	61.62 68	65.25 94	57.18 128	45.42 72	50.79 81	26.49 171	57.81 59	60.05 73	34.51 171	71.24 56	70.20 55	78.36 54	74.44 99

GBi-Net		55.20 120	60.25 87	49.31 128	52.42 161	61.44 123	57.03 129	44.04 87	49.18 91	30.47 146	48.28 108	58.19 79	38.22 150	68.05 87	66.96 78	73.69 116	69.65 121

test1		55.20 120	60.25 87	49.31 128	52.42 161	61.44 123	57.03 129	44.04 87	49.18 91	30.47 146	48.28 108	58.19 79	38.22 150	68.05 87	66.96 78	73.69 116	69.65 121

FMVSNet2		55.04 123	59.95 100	49.31 128	52.42 161	61.44 123	57.03 129	44.08 86	49.55 86	30.40 149	46.89 120	58.84 77	38.22 150	67.07 116	66.21 93	73.69 116	69.65 121

FMVSNet1		54.08 126	58.68 117	48.71 142	50.90 174	61.35 126	56.73 132	43.94 91	45.91 128	29.32 156	42.72 166	56.26 88	37.70 154	68.05 87	66.96 78	73.69 116	69.50 125

FMVSNet3		54.78 124	59.58 108	49.17 131	52.37 164	61.31 127	56.72 133	44.04 87	49.18 91	30.47 146	48.28 108	58.19 79	38.09 153	65.48 142	65.20 124	73.31 122	69.45 129

v748		52.93 133	55.29 142	50.19 123	51.90 168	61.31 127	56.54 134	40.05 163	39.12 191	34.82 135	39.93 183	43.83 173	43.66 128	64.26 149	63.32 147	74.15 110	75.28 86

UGNet		57.03 105	65.25 62	47.44 156	46.54 190	66.73 76	56.30 135	43.28 114	50.06 83	32.99 137	62.57 44	63.26 61	33.31 177	68.25 77	67.58 73	72.20 146	78.29 58
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
Baseline_NR-MVSNet		53.50 130	57.89 124	48.37 145	54.60 146	59.25 151	56.10 136	51.84 39	49.32 90	17.92 200	45.38 138	47.68 133	36.93 163	68.11 84	65.95 96	72.84 128	69.57 124

tpm		48.82 168	51.27 179	45.96 163	54.10 153	47.35 200	56.05 137	30.23 209	46.70 116	43.21 98	52.54 78	47.55 136	37.28 161	54.11 205	50.50 210	54.90 211	60.12 186

pmmvs-eth3d		51.33 150	52.25 170	50.26 122	50.82 175	54.65 176	56.03 138	43.45 112	43.51 149	37.20 125	39.20 187	39.04 198	42.28 135	61.85 165	62.78 154	71.78 151	64.72 165

FC-MVSNet-train		58.40 94	63.15 71	52.85 110	64.29 60	61.84 119	55.98 139	46.47 63	53.06 73	34.96 133	61.95 48	56.37 87	39.49 145	68.67 71	68.36 67	75.92 100	71.81 109

UniMVSNet (Re)		55.15 122	60.39 85	49.03 134	55.31 134	64.59 99	55.77 140	50.63 47	48.66 102	20.95 188	51.47 82	50.40 123	34.41 173	67.81 95	67.89 70	77.11 68	71.88 108

thres200		52.39 139	55.37 141	48.90 139	57.39 103	60.18 135	55.60 141	43.73 99	42.93 156	27.41 167	43.35 154	45.09 158	36.61 165	66.36 126	63.92 141	72.66 137	65.78 154

CDS-MVSNet		52.42 138	57.06 131	47.02 158	53.92 155	58.30 161	55.50 142	46.47 63	42.52 167	29.38 155	49.50 87	52.85 104	28.49 192	66.70 121	66.89 81	68.34 171	62.63 176
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
thres400		52.38 140	55.51 136	48.74 141	57.49 98	60.10 139	55.45 143	43.54 106	42.90 157	26.72 170	43.34 155	45.03 163	36.61 165	66.20 133	64.53 133	72.66 137	66.43 144

tfpn111		52.44 137	55.38 139	49.01 135	57.31 115	60.24 132	55.42 144	43.77 93	42.85 158	27.51 163	42.03 173	45.06 159	37.32 156	66.38 123	64.54 129	72.71 133	66.54 141

conf0.01		52.02 143	54.62 149	49.00 137	57.30 119	60.17 137	55.42 144	43.76 96	42.85 158	27.49 165	43.12 159	39.71 195	37.32 156	66.26 131	64.54 129	72.72 130	65.66 156

conf0.002		51.76 148	54.13 154	49.00 137	57.28 121	60.15 138	55.42 144	43.75 98	42.85 158	27.49 165	43.13 158	37.12 207	37.32 156	66.23 132	64.17 136	72.72 130	65.24 160

conf200view11		52.51 136	55.51 136	49.01 135	57.31 115	60.24 132	55.42 144	43.77 93	42.85 158	27.51 163	43.00 162	45.06 159	37.32 156	66.38 123	64.54 129	72.71 133	66.54 141

tfpn200view9		52.53 135	55.51 136	49.06 133	57.31 115	60.24 132	55.42 144	43.77 93	42.85 158	27.81 161	43.00 162	45.06 159	37.32 156	66.38 123	64.54 129	72.71 133	66.54 141

thres100view900		52.04 142	54.81 148	48.80 140	57.31 115	59.33 147	55.30 149	42.92 122	42.85 158	27.81 161	43.00 162	45.06 159	36.99 162	64.74 147	63.51 144	72.47 141	65.21 161

MDTV_nov1_ep13_2view		47.62 179	49.72 191	45.18 167	48.05 183	53.70 179	54.90 150	33.80 196	39.90 186	29.79 153	38.85 188	41.89 180	39.17 146	58.99 176	55.55 189	65.34 186	59.17 189

view600		51.96 145	55.13 144	48.27 147	57.41 102	60.05 140	54.74 151	43.64 102	42.57 166	25.88 173	43.11 160	44.48 168	35.34 168	66.27 129	63.61 143	72.61 140	65.80 153

thres600view7		51.91 147	55.14 143	48.14 148	57.43 101	60.18 135	54.60 152	43.73 99	42.61 165	25.20 175	43.10 161	44.47 169	35.19 169	66.36 126	63.28 149	72.66 137	66.01 151

tfpnnormal		50.16 159	52.19 171	47.78 154	56.86 128	58.37 160	54.15 153	44.01 90	38.35 199	25.94 172	36.10 192	37.89 201	34.50 172	65.93 135	63.42 145	71.26 154	65.28 159

TransMVSNet (Re)		51.92 146	55.38 139	47.88 152	60.95 72	59.90 141	53.95 154	45.14 74	39.47 188	24.85 177	43.87 148	46.51 148	29.15 188	67.55 101	65.23 123	73.26 124	65.16 162

dps		50.42 156	51.20 182	49.51 127	55.88 131	56.07 168	53.73 155	38.89 167	43.66 146	40.36 112	45.66 135	37.63 204	45.23 123	59.05 175	56.18 184	62.94 192	60.16 185

anonymousdsp		52.84 134	57.78 125	47.06 157	40.24 210	58.95 154	53.70 156	33.54 199	36.51 204	32.69 139	43.88 147	45.40 153	47.97 111	67.17 113	70.28 53	74.22 109	82.29 42

UA-Net		58.50 86	64.68 66	51.30 117	66.97 47	67.13 73	53.68 157	45.65 70	49.51 88	31.58 144	62.91 40	68.47 49	35.85 167	68.20 80	67.28 76	74.03 112	69.24 130

view800		51.55 149	54.89 146	47.66 155	57.37 105	59.77 144	53.62 158	43.72 101	42.22 168	24.94 176	42.80 165	43.81 174	33.94 175	66.09 134	64.38 135	72.39 143	65.14 163

tpmrst		48.08 175	49.88 190	45.98 162	52.71 160	48.11 198	53.62 158	33.70 197	48.70 100	39.74 114	48.96 96	46.23 150	40.29 144	50.14 214	49.28 212	55.80 208	57.71 193

gg-mvs-nofinetune		49.07 167	52.56 166	45.00 168	61.99 67	59.78 143	53.55 160	41.63 142	31.62 214	12.08 210	29.56 207	53.28 96	29.57 187	66.27 129	64.49 134	71.19 155	62.92 173

PatchmatchNet		49.92 161	51.29 178	48.32 146	51.83 169	51.86 187	53.38 161	37.63 178	47.90 107	40.83 110	48.54 105	45.30 155	45.19 124	56.86 189	53.99 201	61.08 197	54.57 202
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
USDC		51.11 151	53.71 156	48.08 150	44.76 197	55.99 169	53.01 162	40.90 153	52.49 76	36.14 129	44.67 143	33.66 211	43.27 133	63.23 152	61.10 163	70.39 160	64.82 164

tfpn		50.58 154	53.65 158	47.00 159	57.34 111	59.31 148	52.41 163	43.76 96	41.81 172	23.86 182	42.49 168	37.80 202	32.63 180	65.68 141	64.02 139	71.99 149	64.41 168

conf0.05thres1000		50.64 153	53.84 155	46.92 160	57.02 127	59.29 149	52.29 164	43.80 92	39.84 187	23.81 183	39.26 186	43.14 177	32.52 181	65.74 138	64.04 137	72.05 148	65.53 157

EPNet_dtu		52.05 141	58.26 121	44.81 169	54.10 153	50.09 193	52.01 165	40.82 156	53.03 74	27.41 167	54.90 68	57.96 82	26.72 197	62.97 153	62.70 156	67.78 175	66.19 149
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
TinyColmap		47.08 184	47.56 198	46.52 161	42.35 206	53.44 180	51.77 166	40.70 157	43.44 151	31.92 142	29.78 206	23.72 226	45.04 125	61.99 163	59.54 175	67.35 177	61.03 181

pm-mvs1		51.02 152	55.55 135	45.73 164	54.16 152	58.52 157	50.92 167	42.56 132	40.32 183	25.67 174	43.66 150	50.34 124	30.06 186	65.85 137	63.97 140	70.99 156	66.21 147

MIMVSNet		43.79 197	48.53 194	38.27 201	41.46 207	48.97 196	50.81 168	32.88 205	44.55 138	22.07 184	32.05 199	47.15 139	24.76 200	58.73 179	56.09 186	57.63 205	52.14 205

PatchMatch-RL		50.11 160	51.56 175	48.43 144	46.23 192	51.94 186	50.21 169	38.62 172	46.62 118	37.51 122	42.43 169	39.38 196	52.24 91	60.98 168	59.56 174	65.76 183	60.01 187

test-LLR		49.28 163	50.29 186	48.10 149	55.26 135	47.16 201	49.52 170	43.48 110	39.22 189	31.98 140	43.65 151	47.93 131	41.29 140	56.80 190	55.36 190	67.08 179	61.94 177

TESTMET0.1,1		46.09 191	50.29 186	41.18 191	36.91 215	47.16 201	49.52 170	20.32 228	39.22 189	31.98 140	43.65 151	47.93 131	41.29 140	56.80 190	55.36 190	67.08 179	61.94 177

pmmvs6		48.35 171	51.64 173	44.51 172	51.92 167	57.94 163	49.44 172	42.17 140	34.45 207	24.62 179	28.87 211	46.90 145	29.07 190	64.60 148	63.08 150	69.83 162	65.68 155

PMMVS		49.20 166	54.28 153	43.28 179	34.13 219	45.70 209	48.98 173	26.09 222	46.31 121	34.92 134	55.22 67	53.47 95	47.48 113	59.43 174	59.04 176	68.05 174	60.77 182

thresconf0.02		48.17 174	51.22 181	44.60 171	55.14 138	55.73 170	48.95 174	41.35 150	43.43 152	21.23 186	42.03 173	37.25 206	31.19 183	62.33 160	60.61 169	69.76 163	57.17 195

GG-mvs-BLEND		36.62 212	53.39 161	17.06 230	0.01 238	58.61 156	48.63 175	0.01 236	47.13 113	0.02 241	43.98 146	60.64 71	0.03 236	54.92 204	51.47 209	53.64 214	56.99 196

CR-MVSNet		50.47 155	52.61 165	47.98 151	49.03 181	52.94 181	48.27 176	38.86 168	44.41 140	39.59 115	44.34 144	44.65 167	46.63 118	58.97 177	60.31 170	65.48 184	62.66 174

Patchmtry						47.61 199	48.27 176	38.86 168		39.59 115

pmmvs5		47.07 185	51.02 184	42.46 184	45.18 196	51.47 188	48.23 178	33.09 203	38.17 200	28.62 159	46.60 124	43.48 175	30.74 184	58.28 183	58.63 180	68.92 169	60.48 183

SixPastTwentyTwo		47.55 183	50.25 188	44.41 173	47.30 188	54.31 178	47.81 179	40.36 160	33.76 208	19.93 192	43.75 149	32.77 213	42.07 136	59.82 173	60.94 166	68.98 168	66.37 146

test-mter		45.30 192	50.37 185	39.38 197	33.65 221	46.99 203	47.59 180	18.59 230	38.75 193	28.00 160	43.28 156	46.82 146	41.50 139	57.28 188	55.78 187	66.93 182	63.70 171

EPMVS		44.66 194	47.86 197	40.92 192	47.97 184	44.70 211	47.58 181	33.27 200	48.11 106	29.58 154	49.65 86	44.38 170	34.65 170	51.71 209	47.90 216	52.49 216	48.57 216

CMPMVS	37.70 17	49.24 164	52.71 164	45.19 166	45.97 193	51.23 189	47.44 182	29.31 211	43.04 154	44.69 87	34.45 197	48.35 128	43.64 129	62.59 155	59.82 173	60.08 198	69.48 126
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
FMVSNet5		40.96 202	45.81 202	35.29 209	34.30 218	44.55 212	47.28 183	28.84 213	40.76 180	21.62 185	29.85 205	42.44 178	24.77 199	57.53 187	55.00 193	54.93 210	50.56 210

tfpn_ndepth		48.34 172	52.27 169	43.76 174	54.35 150	56.46 167	47.24 184	40.92 152	43.45 150	21.04 187	41.16 178	43.22 176	28.90 191	61.57 166	60.65 168	70.12 161	59.34 188

PEN-MVS		49.21 165	54.32 151	43.24 180	54.33 151	59.26 150	47.04 185	51.37 44	41.67 173	9.97 218	46.22 129	41.80 181	22.97 206	60.52 170	64.03 138	73.73 115	66.75 140

CP-MVSNet		48.37 170	53.53 159	42.34 185	51.35 172	58.01 162	46.56 186	50.54 48	41.62 174	10.61 214	46.53 127	40.68 191	23.18 203	58.71 180	61.83 159	71.81 150	67.36 137

PS-CasMVS		48.18 173	53.25 163	42.27 186	51.26 173	57.94 163	46.51 187	50.52 49	41.30 177	10.56 216	45.35 140	40.34 193	23.04 205	58.66 181	61.79 160	71.74 152	67.38 136

tfpn_n400		47.56 181	51.56 175	42.90 182	54.91 141	55.28 173	46.21 188	41.59 143	41.51 175	18.54 195	42.25 170	41.54 183	27.12 194	62.41 158	61.02 164	69.05 166	56.90 197

tfpnconf		47.56 181	51.56 175	42.90 182	54.91 141	55.28 173	46.21 188	41.59 143	41.51 175	18.54 195	42.25 170	41.54 183	27.12 194	62.41 158	61.02 164	69.05 166	56.90 197

tfpnview11		47.58 180	51.57 174	42.92 181	54.94 140	55.30 172	46.21 188	41.58 145	42.10 169	18.54 195	42.25 170	41.54 183	27.12 194	62.29 161	61.12 162	69.15 165	56.40 199

CVMVSNet		46.38 190	52.01 172	39.81 195	42.40 205	50.26 191	46.15 191	37.68 177	40.03 185	15.09 205	46.56 125	47.56 135	33.72 176	56.50 194	55.65 188	63.80 190	67.53 134

PM-MVS		44.55 195	48.13 196	40.37 194	32.85 223	46.82 205	46.11 192	29.28 212	40.48 182	29.99 151	39.98 182	34.39 210	41.80 138	56.08 198	53.88 203	62.19 195	65.31 158

RPMNet		46.41 188	48.72 193	43.72 175	47.77 185	52.94 181	46.02 193	33.92 194	44.41 140	31.82 143	36.89 191	37.42 205	37.41 155	53.88 206	54.02 199	65.37 185	61.47 179

Vis-MVSNet (Re-imp)		50.37 157	57.73 126	41.80 188	57.53 95	54.35 177	45.70 194	45.24 73	49.80 84	13.43 208	58.23 58	56.42 85	20.11 209	62.96 154	63.36 146	68.76 170	58.96 190

FPMVS		38.36 211	40.41 214	35.97 206	38.92 212	39.85 216	45.50 195	25.79 223	41.13 178	18.70 194	30.10 204	24.56 221	31.86 182	49.42 219	46.80 221	55.04 209	51.03 208

RPSCF		46.41 188	54.42 150	37.06 205	25.70 233	45.14 210	45.39 196	20.81 227	62.79 54	35.10 131	44.92 142	55.60 91	43.56 130	56.12 197	52.45 206	51.80 217	63.91 170

TAMVS		44.02 196	49.18 192	37.99 203	47.03 189	45.97 208	45.04 197	28.47 214	39.11 192	20.23 191	43.22 157	48.52 127	28.49 192	58.15 184	57.95 183	58.71 200	51.36 207

CHOSEN 280x420		40.80 203	45.05 205	35.84 208	32.95 222	29.57 229	44.98 198	23.71 225	37.54 202	18.42 198	31.36 202	47.07 142	46.41 120	56.71 192	54.65 197	48.55 222	58.47 192

MDA-MVSNet-bldmvs		41.36 201	43.15 209	39.27 199	28.74 227	52.68 183	44.95 199	40.84 155	32.89 210	18.13 199	31.61 201	22.09 228	38.97 149	50.45 213	56.11 185	64.01 189	56.23 200

WR-MVS_H		47.65 178	53.67 157	40.63 193	51.45 170	59.74 145	44.71 200	49.37 54	40.69 181	7.61 225	46.04 132	44.34 171	17.32 212	57.79 186	61.18 161	73.30 123	65.86 152

DTE-MVSNet		48.03 177	53.28 162	41.91 187	54.64 144	57.50 165	44.63 201	51.66 43	41.02 179	7.97 224	46.26 128	40.90 188	20.24 208	60.45 171	62.89 153	72.33 145	63.97 169

tfpn1000		46.75 187	51.24 180	41.51 189	54.39 149	55.60 171	43.85 202	40.90 153	41.82 171	16.71 202	41.26 177	41.58 182	23.96 201	60.76 169	60.27 172	69.26 164	57.42 194

WR-MVS		48.78 169	55.06 145	41.45 190	55.50 133	60.40 131	43.77 203	49.99 52	41.92 170	8.10 223	45.24 141	45.56 152	17.47 211	61.57 166	64.60 128	73.85 113	66.14 150

Anonymous20231206		42.28 199	45.89 201	38.07 202	51.96 166	48.98 195	43.66 204	38.81 170	38.74 195	14.32 207	26.74 213	40.90 188	20.94 207	56.64 193	54.67 196	58.71 200	54.59 201

LTVRE_ROB	44.17 16	47.06 186	50.15 189	43.44 177	51.39 171	58.42 158	42.90 205	43.51 108	22.27 229	14.85 206	41.94 176	34.57 209	45.43 122	62.28 162	62.77 155	62.56 194	68.83 132
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
test0.0.03 1		43.15 198	46.95 199	38.72 200	55.26 135	50.56 190	42.48 206	43.48 110	38.16 201	15.11 204	35.07 195	44.69 166	16.47 214	55.95 199	54.34 198	59.54 199	49.87 214

LP		40.79 204	41.99 210	39.38 197	40.98 208	46.49 207	42.14 207	33.66 198	35.37 206	29.89 152	29.30 209	27.81 217	32.74 178	52.55 207	52.19 207	56.87 206	50.23 211

ADS-MVSNet		40.67 206	43.38 208	37.50 204	44.36 199	39.79 217	42.09 208	32.67 206	44.34 142	28.87 158	40.76 181	40.37 192	30.22 185	48.34 224	45.87 222	46.81 225	44.21 220

Anonymous20231211		40.75 205	41.57 211	39.80 196	54.71 143	52.32 185	41.42 209	45.09 75	24.45 224	6.80 226	14.58 228	23.43 227	23.08 204	56.20 196	58.74 179	67.68 176	61.31 180

ambc			45.54 204		50.66 177	52.63 184	40.99 210		38.36 198	24.67 178	22.62 220	13.94 233	29.14 189	65.71 140	58.06 182	58.60 202	67.43 135

PatchT		48.08 175	51.03 183	44.64 170	42.96 204	50.12 192	40.36 211	35.09 188	43.17 153	39.59 115	42.00 175	39.96 194	46.63 118	58.97 177	60.31 170	63.21 191	62.66 174

EU-MVSNet		40.63 207	45.65 203	34.78 210	39.11 211	46.94 204	40.02 212	34.03 193	33.50 209	10.37 217	35.57 194	37.80 202	23.65 202	51.90 208	50.21 211	61.49 196	63.62 172

test20.03		40.38 208	44.20 206	35.92 207	53.73 156	49.05 194	38.54 213	43.49 109	32.55 211	9.54 219	27.88 212	39.12 197	12.24 225	56.28 195	54.69 195	57.96 204	49.83 215

MIMVSNet1		35.51 213	41.41 212	28.63 218	27.53 229	43.36 213	38.09 214	33.82 195	32.01 212	6.77 227	21.63 223	35.43 208	11.97 227	55.05 203	53.99 201	53.59 215	48.36 217

gm-plane-assit		44.74 193	45.95 200	43.33 178	60.88 73	46.79 206	36.97 215	32.24 208	24.15 225	11.79 211	29.26 210	32.97 212	46.64 117	65.09 146	62.95 152	71.45 153	60.42 184

testpf		34.85 215	36.16 220	33.31 211	47.49 186	35.56 225	36.85 216	32.31 207	23.08 226	15.63 203	29.39 208	29.48 216	19.62 210	41.38 227	41.07 226	47.95 223	53.18 203

N_pmnet		32.67 219	36.85 217	27.79 220	40.55 209	32.13 228	35.80 217	26.79 220	37.24 203	9.10 220	32.02 200	30.94 214	16.30 215	47.22 225	41.21 225	38.21 228	37.21 225

MVS-HIRNet		42.24 200	41.15 213	43.51 176	44.06 203	40.74 214	35.77 218	35.35 187	35.38 205	38.34 118	25.63 215	38.55 200	43.48 131	50.77 211	47.03 220	64.07 188	49.98 212

testgi		38.71 210	43.64 207	32.95 212	52.30 165	48.63 197	35.59 219	35.05 189	31.58 215	9.03 222	30.29 203	40.75 190	11.19 230	55.30 201	53.47 204	54.53 213	45.48 218

test2356		33.40 217	36.53 218	29.76 215	37.51 214	38.39 220	34.68 220	27.35 216	27.88 217	10.61 214	25.54 216	24.44 222	17.15 213	49.99 216	48.32 214	51.24 218	41.16 224

pmmvs3		35.10 214	38.47 215	31.17 214	26.37 232	40.47 215	34.51 221	18.09 231	24.75 223	16.88 201	23.05 219	26.69 219	32.69 179	50.73 212	51.60 208	58.46 203	51.98 206

PMVS	27.84 18	33.81 216	35.28 221	32.09 213	34.13 219	24.81 232	32.51 222	26.48 221	26.41 222	19.37 193	23.76 218	24.02 225	25.18 198	50.78 210	47.24 219	54.89 212	49.95 213
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
1111		31.35 220	33.52 224	28.83 217	44.28 201	32.44 226	31.71 223	33.25 201	27.87 218	10.92 212	22.18 221	24.05 223	15.89 216	49.03 222	44.09 223	36.94 230	34.96 226

.test1245		22.44 228	22.23 229	22.67 226	44.28 201	32.44 226	31.71 223	33.25 201	27.87 218	10.92 212	22.18 221	24.05 223	15.89 216	49.03 222	0.01 234	0.00 238	0.06 236

FC-MVSNet-test		39.65 209	48.35 195	29.49 216	44.43 198	39.28 218	30.23 225	40.44 158	43.59 147	3.12 236	53.00 75	42.03 179	10.02 232	55.09 202	54.77 194	48.66 221	50.71 209

testmv		30.97 222	34.42 222	26.95 221	36.49 216	37.38 223	29.80 226	27.28 217	22.34 227	4.72 230	20.63 225	20.64 229	13.22 222	49.86 218	47.74 217	50.20 219	42.36 222

test1235678		30.97 222	34.42 222	26.95 221	36.49 216	37.38 223	29.79 227	27.28 217	22.33 228	4.72 230	20.62 226	20.64 229	13.22 222	49.87 217	47.74 217	50.20 219	42.36 222

new-patchmatchnet		33.24 218	37.20 216	28.62 219	44.32 200	38.26 222	29.68 228	36.05 185	31.97 213	6.33 228	26.59 214	27.33 218	11.12 231	50.08 215	41.05 227	44.23 226	45.15 219

testus		31.33 221	36.31 219	25.52 224	37.55 213	38.40 219	25.87 229	23.58 226	26.46 221	5.97 229	24.15 217	24.92 220	12.44 224	49.14 221	48.21 215	47.73 224	42.86 221

no-one		29.19 224	31.89 225	26.05 223	30.96 224	38.33 221	21.54 230	29.86 210	15.84 233	3.56 233	11.28 232	13.03 234	14.44 221	38.96 228	52.83 205	55.96 207	52.92 204

test12356		23.91 226	28.47 226	18.60 227	26.80 231	28.30 230	20.92 231	19.76 229	19.89 230	2.88 238	18.48 227	16.57 232	4.05 233	42.34 226	41.93 224	37.21 229	31.75 227

Gipuma		25.87 225	26.91 228	24.66 225	28.98 226	20.17 233	20.46 232	34.62 191	29.55 216	9.10 220	4.91 236	5.31 238	15.76 218	49.37 220	49.10 213	39.03 227	29.95 229
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
new_pmnet		23.19 227	28.17 227	17.37 228	17.03 234	24.92 231	19.66 233	16.16 233	27.05 220	4.42 232	20.77 224	19.20 231	12.19 226	37.71 229	36.38 228	34.77 231	31.17 228

PMMVS2		15.84 229	19.68 230	11.35 232	15.74 235	16.95 234	13.31 234	17.64 232	16.08 232	0.36 240	13.12 229	11.47 235	1.69 235	28.82 230	27.24 230	19.38 234	24.09 231

EMVS		14.49 231	12.45 233	16.87 231	27.02 230	12.56 237	8.13 235	27.19 219	15.05 234	3.14 235	6.69 234	2.67 240	15.08 220	14.60 234	18.05 232	20.67 233	17.56 234

E-PMN		15.09 230	13.19 232	17.30 229	27.80 228	12.62 236	7.81 236	27.54 215	14.62 235	3.19 234	6.89 233	2.52 241	15.09 219	15.93 232	20.22 231	22.38 232	19.53 232

MVE	12.28 19	13.53 232	15.72 231	10.96 233	7.39 236	15.71 235	6.05 237	23.73 224	10.29 237	3.01 237	5.77 235	3.41 239	11.91 228	20.11 231	29.79 229	13.67 235	24.98 230
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
DeepMVS_CX						6.95 238	5.98 238	2.25 234	11.73 236	2.07 239	11.85 230	5.43 237	11.75 229	11.40 235		8.10 237	18.38 233

tmp_tt				5.40 234	3.97 237	2.35 239	3.26 239	0.44 235	17.56 231	12.09 209	11.48 231	7.14 236	1.98 234	15.68 233	15.49 233	10.69 236

Patchmatch-RL test							1.04 240

testmvs		0.01 233	0.02 234	0.00 235	0.00 239	0.00 240	0.01 241	0.00 237	0.01 238	0.00 242	0.03 238	0.00 242	0.01 237	0.01 236	0.01 234	0.00 238	0.06 236

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

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

test123		0.01 233	0.02 234	0.00 235	0.00 239	0.00 240	0.00 242	0.00 237	0.01 238	0.00 242	0.04 237	0.00 242	0.01 237	0.00 237	0.01 234	0.00 238	0.07 235

MTAPA										65.14 1		80.20 14

MTMP										62.63 11		78.04 21

mPP-MVS					71.67 28							74.36 35

NP-MVS									72.00 35

High-resolution multi-view benchmark