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.

MCST-MVS		74.06 4	77.71 9	69.79 1	78.95 1	81.99 3	76.33 4	62.16 1	75.89 16	52.96 20	64.37 26	73.30 15	65.66 2	77.49 1	77.43 2	82.67 1	93.51 1

CNVR-MVS		73.87 5	78.60 6	68.35 4	73.32 8	81.97 4	76.19 5	59.29 3	80.12 9	56.70 7	67.09 20	76.48 6	64.26 3	75.88 3	75.75 4	80.32 7	92.93 2

ESAPD		76.46 1	83.32 1	68.47 2	78.83 2	83.07 2	77.86 1	58.75 4	86.89 3	56.64 8	89.08 2	83.11 1	60.69 12	74.28 7	74.11 8	78.06 30	92.67 3

DELS-MVS		67.36 19	70.34 33	63.89 17	69.12 26	81.55 6	70.82 20	55.02 17	53.38 65	48.83 33	56.45 40	59.35 50	60.05 18	74.93 5	74.78 6	79.51 13	91.95 4
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
CHOSEN 1792x2688		62.48 49	64.06 53	60.64 35	72.50 10	84.18 1	62.43 62	53.77 24	47.90 81	39.85 71	25.15 195	44.76 97	53.72 48	77.29 2	77.61 1	81.60 4	91.53 5

APDe-MVS		74.59 2	80.23 3	68.01 5	76.51 4	80.20 11	77.39 2	58.18 5	85.31 4	56.84 6	84.89 3	76.08 8	60.66 13	71.85 22	71.76 17	78.47 23	91.49 6

abl_6				63.79 18	70.80 18	81.22 7	65.26 52	53.25 26	77.02 13	53.02 19	65.14 25	73.74 13	60.30 15			80.13 8	90.27 7

NCCC		71.36 11	75.44 13	66.60 7	72.46 11	79.18 14	74.16 7	57.83 6	76.93 14	54.19 14	63.47 27	71.08 20	61.30 9	73.56 10	73.70 10	79.69 11	90.19 8

SMA-MVS		73.13 6	79.49 4	65.71 8	71.08 15	80.36 10	74.71 6	56.31 7	83.60 5	53.34 18	72.81 12	80.52 2	60.83 11	74.88 6	74.53 7	77.96 32	89.72 9

MVS_111021_HR		64.66 37	67.11 42	61.80 27	71.04 16	77.91 22	62.75 61	54.78 18	51.43 69	47.54 38	53.77 47	54.85 59	56.84 31	70.59 27	71.50 21	77.86 33	89.70 10

DeepPCF-MVS	62.48 1	70.07 14	78.36 7	60.39 38	62.38 55	76.96 30	65.54 49	52.23 29	87.46 2	49.07 30	74.05 11	76.19 7	59.01 22	72.79 15	71.61 19	74.13 108	89.49 11

PVSNet_BlendedMVS		62.53 46	66.37 45	58.05 50	58.17 75	75.70 36	61.30 66	48.67 52	58.67 52	50.93 24	55.43 41	49.39 75	53.01 59	69.46 36	66.55 59	76.24 52	89.39 12

PVSNet_Blended		62.53 46	66.37 45	58.05 50	58.17 75	75.70 36	61.30 66	48.67 52	58.67 52	50.93 24	55.43 41	49.39 75	53.01 59	69.46 36	66.55 59	76.24 52	89.39 12

CSCG		72.98 7	76.86 11	68.46 3	78.23 3	81.74 5	77.26 3	60.00 2	75.61 19	59.06 1	62.72 28	77.42 4	56.63 36	74.24 8	77.18 3	79.56 12	89.13 14

HPM-MVS++		72.44 8	78.73 5	65.11 9	71.88 14	77.31 26	71.98 15	55.67 11	83.11 7	53.59 16	75.90 8	78.49 3	61.00 10	73.99 9	73.31 12	76.55 47	88.97 15

APD-MVS		71.86 9	77.91 8	64.80 11	70.39 20	75.69 38	74.02 8	56.14 9	83.59 6	52.92 21	84.67 4	73.46 14	59.30 20	69.47 35	69.66 36	76.02 54	88.84 16
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
SteuartSystems-ACMMP		69.78 15	74.76 15	63.98 14	73.45 7	78.56 19	73.13 11	55.24 16	70.68 29	48.93 32	70.43 16	69.10 22	54.00 47	72.78 17	72.98 13	79.14 17	88.74 17
Skip Steuart: Steuart Systems R&D Blog.
train_agg		70.74 12	76.53 12	63.98 14	70.33 21	75.16 40	72.33 14	55.78 10	75.74 17	50.41 29	80.08 7	73.15 16	57.75 28	71.96 21	70.94 28	77.25 42	88.69 18

DI_MVS_plusplus_trai		61.86 52	65.26 48	57.90 54	57.93 78	74.51 43	66.30 43	46.49 83	49.96 73	41.62 68	42.69 74	61.77 42	58.74 23	70.25 31	69.32 38	76.31 50	88.30 19

CANet		67.21 23	71.83 24	61.83 26	64.51 42	79.25 13	66.72 41	48.73 50	68.49 37	50.63 28	61.40 31	66.47 26	61.44 7	69.31 38	69.90 33	78.94 20	88.00 20

DeepC-MVS	60.65 2	67.33 21	71.52 27	62.44 23	59.79 70	74.84 42	68.89 28	55.56 12	73.91 22	53.50 17	55.00 43	65.63 28	60.08 17	71.99 20	71.33 24	76.85 45	87.94 21
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
ACMMP_Plus		71.50 10	77.27 10	64.77 12	69.64 22	79.26 12	73.53 9	54.73 19	79.32 11	54.23 13	74.81 9	74.61 12	59.40 19	73.00 11	72.17 15	77.10 43	87.72 22

MP-MVS		67.34 20	73.08 16	60.64 35	66.20 34	76.62 32	69.22 27	50.92 36	70.07 30	48.81 34	69.66 17	70.12 21	53.68 50	68.41 46	69.13 41	74.98 73	87.53 23
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
MVS_0304		66.31 28	71.61 26	60.14 39	62.59 54	78.98 16	67.13 38	45.75 88	64.35 45	45.23 47	60.69 33	67.67 25	61.73 6	71.09 25	71.03 26	78.41 26	87.44 24

PVSNet_Blended_VisFu		58.56 64	62.33 61	54.16 69	56.90 80	73.92 47	57.72 96	46.16 86	44.23 87	42.73 59	46.26 65	51.06 69	46.28 115	67.99 51	65.38 68	75.18 68	87.44 24

CostFormer		62.45 50	65.68 47	58.67 46	63.29 46	77.65 23	67.62 34	38.42 166	54.04 63	46.00 43	48.27 62	57.89 54	56.97 30	67.03 58	67.79 52	79.74 9	87.09 26

HQP-MVS		67.22 22	72.08 22	61.56 30	66.76 32	73.58 51	71.41 17	52.98 27	69.92 32	43.85 52	70.58 15	58.75 52	56.76 33	72.90 13	71.88 16	77.57 36	86.94 27

CDPH-MVS		67.03 24	71.64 25	61.65 29	69.10 27	76.84 31	71.35 19	55.42 14	67.02 40	42.83 56	65.27 24	64.60 32	53.16 57	69.70 34	71.40 22	78.02 31	86.67 28

GG-mvs-BLEND		44.87 170	64.59 51	21.86 224	0.01 240	73.70 49	55.99 113	0.01 237	50.70 70	0.01 242	49.18 58	63.61 35	0.01 237	63.83 86	64.50 93	75.13 70	86.62 29

TSAR-MVS + GP.		66.77 27	72.21 21	60.44 37	61.23 61	70.00 70	64.26 56	47.79 60	72.98 24	56.32 9	71.35 14	72.33 17	55.68 39	65.49 64	66.66 58	77.35 38	86.62 29

PHI-MVS		65.17 35	72.07 23	57.11 60	63.02 49	77.35 25	67.04 39	48.14 59	68.03 38	37.56 79	66.00 22	65.39 29	53.19 56	70.68 26	70.57 32	73.72 114	86.46 31

zzz-MVS		67.78 18	72.46 20	62.33 25	66.09 35	74.21 44	70.05 23	51.54 32	77.27 12	54.61 11	60.30 35	71.51 19	56.73 34	69.19 39	68.63 47	74.96 74	86.11 32

DeepC-MVS_fast	60.18 3	66.84 26	70.69 31	62.36 24	62.76 50	73.21 54	67.96 31	52.31 28	72.26 26	51.03 23	56.50 39	64.26 33	63.37 4	71.64 23	70.85 29	76.70 46	86.10 33
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
PCF-MVS	55.99 6	62.31 51	66.60 44	57.32 58	59.12 74	73.68 50	67.53 35	48.71 51	61.35 49	42.83 56	51.33 53	63.48 36	53.48 55	65.64 62	64.87 72	72.22 135	85.83 34
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
HyFIR lowres test		57.12 70	59.11 72	54.80 68	61.55 60	77.55 24	59.02 91	45.00 95	41.84 101	33.93 101	22.44 202	49.16 78	51.02 68	68.39 47	68.71 45	78.26 28	85.70 35

MVS_Test		63.75 40	67.24 40	59.68 42	60.01 67	76.99 29	68.13 30	45.17 93	57.45 55	43.74 53	53.07 48	56.16 58	61.33 8	70.27 30	71.11 25	79.72 10	85.63 36

QAPM		65.47 33	67.82 38	62.72 22	72.56 9	81.17 8	67.43 36	55.38 15	56.07 61	43.29 54	43.60 73	65.38 30	59.10 21	72.20 19	70.76 30	78.56 22	85.59 37

IB-MVS	53.15 10	57.33 68	59.02 74	55.37 66	60.83 63	77.11 28	54.51 119	50.10 43	43.22 90	42.82 58	40.50 83	37.61 119	44.67 127	59.27 150	69.81 34	79.29 15	85.59 37
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
canonicalmvs		65.55 32	70.75 30	59.49 44	62.11 57	78.26 21	66.52 42	43.82 119	71.54 28	47.84 37	61.30 32	61.68 43	58.48 25	67.56 52	69.67 35	78.16 29	85.25 39

CP-MVS		64.37 39	69.48 35	58.39 48	62.21 56	71.81 62	67.27 37	49.51 45	69.40 35	45.76 44	60.41 34	64.96 31	51.84 65	67.33 56	67.57 53	73.78 113	84.89 40

HFP-MVS		68.75 16	72.84 17	63.98 14	68.87 28	75.09 41	71.87 16	51.22 34	73.50 23	58.17 4	68.05 19	68.67 23	57.79 27	70.49 29	69.23 39	75.98 55	84.84 41

tpmrst		57.23 69	59.08 73	55.06 67	59.91 68	70.65 67	60.71 69	35.38 187	47.91 80	42.58 62	39.78 87	45.45 95	54.44 45	62.19 111	62.82 110	77.37 37	84.73 42

test-LLR		54.62 80	58.66 77	49.89 116	51.68 136	65.89 106	47.88 153	46.35 84	42.51 93	29.84 128	41.41 79	48.87 79	45.20 120	62.91 106	64.43 95	78.43 24	84.62 43

TESTMET0.1,1		53.30 103	58.66 77	47.04 133	44.94 175	65.89 106	47.88 153	35.95 181	42.51 93	29.84 128	41.41 79	48.87 79	45.20 120	62.91 106	64.43 95	78.43 24	84.62 43

diffmvs		61.00 57	65.21 49	56.08 65	55.72 96	74.18 45	64.62 53	43.25 125	56.71 56	42.71 60	50.01 57	62.20 41	54.45 44	64.56 73	64.30 97	75.75 60	84.53 45

CPTT-MVS		59.54 63	64.47 52	53.79 72	54.99 101	67.63 94	65.48 50	44.59 109	64.81 44	37.74 77	51.55 51	59.90 49	49.77 79	61.83 116	61.26 135	70.18 154	84.31 46

DWT-MVSNet_training		61.22 56	63.52 56	58.53 47	65.00 40	76.55 33	59.50 88	48.22 56	51.79 68	42.14 64	47.85 63	50.21 71	55.46 40	66.16 60	67.92 51	80.85 5	84.14 47

ACMMPR		66.20 29	71.51 28	60.00 40	65.34 39	74.04 46	69.39 26	50.92 36	71.97 27	46.04 42	66.79 21	65.68 27	53.07 58	68.93 43	69.12 42	75.21 67	84.05 48

Effi-MVS+		59.63 62	61.78 67	57.12 59	61.56 59	71.63 63	63.61 57	47.59 64	47.18 82	37.79 76	45.29 69	49.93 73	56.27 37	67.45 53	67.06 55	75.91 57	83.93 49

EPNet		64.39 38	70.93 29	56.77 62	60.58 66	75.77 35	59.28 90	50.58 39	69.93 31	40.73 70	68.59 18	61.60 45	53.72 48	68.65 44	68.07 49	75.75 60	83.87 50
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
Anonymous20240521		60.61 58	61.52 68	59.54 43	68.33 29	77.12 27	65.45 51	54.67 20	48.63 78	35.89 85	36.56 107	43.34 100	53.61 53	69.82 33	71.55 20	79.29 15	83.63 51

tpm		54.94 77	57.86 87	51.54 97	59.48 72	67.04 97	58.34 94	34.60 191	41.93 100	34.41 94	42.40 75	47.14 88	49.07 91	61.46 120	61.67 128	73.31 123	83.39 52

ACMMP		63.27 42	67.85 37	57.93 52	62.64 53	72.30 60	68.23 29	48.77 49	66.50 41	43.05 55	62.07 29	57.84 55	49.98 74	66.58 59	66.46 62	74.93 78	83.17 53
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
X-MVS		63.53 41	68.62 36	57.60 55	64.77 41	73.06 55	65.82 46	50.53 40	65.77 42	42.02 65	58.20 37	63.42 37	47.83 107	68.25 50	68.50 48	74.61 90	83.16 54

PGM-MVS		65.35 34	70.43 32	59.43 45	65.78 37	73.75 48	69.41 25	48.18 57	68.80 36	45.37 46	65.88 23	64.04 34	52.68 63	68.94 42	68.68 46	75.18 68	82.93 55

v7		53.99 88	55.20 102	52.57 83	52.08 121	69.46 79	59.68 84	44.92 97	35.90 133	34.33 95	33.99 138	35.55 142	50.08 73	64.38 77	64.67 80	74.32 99	82.47 56

v1144		53.47 98	54.65 111	52.10 90	51.93 123	69.81 72	59.32 89	44.77 106	33.21 160	32.52 114	33.55 145	34.34 154	49.29 89	64.58 72	64.81 75	74.74 83	82.27 57

OPM-MVS		61.59 54	62.30 62	60.76 34	66.53 33	73.35 52	71.41 17	54.18 22	40.82 103	41.57 69	45.70 68	54.84 60	54.43 46	69.92 32	69.19 40	76.45 48	82.25 58

test-mter		48.31 149	55.04 105	40.45 176	34.12 216	59.02 173	41.77 181	28.05 218	38.43 113	22.67 161	39.35 95	44.40 98	41.88 136	60.30 137	61.68 127	74.20 104	82.12 59

MVSTER		62.51 48	67.22 41	57.02 61	55.05 99	69.23 84	63.02 59	46.88 78	61.11 50	43.95 51	59.20 36	58.86 51	56.80 32	69.13 40	70.98 27	76.41 49	82.04 60

v144192		52.43 113	53.63 120	51.03 101	51.06 141	69.60 77	56.94 102	44.84 101	32.15 165	30.88 123	32.45 150	32.71 162	48.36 101	62.98 105	63.52 104	74.10 111	82.02 61

v1neww		54.32 82	55.60 99	52.81 75	52.11 118	69.43 81	60.57 73	44.86 98	37.13 127	35.34 89	34.95 121	37.46 125	49.53 84	63.69 91	64.59 84	74.47 94	81.99 62

v7new		54.32 82	55.60 99	52.81 75	52.11 118	69.43 81	60.57 73	44.86 98	37.13 127	35.34 89	34.95 121	37.46 125	49.53 84	63.69 91	64.59 84	74.47 94	81.99 62

v6		54.32 82	55.61 98	52.82 74	52.11 118	69.44 80	60.57 73	44.86 98	37.15 126	35.40 87	34.96 119	37.47 124	49.52 86	63.68 94	64.59 84	74.47 94	81.99 62

v2v482		54.00 86	55.12 103	52.69 81	51.73 134	69.42 83	60.65 71	45.09 94	34.56 144	33.73 105	35.29 115	35.36 144	49.92 75	64.05 82	65.16 69	75.00 72	81.98 65

TSAR-MVS + MP.		70.28 13	75.09 14	64.66 13	69.34 24	64.61 119	72.60 12	56.29 8	80.73 8	58.36 3	84.56 5	75.22 10	55.37 41	69.11 41	69.45 37	75.97 56	81.97 66

ACMP	56.21 5	59.78 59	61.81 66	57.41 57	61.15 62	68.88 86	65.98 45	48.85 48	58.56 54	44.19 50	48.89 59	46.31 91	48.56 99	63.61 98	64.49 94	75.75 60	81.91 67
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
divwei89l23v2f112		53.78 90	55.02 106	52.33 86	51.89 126	69.74 76	60.25 77	44.82 102	34.06 150	33.33 110	34.56 131	36.18 137	48.75 95	63.90 83	64.82 73	74.94 75	81.88 68

v1		53.78 90	55.02 106	52.34 85	51.89 126	69.75 74	60.24 78	44.81 103	34.05 151	33.39 108	34.57 128	36.18 137	48.75 95	63.90 83	64.82 73	74.94 75	81.88 68

v1141		53.78 90	55.02 106	52.33 86	51.89 126	69.75 74	60.24 78	44.81 103	34.04 152	33.34 109	34.57 128	36.17 139	48.76 94	63.90 83	64.81 75	74.94 75	81.87 70

v1192		52.69 107	53.86 118	51.31 99	51.22 140	69.76 73	57.37 99	44.39 110	32.21 164	31.39 121	32.41 151	32.44 166	49.19 90	64.25 78	64.17 98	74.31 102	81.81 71

tpm cat1		57.41 67	58.26 82	56.42 63	60.80 64	72.56 59	64.35 55	38.43 165	49.18 77	46.36 40	36.69 106	43.50 99	54.47 43	61.39 122	62.64 114	74.11 110	81.81 71

SD-MVS		68.30 17	72.58 19	63.31 20	69.24 25	67.85 91	70.81 21	53.65 25	79.64 10	58.52 2	74.31 10	75.37 9	53.52 54	65.63 63	63.56 103	74.13 108	81.73 73

TSAR-MVS + ACMM		65.95 31	72.83 18	57.93 52	69.35 23	65.85 108	73.36 10	39.84 154	76.00 15	48.69 35	82.54 6	75.03 11	49.38 88	65.33 66	63.42 105	66.94 177	81.67 74

v1921920		51.95 119	53.19 123	50.51 107	50.82 146	69.14 85	55.45 114	44.34 114	31.53 170	30.53 125	31.96 154	31.67 173	48.31 103	63.12 103	63.28 106	73.59 115	81.60 75

MVS_111021_LR		57.06 71	60.60 70	52.93 73	56.25 86	65.14 114	55.16 117	41.21 147	52.32 67	44.89 48	53.92 46	49.27 77	52.16 64	61.46 120	60.54 144	67.92 166	81.53 76

MAR-MVS		66.85 25	69.81 34	63.39 19	73.56 6	80.51 9	69.87 24	51.51 33	67.78 39	46.44 39	51.09 54	61.60 45	60.38 14	72.67 18	73.61 11	78.59 21	81.44 77
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
EPP-MVSNet		52.91 105	58.91 75	45.91 139	54.99 101	68.84 87	49.27 145	42.71 134	37.53 119	20.20 172	46.09 66	56.19 57	36.90 149	61.37 123	60.90 139	71.41 144	81.41 78

V42		52.63 108	55.08 104	49.76 118	44.93 176	67.49 96	60.19 80	42.13 140	37.21 124	34.08 99	34.57 128	37.30 127	47.29 110	63.48 101	64.15 99	69.96 157	81.38 79

tpmp4_e23		59.70 60	61.03 69	58.14 49	63.70 43	73.33 53	65.69 47	39.53 156	52.56 66	46.23 41	41.59 78	47.46 86	57.38 29	65.01 68	65.89 64	76.31 50	81.36 80

3Dnovator	58.39 4	65.97 30	66.85 43	64.94 10	73.72 5	79.03 15	67.73 33	54.25 21	61.52 48	52.79 22	42.27 76	60.73 48	62.01 5	71.29 24	71.75 18	79.12 18	81.34 81

CANet_DTU		57.87 66	63.63 55	51.15 100	52.18 116	70.20 69	58.14 95	37.32 172	56.49 57	31.06 122	57.38 38	50.05 72	53.67 51	64.98 69	65.04 70	74.57 91	81.29 82

LGP-MVS_train		59.69 61	62.59 59	56.31 64	61.94 58	68.15 90	66.90 40	48.15 58	59.75 51	38.47 74	50.38 56	48.34 83	46.87 112	65.39 65	64.93 71	75.51 65	81.21 83

gm-plane-assit		45.41 166	48.03 169	42.34 167	56.49 83	40.48 218	24.54 223	34.15 200	14.44 227	6.59 217	17.82 212	35.32 145	49.82 77	72.93 12	74.11 8	82.47 2	81.12 84

v1240		51.42 124	52.66 136	49.97 114	50.31 149	68.70 88	54.05 123	43.85 118	30.78 175	30.22 126	31.43 158	31.03 182	47.98 105	62.62 110	63.16 108	73.40 120	80.93 85

UGNet		51.04 128	58.79 76	42.00 169	40.59 192	65.32 113	46.65 164	39.26 160	39.90 105	27.30 136	54.12 45	52.03 64	30.93 173	59.85 146	59.62 152	67.23 173	80.70 86
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
gg-mvs-nofinetune		50.82 132	55.83 95	44.97 147	60.63 65	75.69 38	53.40 126	34.48 193	20.05 218	6.93 216	18.27 211	52.70 63	33.57 160	70.50 28	72.93 14	80.84 6	80.68 87

OpenMVS	55.62 8	62.57 45	63.76 54	61.19 32	72.13 12	78.84 17	64.42 54	50.51 41	56.44 58	45.67 45	36.88 104	56.51 56	56.66 35	68.28 49	68.96 43	77.73 35	80.44 88

3Dnovator+	55.76 7	62.70 44	65.10 50	59.90 41	65.89 36	72.15 61	62.94 60	49.82 44	62.77 47	49.06 31	43.62 72	61.47 47	58.60 24	68.51 45	66.75 57	73.08 128	80.40 89

Fast-Effi-MVS+-dtu		52.47 112	55.89 94	48.48 128	56.25 86	65.07 115	58.75 93	23.79 226	41.27 102	27.07 139	37.95 98	41.34 111	50.85 70	62.90 108	62.34 120	74.17 106	80.37 90

PMMVS		55.74 72	62.68 57	47.64 131	44.34 181	65.58 112	47.22 159	37.96 168	56.43 59	34.11 98	61.51 30	47.41 87	54.55 42	65.88 61	62.49 118	67.67 169	79.48 91

HSP-MVS		74.54 3	81.12 2	66.86 6	71.93 13	78.65 18	72.60 12	55.44 13	89.94 1	54.35 12	92.24 1	77.08 5	69.84 1	75.48 4	75.01 5	76.99 44	79.45 92

Fast-Effi-MVS+		55.73 73	58.26 82	52.76 78	54.33 104	68.19 89	57.05 100	34.66 189	46.92 83	38.96 73	40.53 82	41.55 108	55.69 38	65.31 67	65.99 63	75.90 58	79.34 93

v10		53.44 99	54.40 112	52.31 88	52.08 121	66.99 98	59.68 84	43.41 122	35.90 133	34.30 96	33.98 139	35.56 141	50.10 72	64.39 76	64.67 80	74.32 99	79.30 94

v148		51.72 120	53.15 125	50.05 112	50.15 150	67.51 95	56.98 101	42.85 131	32.60 163	32.41 116	33.88 140	34.71 151	44.45 128	61.06 126	63.00 109	73.45 119	79.24 95

v8		53.77 93	54.82 110	52.54 84	52.12 117	66.95 100	60.56 76	43.23 126	37.17 125	35.35 88	34.96 119	37.50 123	49.51 87	63.67 96	64.59 84	74.48 93	78.91 96

GA-MVS		53.77 93	56.41 92	50.70 104	51.63 138	69.96 71	57.55 98	44.39 110	34.31 146	27.15 137	40.99 81	36.40 131	47.65 109	67.45 53	67.16 54	75.83 59	78.60 97

thres100view900		52.33 115	53.91 117	50.48 108	56.10 88	67.79 92	56.18 112	49.18 46	35.86 136	25.22 145	34.74 125	34.10 155	42.41 135	64.45 75	62.62 115	73.81 112	77.85 98

thres200		50.76 133	52.52 137	48.70 127	55.98 92	64.60 120	55.29 116	47.34 70	33.91 154	24.36 154	34.33 134	33.90 159	37.27 147	60.84 131	62.41 119	71.99 138	77.63 99

IS_MVSNet		51.53 123	57.98 85	44.01 154	55.96 93	66.16 104	47.65 155	42.84 132	39.82 106	19.09 179	44.97 70	50.28 70	27.20 188	63.43 102	63.84 100	71.33 145	77.33 100

Vis-MVSNet		51.13 127	58.04 84	43.06 163	47.68 163	67.71 93	49.10 148	39.09 162	37.75 117	22.57 162	51.03 55	48.78 81	32.42 168	62.12 113	61.80 124	67.49 171	77.12 101
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
v18		53.63 95	54.35 113	52.80 77	52.25 113	62.94 139	60.80 68	42.78 133	39.23 108	36.81 81	35.07 118	37.78 118	49.82 77	63.69 91	64.65 82	74.32 99	77.07 102

Effi-MVS+-dtu		53.63 95	54.85 109	52.20 89	59.32 73	61.33 159	56.42 110	40.24 152	43.84 89	34.22 97	39.49 92	46.18 92	53.00 61	58.72 155	57.49 163	69.99 156	76.91 103

v16		53.50 97	54.17 114	52.73 80	52.24 114	62.90 140	60.67 70	42.67 136	38.72 111	36.70 82	34.84 124	37.59 121	49.69 81	63.72 88	64.68 78	74.39 97	76.72 104

v17		53.40 100	54.09 115	52.60 82	52.22 115	62.90 140	60.64 72	42.66 137	38.24 114	36.04 84	34.85 123	37.59 121	49.63 82	63.70 90	64.68 78	74.39 97	76.70 105

AdaColmap		62.79 43	62.63 58	62.98 21	70.82 17	72.90 58	67.84 32	54.09 23	65.14 43	50.71 26	41.78 77	47.64 85	60.17 16	67.41 55	66.83 56	74.28 103	76.69 106

anonymousdsp		43.03 183	47.19 171	38.18 188	36.00 212	56.92 184	38.44 188	34.56 192	24.22 207	22.53 163	29.69 177	29.92 187	35.21 154	53.96 183	58.98 155	62.32 197	76.66 107

IterMVS-LS		53.36 102	55.65 96	50.68 106	55.34 97	59.04 172	55.00 118	39.98 153	38.72 111	33.22 112	44.52 71	47.05 89	49.63 82	61.82 117	61.77 125	70.92 148	76.61 108
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
MSLP-MVS++		61.81 53	62.19 63	61.37 31	68.33 29	63.08 137	70.75 22	38.89 163	63.96 46	57.51 5	48.59 60	61.66 44	53.67 51	62.04 114	59.92 150	79.03 19	76.08 109

conf0.002		51.63 122	53.79 119	49.12 122	56.33 84	64.84 117	53.05 130	47.38 68	35.86 136	24.83 147	37.86 99	38.15 116	41.08 137	61.04 128	62.70 111	72.05 136	76.06 110

conf0.01		51.32 125	53.22 122	49.11 123	56.29 85	64.78 118	53.05 130	47.37 69	35.86 136	24.83 147	36.85 105	36.64 130	41.08 137	61.01 129	61.37 130	72.03 137	76.01 111

v15		52.60 109	53.17 124	51.94 91	51.92 124	62.76 142	60.04 81	42.19 139	34.35 145	33.96 100	34.37 133	36.40 131	48.73 97	63.75 87	64.59 84	74.79 79	75.97 112

ACMM	53.73 9	57.91 65	58.27 81	57.49 56	63.10 47	66.45 102	65.65 48	49.02 47	53.69 64	42.67 61	36.41 109	46.07 93	50.38 71	64.74 71	64.63 83	74.14 107	75.91 113
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
thres400		50.39 134	52.22 141	48.26 129	55.02 100	66.32 103	52.97 134	48.33 55	32.68 162	22.94 160	33.21 147	33.38 161	37.27 147	62.74 109	61.38 129	73.04 129	75.81 114

FC-MVSNet-train		55.68 74	57.00 90	54.13 70	63.37 44	66.16 104	46.77 162	52.14 30	42.36 96	37.67 78	48.50 61	41.42 110	51.28 67	61.58 119	63.22 107	73.56 116	75.76 115

v11		52.23 117	52.83 131	51.53 98	51.73 134	62.49 148	58.82 92	41.81 145	33.53 158	33.23 111	33.73 142	35.10 147	49.07 91	64.49 74	64.71 77	74.49 92	75.75 116

tfpn111		51.00 129	52.68 135	49.04 125	56.10 88	64.52 124	53.05 130	47.31 72	35.86 136	24.79 150	36.35 110	34.10 155	41.08 137	60.84 131	61.37 130	71.90 140	75.70 117

conf200view11		50.87 131	52.45 139	49.04 125	56.10 88	64.52 124	53.05 130	47.31 72	35.86 136	24.79 150	34.74 125	34.10 155	41.08 137	60.84 131	61.37 130	71.90 140	75.70 117

tfpn200view9		50.91 130	52.45 139	49.11 123	56.10 88	64.53 122	53.06 129	47.31 72	35.86 136	25.22 145	34.74 125	34.10 155	41.08 137	60.84 131	61.37 130	71.90 140	75.70 117

V14		52.51 111	53.06 126	51.87 92	51.91 125	62.70 144	59.93 82	42.13 140	34.16 148	33.88 102	34.22 135	36.35 134	48.62 98	63.72 88	64.59 84	74.77 80	75.69 120

V9		52.42 114	52.94 129	51.81 93	51.89 126	62.64 146	59.81 83	42.08 142	33.97 153	33.85 103	34.05 137	36.28 135	48.49 100	63.68 94	64.59 84	74.76 81	75.41 121

IterMVS		50.23 135	53.27 121	46.68 135	47.59 165	60.58 164	53.10 128	36.62 175	36.07 132	25.89 142	39.42 93	40.05 114	43.65 130	60.22 141	61.35 134	73.23 124	75.23 122
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
MS-PatchMatch		61.41 55	61.88 64	60.85 33	70.57 19	75.98 34	66.29 44	46.91 77	50.56 71	48.28 36	36.30 112	51.64 65	50.95 69	72.89 14	70.65 31	82.13 3	75.17 123

v12		52.31 116	52.81 132	51.72 95	51.87 130	62.56 147	59.66 86	42.02 143	33.77 155	33.70 106	33.88 140	36.20 136	48.36 101	63.64 97	64.56 91	74.73 86	75.08 124

v13		52.22 118	52.69 134	51.67 96	51.85 132	62.48 149	59.57 87	41.97 144	33.61 157	33.69 107	33.71 143	36.11 140	48.23 104	63.57 99	64.55 92	74.72 87	74.77 125

view600		48.85 145	50.58 154	46.82 134	54.19 105	64.94 116	50.81 140	47.53 67	31.78 168	21.59 166	32.31 152	32.63 164	34.28 157	61.06 126	57.41 164	72.54 132	73.96 126

FMVSNet3		55.66 75	59.68 71	50.96 102	50.59 147	66.49 101	57.57 97	46.61 79	49.30 74	28.77 131	39.61 88	51.42 66	43.85 129	68.29 48	68.80 44	78.35 27	73.86 127

v7n		47.22 156	48.38 167	45.87 140	48.20 162	63.58 132	50.69 141	40.93 150	26.60 197	26.44 141	26.52 187	29.65 189	38.19 145	58.22 156	60.23 147	70.79 149	73.83 128

OMC-MVS		55.48 76	61.85 65	48.04 130	41.55 190	60.32 166	56.80 104	31.78 212	75.67 18	42.30 63	51.52 52	54.15 61	49.91 76	60.28 139	57.59 161	65.91 180	73.42 129

thres600view7		48.44 148	50.23 155	46.35 137	54.05 107	64.60 120	50.18 143	47.34 70	31.73 169	20.74 169	32.28 153	32.62 165	33.79 159	60.84 131	56.11 174	71.99 138	73.40 130

CHOSEN 280x420		42.39 188	47.40 170	36.54 193	33.56 217	39.66 221	40.67 183	26.88 221	34.66 143	18.03 186	30.09 172	45.59 94	44.82 126	54.46 178	54.00 186	55.28 214	73.32 131

CDS-MVSNet		49.25 142	53.97 116	43.75 156	47.53 166	64.53 122	48.59 149	42.27 138	33.77 155	26.64 140	40.46 84	42.26 105	30.01 176	61.77 118	61.71 126	67.48 172	73.28 132
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
GBi-Net		54.66 78	58.42 79	50.26 110	49.36 155	65.81 109	56.80 104	46.61 79	49.30 74	28.77 131	39.61 88	51.42 66	42.71 132	64.25 78	65.54 65	77.32 39	73.03 133

test1		54.66 78	58.42 79	50.26 110	49.36 155	65.81 109	56.80 104	46.61 79	49.30 74	28.77 131	39.61 88	51.42 66	42.71 132	64.25 78	65.54 65	77.32 39	73.03 133

FMVSNet2		53.94 89	57.29 88	50.03 113	49.36 155	65.81 109	56.80 104	45.95 87	43.13 91	28.04 135	35.68 113	48.18 84	42.71 132	67.23 57	67.95 50	77.32 39	73.03 133

pmmvs4		51.28 126	52.50 138	49.85 117	49.54 154	63.02 138	52.83 135	43.41 122	44.65 86	35.71 86	34.38 132	32.25 167	45.14 123	60.21 142	60.03 148	72.44 134	72.98 136

UniMVSNet (Re)		46.89 160	51.65 146	41.34 174	45.60 171	62.71 143	44.05 175	47.10 76	37.24 123	13.55 195	36.90 103	34.54 153	26.76 189	57.56 160	59.90 151	70.98 147	72.69 137

UniMVSNet_NR-MVSNet		49.56 141	53.04 127	45.49 142	51.59 139	64.42 128	46.97 160	51.01 35	37.87 115	16.42 189	39.87 86	34.91 149	33.43 164	59.59 147	62.70 111	73.52 117	71.94 138

DU-MVS		47.33 155	50.86 151	43.20 161	44.43 179	60.64 162	46.97 160	47.63 62	37.26 121	16.42 189	37.31 101	31.39 176	33.43 164	57.53 161	59.98 149	70.35 152	71.94 138

pmmvs5		47.02 159	50.02 156	43.51 158	43.48 186	62.65 145	47.24 158	37.78 170	30.59 177	24.80 149	35.26 116	30.43 186	34.36 156	59.05 151	60.28 146	73.40 120	71.92 140

view800		47.68 153	49.78 157	45.24 145	53.39 109	63.19 136	48.13 152	46.57 82	30.98 174	20.25 171	31.52 157	31.90 171	31.52 170	59.37 148	59.61 153	71.56 143	71.89 141

NR-MVSNet		48.84 146	51.76 143	45.44 143	57.66 79	60.64 162	47.39 156	47.63 62	37.26 121	13.31 196	37.31 101	29.64 190	33.53 161	63.52 100	62.09 123	73.10 127	71.89 141

TSAR-MVS + COLMAP		54.37 81	62.43 60	44.98 146	34.33 215	58.94 174	54.11 122	34.15 200	74.06 21	34.57 93	71.63 13	42.03 107	47.88 106	61.26 124	57.33 166	64.83 185	71.74 143

CNLPA		54.00 86	57.08 89	50.40 109	49.83 152	61.75 154	53.47 125	37.27 173	74.55 20	44.85 49	33.58 144	45.42 96	52.94 62	58.89 152	53.66 187	64.06 187	71.68 144

tfpnnormal		46.61 162	46.82 175	46.37 136	52.70 111	62.31 150	50.39 142	47.17 75	25.74 202	21.80 165	23.13 200	24.15 210	33.45 163	60.28 139	60.77 142	72.70 131	71.39 145

EG-PatchMatch MVS		50.23 135	50.89 150	49.47 119	59.54 71	70.88 64	52.46 136	44.01 117	26.22 199	31.91 117	24.97 196	31.45 175	33.48 162	64.79 70	66.51 61	75.40 66	71.39 145

v52		44.41 173	46.76 176	41.67 170	38.93 196	60.06 167	49.26 146	36.02 178	25.57 203	24.73 153	25.66 192	31.34 177	35.93 151	55.52 173	57.99 159	65.14 182	71.21 147

V4		44.41 173	46.76 176	41.67 170	38.93 196	60.05 168	49.26 146	36.02 178	25.55 204	24.75 152	25.66 192	31.33 178	35.93 151	55.52 173	57.99 159	65.14 182	71.21 147

v748		44.90 169	46.14 180	43.46 159	45.37 174	60.89 161	48.15 151	39.42 157	25.81 201	24.36 154	25.90 191	28.48 194	34.44 155	53.39 187	57.35 165	69.00 161	71.14 149

PatchmatchNet		53.37 101	55.62 97	50.75 103	55.93 94	70.54 68	51.39 139	36.41 176	44.85 85	37.26 80	39.40 94	42.54 103	47.83 107	60.29 138	60.88 141	75.69 63	70.87 150
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
MDTV_nov1_ep13		52.99 104	55.59 101	49.95 115	54.08 106	70.69 66	56.47 109	38.42 166	42.78 92	30.19 127	39.56 91	43.31 101	45.78 117	60.07 143	62.11 122	74.74 83	70.62 151

Baseline_NR-MVSNet		47.14 158	50.83 152	42.84 165	44.43 179	63.31 135	44.50 174	50.36 42	37.71 118	11.25 203	30.84 165	32.09 168	30.96 172	57.53 161	63.73 102	75.53 64	70.60 152

TranMVSNet+NR-MVSNet		48.06 151	51.36 148	44.21 152	50.38 148	62.09 152	47.28 157	50.88 38	36.11 131	13.25 197	37.51 100	31.60 174	30.70 174	59.34 149	62.53 117	72.81 130	70.31 153

pm-mvs1		46.14 164	49.34 163	42.41 166	48.93 158	62.22 151	44.98 172	42.68 135	27.66 190	20.76 168	29.88 175	34.96 148	26.41 190	60.03 144	60.42 145	70.70 150	70.20 154

FMVSNet1		50.14 137	52.78 133	47.06 132	45.56 172	63.56 133	54.22 121	43.74 120	34.10 149	25.37 144	29.79 176	42.06 106	38.70 143	64.25 78	65.54 65	74.75 82	70.18 155

EPMVS		54.07 85	56.06 93	51.75 94	56.74 82	70.80 65	55.32 115	34.20 197	46.46 84	36.59 83	40.38 85	42.55 102	49.77 79	61.25 125	60.90 139	77.86 33	70.08 156

ACMH+	47.85 12	49.13 144	48.86 166	49.44 120	56.75 81	62.01 153	56.62 108	47.55 66	37.49 120	23.98 156	26.68 186	29.46 191	43.12 131	57.45 163	58.85 156	68.62 164	70.05 157

ACMH	47.82 13	50.10 138	49.60 159	50.69 105	63.36 45	66.99 98	56.83 103	52.13 31	31.06 173	17.74 187	28.22 182	26.24 201	45.17 122	60.88 130	63.80 101	68.91 162	70.00 158
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
tfpn		46.62 161	49.07 164	43.75 156	52.70 111	61.49 158	45.65 169	45.68 89	30.25 179	18.84 182	30.87 164	33.67 160	29.22 183	57.80 159	59.49 154	70.44 151	69.95 159

CLD-MVS		64.69 36	67.25 39	61.69 28	68.22 31	78.33 20	63.09 58	47.59 64	69.64 33	53.98 15	54.87 44	53.94 62	57.87 26	72.79 15	71.34 23	79.40 14	69.87 160
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
CR-MVSNet		48.82 147	51.85 142	45.29 144	46.74 168	55.95 186	52.06 137	34.21 195	42.17 97	31.74 118	32.92 149	42.53 104	45.00 124	58.80 153	61.11 137	61.99 198	69.47 161

PatchT		48.11 150	51.27 149	44.43 149	50.13 151	61.58 155	33.59 200	32.92 206	40.38 104	31.74 118	30.60 168	36.93 128	45.00 124	58.80 153	61.11 137	73.19 125	69.47 161

RPMNet		43.70 179	48.17 168	38.48 186	45.52 173	55.95 186	37.66 192	26.63 222	42.17 97	25.47 143	29.59 178	37.61 119	33.87 158	50.85 194	52.02 194	61.75 200	69.00 163

dps		52.84 106	52.92 130	52.74 79	59.89 69	69.49 78	54.47 120	37.38 171	42.49 95	39.53 72	35.33 114	32.71 162	51.83 66	60.45 136	61.12 136	73.33 122	68.86 164

EPNet_dtu		49.85 139	56.99 91	41.52 173	52.79 110	57.06 183	41.44 182	43.13 127	56.13 60	19.24 178	52.11 49	48.38 82	22.14 198	58.19 157	58.38 158	70.35 152	68.71 165
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
conf0.05thres1000		45.26 168	46.99 173	43.24 160	51.87 130	60.52 165	45.17 171	45.24 92	27.06 194	18.60 183	26.24 188	31.23 181	28.82 185	56.88 167	58.52 157	69.71 158	68.50 166

TAPA-MVS	47.92 11	51.66 121	57.88 86	44.40 150	36.46 209	58.42 181	53.82 124	30.83 214	69.51 34	34.97 92	46.90 64	49.67 74	46.99 111	58.00 158	54.64 184	63.33 193	68.00 167
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
TAMVS		44.27 177	49.35 162	38.35 187	44.74 177	61.04 160	39.07 186	31.82 211	29.95 181	18.34 184	33.55 145	39.94 115	30.01 176	56.85 168	57.58 162	66.13 179	66.54 168

pmmvs6		41.90 189	44.01 190	39.43 182	44.45 178	58.77 175	41.92 180	39.22 161	21.74 210	19.08 180	17.40 215	31.33 178	24.28 194	55.94 171	56.67 171	67.60 170	66.24 169

pmmvs-eth3d		44.67 171	45.27 184	43.98 155	42.56 189	55.72 189	44.97 173	40.81 151	31.96 167	29.13 130	26.09 190	25.27 207	36.69 150	55.13 177	56.62 172	69.68 159	66.12 170

TransMVSNet (Re)		47.46 154	48.94 165	45.74 141	57.96 77	64.29 130	48.26 150	48.47 54	26.33 198	19.33 176	29.45 179	31.28 180	25.31 192	63.05 104	62.70 111	75.10 71	65.47 171

Anonymous20231206		40.63 191	43.29 192	37.53 191	48.88 160	55.81 188	34.99 195	44.98 96	28.16 187	10.16 207	17.26 216	27.50 197	18.28 203	54.00 182	55.07 181	67.85 168	65.23 172

tfpn_ndepth		46.53 163	49.41 161	43.18 162	54.66 103	61.56 156	42.25 179	45.66 90	35.68 142	18.31 185	36.55 108	34.84 150	28.88 184	55.45 175	57.01 169	69.32 160	64.78 173

MSDG		52.58 110	51.40 147	53.95 71	65.48 38	64.31 129	61.44 65	44.02 116	44.17 88	32.92 113	30.40 169	31.81 172	46.35 114	62.13 112	62.55 116	73.49 118	64.41 174

PLC	44.22 14	49.14 143	51.75 144	46.10 138	42.78 188	55.60 190	53.11 127	34.46 194	55.69 62	32.47 115	34.16 136	41.45 109	48.91 93	57.13 165	54.09 185	64.84 184	64.10 175
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
MIMVSNet		45.62 165	49.56 160	41.02 175	38.17 199	64.43 127	49.48 144	35.43 186	36.53 129	20.06 174	22.58 201	35.16 146	28.75 186	61.97 115	62.20 121	74.20 104	64.07 176

MDTV_nov1_ep13_2view		44.44 172	45.75 181	42.91 164	46.13 169	63.43 134	46.53 165	34.20 197	29.08 186	19.95 175	26.23 189	27.89 196	35.88 153	53.36 188	56.43 173	74.74 83	63.86 177

Vis-MVSNet (Re-imp)		44.31 176	51.67 145	35.72 195	51.82 133	55.24 191	34.57 199	41.63 146	39.10 109	8.84 211	45.93 67	46.63 90	14.45 213	54.09 181	57.03 168	63.00 194	63.65 178

thresconf0.02		47.89 152	50.76 153	44.54 148	53.86 108	63.96 131	46.23 166	47.72 61	33.00 161	17.08 188	36.35 110	37.80 117	29.86 178	60.01 145	60.57 143	72.49 133	63.62 179

CVMVSNet		38.91 192	44.49 189	32.40 204	34.57 213	47.20 207	34.81 197	34.20 197	31.45 171	8.95 210	38.86 96	36.38 133	24.30 193	47.77 200	46.94 213	57.59 209	62.85 180

CMPMVS	33.64 16	44.39 175	46.41 179	42.03 168	44.21 183	56.50 185	46.73 163	26.48 223	34.20 147	35.14 91	24.22 197	34.64 152	40.52 142	56.50 170	56.07 175	59.12 205	62.74 181
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
MVS-HIRNet		43.98 178	43.63 191	44.39 151	47.66 164	59.31 171	32.66 206	33.88 202	30.15 180	33.75 104	16.82 218	28.39 195	45.25 119	53.92 186	55.00 183	73.16 126	61.80 182

USDC		42.80 184	45.57 182	39.58 179	34.55 214	51.13 195	42.61 178	36.21 177	39.59 107	23.65 158	33.13 148	20.87 216	37.86 146	55.35 176	57.16 167	62.61 195	61.75 183

PM-MVS		34.96 206	38.17 203	31.22 209	22.78 227	40.82 217	33.56 201	23.61 227	29.16 185	21.43 167	28.00 183	21.43 212	31.90 169	44.33 210	42.12 220	54.07 216	61.34 184

ADS-MVSNet		45.39 167	46.42 178	44.19 153	48.74 161	57.52 182	43.91 176	31.93 210	35.89 135	27.11 138	30.12 171	32.06 169	45.30 118	53.13 189	55.19 180	68.15 165	61.07 185

test0.0.03 1		43.07 182	46.95 174	38.54 185	51.68 136	58.77 175	35.28 194	46.35 84	32.05 166	12.44 198	28.53 181	35.52 143	14.40 214	57.12 166	56.93 170	71.11 146	59.69 186

CP-MVSNet		37.09 198	40.62 196	32.99 199	37.56 202	48.25 204	32.75 204	43.05 128	27.88 189	5.93 219	31.27 159	25.82 204	15.09 209	43.37 212	48.82 201	63.54 191	58.90 187

ambc			35.52 213		38.36 198	40.40 219	28.38 216		25.20 206	14.87 191	13.22 224	7.54 235	19.34 201	55.63 172	47.79 209	47.91 223	58.89 188

PEN-MVS		38.23 194	41.72 194	34.15 197	40.56 193	50.07 199	33.17 203	44.35 113	27.64 192	5.54 223	30.84 165	26.67 199	14.99 211	45.64 203	52.38 193	66.29 178	58.83 189

testpf		31.84 214	34.86 214	28.32 214	48.89 159	32.91 228	26.53 219	25.77 225	21.99 209	10.05 208	23.39 198	25.55 205	14.07 215	39.23 223	42.32 219	44.58 225	58.65 190

PatchMatch-RL		43.37 180	44.93 188	41.56 172	37.94 200	51.70 193	40.02 184	35.75 182	39.04 110	30.71 124	35.14 117	27.43 198	46.58 113	51.99 190	50.55 199	58.38 207	58.64 191

PS-CasMVS		36.84 200	40.23 200	32.89 200	37.44 203	48.09 205	32.68 205	42.97 130	27.36 193	5.89 220	30.08 173	25.48 206	14.96 212	43.28 213	48.71 202	63.39 192	58.63 192

LS3D		49.59 140	49.75 158	49.40 121	55.88 95	59.86 170	56.31 111	45.33 91	48.57 79	28.32 134	31.54 156	36.81 129	46.27 116	57.17 164	55.88 176	64.29 186	58.42 193

tfpn1000		41.76 190	45.01 187	37.96 189	50.95 143	58.44 180	34.94 196	44.09 115	30.68 176	12.08 199	30.14 170	31.96 170	18.67 202	51.96 191	53.45 188	67.05 176	58.40 194

WR-MVS		37.61 196	42.15 193	32.31 206	43.64 185	51.85 192	29.39 212	43.35 124	27.65 191	4.40 225	29.90 174	29.80 188	10.46 221	46.73 202	51.98 195	62.60 196	57.16 195

EU-MVSNet		33.00 213	36.49 211	28.92 212	33.10 218	42.86 216	29.32 214	35.99 180	22.94 208	5.83 221	25.29 194	24.43 208	15.21 208	41.22 221	41.65 222	54.08 215	57.01 196

tfpnview11		42.71 185	45.29 183	39.71 178	51.06 141	58.61 179	38.47 187	44.80 105	30.44 178	13.60 192	31.25 160	30.97 183	22.40 195	54.20 179	55.04 182	67.90 167	56.51 197

UA-Net		47.19 157	53.02 128	40.38 177	55.31 98	60.02 169	38.41 189	38.68 164	36.42 130	22.47 164	51.95 50	58.72 53	25.62 191	54.11 180	53.40 189	61.79 199	56.51 197

FMVSNet5		43.29 181	47.07 172	38.87 183	30.46 222	50.99 196	45.87 167	37.19 174	42.17 97	19.32 177	26.77 185	40.51 112	30.26 175	56.82 169	55.81 177	70.10 155	56.46 199

tfpn_n400		42.55 186	45.11 185	39.55 180	50.95 143	58.68 177	38.40 190	44.75 107	29.29 183	13.60 192	31.25 160	30.97 183	22.38 196	53.96 183	55.66 178	67.20 174	56.00 200

tfpnconf		42.55 186	45.11 185	39.55 180	50.95 143	58.68 177	38.40 190	44.75 107	29.29 183	13.60 192	31.25 160	30.97 183	22.38 196	53.96 183	55.66 178	67.20 174	56.00 200

WR-MVS_H		36.29 201	40.35 199	31.55 208	37.80 201	49.94 201	30.57 211	41.11 149	26.90 196	4.14 227	30.72 167	28.85 192	10.45 222	42.47 216	47.99 207	65.24 181	55.54 202

LTVRE_ROB	32.83 17	35.10 205	37.46 206	32.35 205	43.12 187	49.99 200	28.52 215	33.23 205	12.73 231	8.18 212	27.71 184	21.34 213	32.64 167	46.92 201	48.11 205	48.41 221	55.45 203
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
DTE-MVSNet		36.91 199	40.44 197	32.79 202	40.74 191	47.55 206	30.71 210	44.39 110	27.03 195	4.32 226	30.88 163	25.99 202	12.73 216	45.58 204	50.80 198	63.86 188	55.23 204

pmmvs3		31.22 215	33.62 216	28.43 213	22.82 226	40.26 220	26.40 220	22.05 229	16.89 225	10.99 205	14.72 220	16.26 222	29.70 180	44.82 206	47.39 210	58.61 206	54.98 205

SixPastTwentyTwo		36.11 202	37.80 204	34.13 198	37.13 206	46.72 208	34.58 198	34.96 188	21.20 214	11.66 200	29.15 180	19.88 217	29.77 179	44.93 205	48.34 204	56.67 211	54.41 206

test2356		34.09 210	36.84 210	30.87 211	36.25 211	43.59 214	27.92 218	35.44 185	21.73 211	6.94 215	19.31 208	18.23 219	17.77 204	49.28 196	51.58 196	60.94 201	54.17 207

test20.03		36.00 203	38.92 201	32.60 203	45.92 170	50.99 196	28.05 217	43.69 121	21.62 212	6.03 218	17.61 214	25.91 203	8.34 228	51.26 192	52.60 192	63.58 189	52.46 208

TinyColmap		37.18 197	37.37 209	36.95 192	31.17 221	45.21 211	39.71 185	34.65 190	29.83 182	20.20 172	18.54 210	13.72 227	38.27 144	50.33 195	51.57 197	57.71 208	52.42 209

Anonymous20231211		34.77 207	32.10 220	37.88 190	44.26 182	46.14 209	29.39 212	39.72 155	14.08 230	38.09 75	7.94 231	13.30 228	16.12 207	43.07 214	47.88 208	59.72 203	52.28 210

TDRefinement		35.76 204	38.23 202	32.88 201	19.09 233	46.04 210	43.29 177	29.49 215	33.49 159	19.04 181	22.29 203	17.82 221	29.69 181	48.60 198	47.24 211	56.65 212	52.12 211

MDA-MVSNet-bldmvs		34.31 209	34.11 215	34.54 196	24.73 225	49.66 202	33.42 202	43.03 129	21.59 213	11.10 204	19.81 206	12.68 229	31.41 171	35.59 224	48.05 206	63.56 190	51.39 212

COLMAP_ROB	34.79 15	38.65 193	40.72 195	36.23 194	36.41 210	49.22 203	45.51 170	27.60 220	37.81 116	20.54 170	23.37 199	24.25 209	28.11 187	51.02 193	48.55 203	59.22 204	50.82 213
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
RPSCF		33.61 212	40.43 198	25.65 218	16.00 235	32.41 229	31.73 209	13.33 234	50.13 72	23.12 159	31.56 155	40.09 113	32.73 166	41.14 222	37.05 224	36.99 229	50.63 214

LP		38.21 195	37.72 205	38.79 184	44.07 184	51.16 194	35.54 193	31.37 213	25.38 205	23.73 157	18.64 209	18.03 220	29.31 182	47.85 199	52.63 191	68.71 163	50.34 215

new-patchmatchnet		30.47 217	32.80 218	27.75 216	36.81 208	43.98 212	24.85 222	39.29 158	20.52 215	4.06 228	15.94 219	16.05 223	9.57 223	41.32 220	42.05 221	51.94 219	49.74 216

N_pmnet		34.09 210	35.74 212	32.17 207	37.25 205	43.17 215	32.26 208	35.57 184	26.22 199	10.60 206	20.44 205	19.38 218	20.20 200	44.59 209	47.00 212	57.13 210	49.35 217

testmv		27.97 221	29.98 222	25.62 219	32.54 219	36.86 225	20.53 227	33.33 203	14.11 228	2.64 232	12.76 226	11.77 230	11.07 219	42.34 217	45.44 216	53.60 217	46.60 218

test1235678		27.96 222	29.97 223	25.62 219	32.54 219	36.83 226	20.53 227	33.33 203	14.10 229	2.64 232	12.75 227	11.76 231	11.07 219	42.34 217	45.43 217	53.60 217	46.59 219

testus		29.45 219	32.20 219	26.23 217	37.01 207	37.90 222	17.56 231	35.70 183	18.23 221	3.39 230	17.04 217	14.78 224	11.78 218	42.48 215	49.38 200	51.92 220	45.62 220

MIMVSNet1		29.60 218	33.37 217	25.20 222	19.52 231	43.94 213	26.29 221	37.92 169	19.95 219	3.79 229	12.64 228	21.99 211	7.70 231	43.83 211	46.32 215	55.97 213	44.92 221

FC-MVSNet-test		30.97 216	37.38 208	23.49 223	37.42 204	33.68 227	19.43 230	39.27 159	31.37 172	1.67 237	38.56 97	28.85 192	6.06 233	41.40 219	43.80 218	37.10 228	44.03 222

testgi		34.51 208	37.42 207	31.12 210	47.37 167	50.34 198	24.38 224	41.21 147	20.32 216	5.64 222	20.56 204	26.55 200	8.06 229	49.28 196	52.65 190	60.05 202	42.23 223

no-one		21.91 225	22.52 228	21.20 225	25.97 224	30.78 230	13.29 234	32.75 207	9.08 234	1.84 234	6.18 233	7.00 236	8.03 230	25.56 230	40.16 223	45.29 224	38.83 224

FPMVS		26.87 223	28.19 224	25.32 221	27.09 223	29.49 231	32.28 207	17.79 231	28.09 188	11.33 201	19.38 207	14.69 225	20.88 199	35.11 225	32.82 227	42.56 226	37.75 225

PMVS	18.18 18	21.95 224	22.85 225	20.90 226	21.92 229	14.78 234	19.95 229	17.31 232	15.69 226	11.32 202	13.70 221	13.91 226	15.02 210	34.92 226	31.72 228	39.85 227	35.20 226
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
1111		29.41 220	30.75 221	27.85 215	39.46 194	37.63 223	22.26 225	32.15 208	17.93 223	7.92 213	13.48 222	20.98 214	17.30 205	44.76 207	46.51 214	47.99 222	33.96 227

test12356		20.09 226	22.80 226	16.93 228	22.59 228	24.43 232	13.32 233	25.93 224	12.67 232	1.58 239	11.53 229	9.25 233	2.29 234	33.15 228	37.05 224	35.85 230	31.54 228

new_pmnet		19.10 228	22.71 227	14.89 230	10.93 237	24.08 233	14.22 232	13.94 233	18.68 220	2.93 231	12.84 225	11.27 232	11.94 217	30.57 229	30.58 229	35.38 231	30.93 229

PMMVS2		12.25 230	14.17 231	10.00 233	11.39 236	14.35 235	8.21 236	19.29 230	9.31 233	0.19 241	7.38 232	6.19 237	1.10 236	19.26 231	21.13 231	19.85 233	21.56 230

Gipuma		17.16 229	17.83 230	16.36 229	18.76 234	12.15 237	11.97 235	27.78 219	17.94 222	4.86 224	2.53 237	2.73 239	8.90 226	34.32 227	36.09 226	25.92 232	19.06 231
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
MVE	10.35 19	9.76 233	11.08 232	8.22 234	4.43 238	13.04 236	3.36 240	23.57 228	5.74 238	1.76 235	3.09 236	1.75 240	6.78 232	12.78 233	23.04 230	9.44 236	18.09 232
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
DeepMVS_CX						5.87 240	4.32 238	1.74 235	9.04 235	1.30 240	7.97 230	3.16 238	8.56 227	9.74 234		6.30 238	14.51 233

E-PMN		10.66 231	8.30 233	13.42 231	19.91 230	7.87 238	4.30 239	29.47 216	8.37 237	1.70 236	3.67 234	1.29 242	9.12 225	8.98 235	13.59 232	16.03 234	14.30 234

EMVS		10.15 232	7.67 234	13.05 232	19.22 232	7.77 239	4.48 237	29.34 217	8.65 236	1.67 237	3.55 235	1.36 241	9.15 224	8.15 236	11.79 234	14.44 235	12.43 235

.test1245		19.53 227	19.26 229	19.85 227	39.46 194	37.63 223	22.26 225	32.15 208	17.93 223	7.92 213	13.48 222	20.98 214	17.30 205	44.76 207	0.01 235	0.00 239	0.03 236

testmvs		0.01 234	0.01 235	0.00 236	0.00 241	0.00 242	0.00 243	0.00 238	0.01 239	0.00 243	0.02 238	0.00 243	0.00 239	0.01 237	0.01 235	0.00 239	0.03 236

test123		0.01 234	0.01 235	0.00 236	0.00 241	0.00 242	0.00 243	0.00 238	0.01 239	0.00 243	0.02 238	0.00 243	0.01 237	0.00 238	0.01 235	0.00 239	0.03 236

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

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

our_test_3					49.68 153	61.50 157	45.84 168

MTAPA										54.82 10		71.98 18

MTMP										50.64 27		68.31 24

Patchmatch-RL test							0.69 242

tmp_tt				4.41 235	2.56 239	1.81 241	2.61 241	0.27 236	20.12 217	9.81 209	17.69 213	9.04 234	1.96 235	12.88 232	12.11 233	9.23 237

XVS					62.70 51	73.06 55	61.80 63			42.02 65		63.42 37				74.68 88

X-MVStestdata					62.70 51	73.06 55	61.80 63			42.02 65		63.42 37				74.68 88

mPP-MVS					63.08 48							62.34 40

NP-MVS									72.62 25

Patchmtry						64.49 126	52.06 137	34.21 195		31.74 118

High-resolution multi-view benchmark