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.

CHOSEN 1792x2688		72.55 67	71.98 76	73.22 57	86.57 43	92.41 6	75.63 67	66.77 48	62.08 84	52.32 84	30.27 183	50.74 125	66.14 80	86.22 10	85.41 7	91.90 1	96.75 12

MCST-MVS		85.75 6	86.99 10	84.31 3	94.07 1	92.80 4	88.15 5	79.10 1	85.66 20	70.72 27	76.50 29	80.45 17	82.17 2	88.35 2	87.49 3	91.63 2	97.65 3

gg-mvs-nofinetune		62.34 130	66.19 114	57.86 155	76.15 99	88.61 36	71.18 104	41.24 199	25.74 201	13.16 202	22.91 196	63.97 65	54.52 141	85.06 14	85.25 10	90.92 3	91.78 58

gm-plane-assit		54.99 170	57.99 166	51.49 178	69.27 138	54.42 202	32.32 205	42.59 192	21.18 205	13.71 200	23.61 193	43.84 139	60.21 112	87.09 5	86.55 5	90.81 4	89.28 86

MVS_Test		75.22 53	76.69 54	73.51 54	79.30 75	88.82 34	80.06 46	58.74 109	69.77 61	57.50 71	59.78 66	61.35 75	75.31 23	82.07 41	83.60 24	90.13 5	91.41 61

DPM-MVS		85.41 8	86.72 13	83.89 7	91.66 9	91.92 11	90.49 1	78.09 2	86.90 15	73.95 17	74.52 31	82.01 12	79.29 9	90.24 1	90.65 1	89.86 6	90.78 69

baseline1		71.47 72	72.02 75	70.82 71	80.56 70	84.51 76	76.61 64	66.93 47	56.22 106	48.66 100	55.40 74	60.43 81	62.55 98	83.35 25	80.99 52	89.60 7	83.28 136

DWT-MVSNet_training		72.81 65	73.98 66	71.45 67	81.26 63	86.37 62	72.08 91	59.82 106	69.13 67	58.15 67	54.71 76	61.33 77	67.81 73	76.86 84	78.63 69	89.59 8	90.86 66

MS-PatchMatch		70.34 81	69.00 94	71.91 66	85.20 49	85.35 70	77.84 57	61.77 89	58.01 98	55.40 78	41.26 136	58.34 93	61.69 102	81.70 47	78.29 74	89.56 9	80.02 153

CANet		80.90 25	82.93 26	78.53 28	86.83 42	92.26 8	81.19 39	66.95 46	81.60 33	69.90 30	66.93 43	74.80 29	76.79 17	84.68 17	84.77 15	89.50 10	95.50 19

QAPM		77.50 42	77.43 47	77.59 33	91.52 10	92.00 10	81.41 37	70.63 24	66.22 71	58.05 68	54.70 77	71.79 41	74.49 30	82.46 34	82.04 33	89.46 11	92.79 50

MVS_0304		79.43 30	82.20 28	76.20 39	84.22 50	91.79 13	81.82 34	63.81 67	76.83 47	61.71 52	66.37 46	75.52 28	76.38 20	85.54 12	85.03 12	89.28 12	94.32 32

DELS-MVS		79.49 28	79.84 37	79.08 25	88.26 34	92.49 5	84.12 22	70.63 24	65.27 78	69.60 33	61.29 59	66.50 57	72.75 39	88.07 3	88.03 2	89.13 13	97.22 6
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
casdiffmvs		75.20 54	75.69 60	74.63 52	79.26 76	89.07 32	78.47 52	63.59 70	67.05 69	63.79 44	55.72 73	60.32 82	73.58 33	82.16 40	81.78 38	89.08 14	93.72 41

MSLP-MVS++		78.57 35	77.33 48	80.02 19	88.39 31	84.79 74	84.62 19	66.17 53	75.96 49	78.40 7	61.59 57	71.47 42	73.54 35	78.43 71	78.88 68	88.97 15	90.18 77

baseline2		71.22 76	73.01 73	69.13 81	75.76 102	86.34 63	71.23 102	62.78 81	62.62 81	52.85 83	57.32 69	54.31 112	63.27 94	79.74 61	79.31 64	88.89 16	91.43 59

abl_6				79.06 26	89.68 21	92.14 9	77.70 58	69.68 30	86.87 16	71.88 22	74.29 32	80.06 19	76.56 19			88.84 17	95.82 13

CNVR-MVS		85.96 5	87.58 8	84.06 5	92.58 3	92.40 7	87.62 7	77.77 3	88.44 11	75.93 14	79.49 22	81.97 13	81.65 3	87.04 6	86.58 4	88.79 18	97.18 7

IB-MVS	64.48 11	69.02 85	68.97 95	69.09 83	81.75 60	89.01 33	64.50 142	64.91 61	56.65 102	62.59 48	47.89 102	45.23 136	51.99 146	69.18 161	81.88 37	88.77 19	92.93 47
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
GG-mvs-BLEND		54.54 173	77.58 46	27.67 201	0.03 214	90.09 26	77.20 62	0.02 211	66.83 70	0.05 214	59.90 64	73.33 33	0.04 210	78.40 72	79.30 65	88.65 20	95.20 24

FMVSNet3		70.41 79	71.89 78	68.68 85	70.89 129	79.42 120	75.63 67	60.97 97	65.32 75	51.06 89	47.37 106	62.05 69	64.90 85	82.49 33	82.27 32	88.64 21	84.34 127

test-LLR		68.23 91	71.61 80	64.28 112	71.37 124	81.32 102	63.98 147	61.03 95	58.62 95	42.96 125	52.74 86	61.65 73	57.74 130	75.64 96	78.09 78	88.61 22	93.21 43

TESTMET0.1,1		67.38 98	71.61 80	62.45 128	66.05 157	81.32 102	63.98 147	55.36 145	58.62 95	42.96 125	52.74 86	61.65 73	57.74 130	75.64 96	78.09 78	88.61 22	93.21 43

3Dnovator	70.49 5	78.42 36	76.77 53	80.35 17	91.43 11	90.27 22	81.84 33	70.79 23	72.10 53	71.95 21	50.02 95	67.86 53	77.47 15	82.89 28	84.24 18	88.61 22	89.99 78

MVSTER		76.92 46	79.92 36	73.42 56	74.98 107	82.97 87	78.15 54	63.41 71	78.02 43	64.41 43	67.54 41	72.80 34	71.05 50	83.29 26	83.73 21	88.53 25	91.12 64

DVP-MVS		88.07 1	90.73 1	84.97 1	91.98 6	95.01 1	87.86 6	76.88 4	93.90 1	85.15 1	90.11 5	86.90 1	79.46 8	86.26 9	84.67 16	88.50 26	98.25 2

NCCC		84.16 13	85.46 18	82.64 8	92.34 4	90.57 20	86.57 10	76.51 5	86.85 17	72.91 20	77.20 28	78.69 23	79.09 11	84.64 18	84.88 14	88.44 27	95.41 21

CLD-MVS		77.36 44	77.29 49	77.45 34	82.21 57	88.11 43	81.92 32	68.96 36	77.97 44	69.62 32	62.08 55	59.44 87	73.57 34	81.75 46	81.27 48	88.41 28	90.39 74
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
MSP-MVS		87.87 2	90.57 2	84.73 2	89.38 24	91.60 14	88.24 4	74.15 9	93.55 2	82.28 2	94.99 1	83.21 7	85.96 1	87.67 4	84.67 16	88.32 29	98.29 1

HyFIR lowres test		68.39 89	68.28 100	68.52 86	80.85 66	88.11 43	71.08 106	58.09 114	54.87 117	47.80 105	27.55 189	55.80 105	64.97 84	79.11 65	79.14 66	88.31 30	93.35 42

EPMVS		66.21 102	67.49 105	64.73 106	75.81 101	84.20 82	68.94 119	44.37 186	61.55 85	48.07 104	49.21 98	54.87 111	62.88 95	71.82 137	71.40 150	88.28 31	79.37 156

MVS_111021_HR		77.42 43	78.40 44	76.28 38	86.95 40	90.68 18	77.41 60	70.56 27	66.21 72	62.48 49	66.17 47	63.98 64	72.08 44	82.87 29	83.15 26	88.24 32	95.71 16

PVSNet_BlendedMVS		76.84 47	78.47 42	74.95 48	82.37 55	89.90 27	75.45 71	65.45 58	74.99 50	70.66 28	63.07 52	58.27 94	67.60 74	84.24 20	81.70 40	88.18 33	97.10 8

PVSNet_Blended		76.84 47	78.47 42	74.95 48	82.37 55	89.90 27	75.45 71	65.45 58	74.99 50	70.66 28	63.07 52	58.27 94	67.60 74	84.24 20	81.70 40	88.18 33	97.10 8

CS-MVS		75.18 55	78.59 41	71.20 68	77.74 84	87.69 48	73.93 86	58.81 108	69.17 66	55.73 74	67.86 40	66.89 55	72.87 38	82.50 32	81.29 47	88.15 35	94.71 27

TSAR-MVS + GP.		82.27 20	85.98 16	77.94 30	80.72 69	88.25 41	81.12 40	67.71 42	87.10 13	73.31 18	85.23 12	83.68 6	76.64 18	80.43 57	81.47 44	88.15 35	95.66 17

DeepC-MVS_fast	75.41 2	81.69 21	82.10 30	81.20 14	91.04 13	87.81 47	83.42 24	74.04 10	83.77 24	71.09 25	66.88 44	72.44 35	79.48 7	85.08 13	84.97 13	88.12 37	93.78 39
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
EPNet		79.28 34	82.25 27	75.83 42	88.31 33	90.14 24	79.43 49	68.07 40	81.76 32	61.26 54	77.26 27	70.08 47	70.06 56	82.43 36	82.00 35	87.82 38	92.09 54
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
SteuartSystems-ACMMP		82.51 18	85.35 19	79.20 23	90.25 14	89.39 30	84.79 18	70.95 22	82.86 26	68.32 35	86.44 11	77.19 24	73.07 36	83.63 23	83.64 22	87.82 38	94.34 31
Skip Steuart: Steuart Systems R&D Blog.
diffmvs		74.32 57	75.42 61	73.04 58	75.60 104	87.27 53	78.20 53	62.96 75	68.66 68	61.89 50	59.79 65	59.84 85	71.80 45	78.30 74	79.87 61	87.80 40	94.23 34

ETV-MVS		76.16 50	79.12 40	72.71 60	78.08 81	87.54 51	74.15 85	61.37 93	70.55 58	59.25 64	68.75 38	67.36 54	75.64 22	81.94 44	81.90 36	87.78 41	95.79 14

tpmrst		67.15 100	68.12 102	66.03 99	76.21 98	80.98 105	71.27 101	45.05 182	60.69 89	50.63 93	46.95 114	54.15 114	65.30 82	71.80 138	71.77 144	87.72 42	90.48 73

APDe-MVS		86.37 4	88.41 5	84.00 6	91.43 11	91.83 12	88.34 3	74.67 7	91.19 5	81.76 3	91.13 3	81.94 14	80.07 5	83.38 24	82.58 31	87.69 43	96.78 10

CostFormer		72.18 68	73.90 68	70.18 76	79.47 73	86.19 67	76.94 63	48.62 171	66.07 74	60.40 61	54.14 83	65.82 58	67.98 71	75.84 94	76.41 91	87.67 44	92.83 49

DPE-MVS		87.60 3	90.44 3	84.29 4	92.09 5	93.44 2	88.69 2	75.11 6	93.06 4	80.80 4	94.23 2	86.70 2	81.44 4	84.84 16	83.52 25	87.64 45	97.28 5

CANet_DTU		72.84 64	76.63 55	68.43 87	76.81 94	86.62 59	75.54 70	54.71 153	72.06 54	43.54 120	67.11 42	58.46 91	72.40 41	81.13 53	80.82 57	87.57 46	90.21 76

OpenMVS	67.62 8	74.92 56	73.91 67	76.09 41	90.10 17	90.38 21	78.01 55	66.35 51	66.09 73	62.80 46	46.33 118	64.55 63	71.77 46	79.92 60	80.88 56	87.52 47	89.20 87

GBi-Net		69.21 82	70.40 87	67.81 90	69.49 134	78.65 125	74.54 78	60.97 97	65.32 75	51.06 89	47.37 106	62.05 69	63.43 91	77.49 76	78.22 75	87.37 48	83.73 130

test1		69.21 82	70.40 87	67.81 90	69.49 134	78.65 125	74.54 78	60.97 97	65.32 75	51.06 89	47.37 106	62.05 69	63.43 91	77.49 76	78.22 75	87.37 48	83.73 130

FMVSNet2		68.06 92	68.57 97	67.45 94	69.49 134	78.65 125	74.54 78	60.23 105	56.29 105	49.64 98	42.13 132	57.08 99	63.43 91	81.15 52	80.99 52	87.37 48	83.73 130

canonicalmvs		77.65 40	79.59 38	75.39 44	81.52 61	89.83 29	81.32 38	60.74 101	80.05 38	66.72 37	68.43 39	65.09 60	74.72 29	78.87 67	82.73 29	87.32 51	92.16 53

HPM-MVS++		85.64 7	88.43 4	82.39 9	92.65 2	90.24 23	85.83 13	74.21 8	90.68 7	75.63 15	86.77 10	84.15 5	78.68 12	86.33 7	85.26 9	87.32 51	95.60 18

ET-MVSNet_ETH3D		71.38 74	74.70 64	67.51 93	51.61 198	88.06 45	77.29 61	60.95 100	63.61 80	48.36 102	66.60 45	60.67 80	79.55 6	73.56 118	80.58 58	87.30 53	89.80 80

CSCG		82.90 17	84.52 20	81.02 15	91.85 7	93.43 3	87.14 8	74.01 11	81.96 30	76.14 12	70.84 34	82.49 9	69.71 58	82.32 38	85.18 11	87.26 54	95.40 22

SMA-MVS		85.24 9	88.27 6	81.72 12	91.74 8	90.71 17	86.71 9	73.16 16	90.56 8	74.33 16	83.07 15	85.88 3	77.16 16	86.28 8	85.58 6	87.23 55	95.77 15

CDPH-MVS		79.39 33	82.13 29	76.19 40	89.22 27	88.34 39	84.20 21	71.00 21	79.67 40	56.97 72	77.77 25	72.24 39	68.50 70	81.33 49	82.74 28	87.23 55	92.84 48

EIA-MVS		73.48 61	76.05 57	70.47 74	78.12 80	87.21 54	71.78 95	60.63 102	69.66 62	55.56 77	64.86 49	60.69 79	69.53 61	77.35 80	78.59 70	87.22 57	94.01 37

test-mter		64.06 119	69.24 92	58.01 153	59.07 185	77.40 138	59.13 169	48.11 174	55.64 111	39.18 143	51.56 91	58.54 89	55.38 138	73.52 119	76.00 96	87.22 57	92.05 56

train_agg		83.35 15	86.93 12	79.17 24	89.70 20	88.41 38	85.60 16	72.89 18	86.31 18	66.58 38	90.48 4	82.24 11	73.06 37	83.10 27	82.64 30	87.21 59	95.30 23

DI_MVS_plusplus_trai		73.94 60	74.85 63	72.88 59	76.57 96	86.80 57	80.41 44	61.47 91	62.35 83	59.44 63	47.91 101	68.12 50	72.24 42	82.84 30	81.50 43	87.15 60	94.42 30

MAR-MVS		77.19 45	78.37 45	75.81 43	89.87 18	90.58 19	79.33 50	65.56 57	77.62 46	58.33 66	59.24 67	67.98 51	74.83 26	82.37 37	83.12 27	86.95 61	87.67 103
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
baseline		72.89 63	74.46 65	71.07 69	75.99 100	87.50 52	74.57 77	60.49 103	70.72 57	57.60 69	60.63 62	60.97 78	70.79 52	75.27 99	76.33 92	86.94 62	89.79 81

APD-MVS		84.83 10	87.00 9	82.30 10	89.61 22	89.21 31	86.51 11	73.64 13	90.98 6	77.99 9	89.89 6	80.04 20	79.18 10	82.00 43	81.37 45	86.88 63	95.49 20
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
MIMVSNet		57.78 163	59.71 158	55.53 167	54.79 193	77.10 141	63.89 149	45.02 183	46.59 145	36.79 155	28.36 187	40.77 150	45.84 168	74.97 101	76.58 88	86.87 64	73.60 173

DeepC-MVS	74.46 3	80.30 27	81.05 33	79.42 21	87.42 38	88.50 37	83.23 25	73.27 15	82.78 27	71.01 26	62.86 54	69.93 48	74.80 27	84.30 19	84.20 19	86.79 65	94.77 25
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
TSAR-MVS + MP.		84.39 11	86.58 14	81.83 11	88.09 36	86.47 61	85.63 15	73.62 14	90.13 9	79.24 6	89.67 7	82.99 8	77.72 14	81.22 50	80.92 55	86.68 66	94.66 28
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
ACMMP_NAP		83.54 14	86.37 15	80.25 18	89.57 23	90.10 25	85.27 17	71.66 20	87.38 12	73.08 19	84.23 14	80.16 18	75.31 23	84.85 15	83.64 22	86.57 67	94.21 35

tpm		64.85 112	66.02 116	63.48 118	74.52 109	78.38 128	70.98 108	44.99 184	51.61 125	43.28 124	47.66 104	53.18 118	60.57 108	70.58 149	71.30 153	86.54 68	89.45 85

PVSNet_Blended_VisFu		71.76 71	73.54 70	69.69 77	79.01 77	87.16 55	72.05 92	61.80 88	56.46 104	59.66 62	53.88 85	62.48 67	59.08 121	81.17 51	78.90 67	86.53 69	94.74 26

Fast-Effi-MVS+		67.59 94	67.56 104	67.62 92	73.67 112	81.14 104	71.12 105	54.79 152	58.88 94	50.61 94	46.70 116	47.05 132	69.12 67	76.06 92	76.44 90	86.43 70	86.65 108

GA-MVS		64.55 115	65.76 118	63.12 120	69.68 133	81.56 98	69.59 115	58.16 113	45.23 151	35.58 162	47.01 113	41.82 142	59.41 117	79.62 62	78.54 71	86.32 71	86.56 109

HFP-MVS		82.48 19	84.12 21	80.56 16	90.15 15	87.55 50	84.28 20	69.67 31	85.22 21	77.95 10	84.69 13	75.94 27	75.04 25	81.85 45	81.17 49	86.30 72	92.40 52

HQP-MVS		78.26 37	80.91 34	75.17 47	85.67 47	84.33 80	83.01 27	69.38 32	79.88 39	55.83 73	79.85 21	64.90 62	70.81 51	82.46 34	81.78 38	86.30 72	93.18 46

Effi-MVS+		70.42 77	71.23 82	69.47 78	78.04 82	85.24 71	75.57 69	58.88 107	59.56 92	48.47 101	52.73 88	54.94 110	69.69 59	78.34 73	77.06 84	86.18 74	90.73 72

PGM-MVS		79.42 32	81.84 31	76.60 37	88.38 32	86.69 58	82.97 28	65.75 55	80.39 37	64.94 41	81.95 19	72.11 40	71.41 49	80.45 56	80.55 59	86.18 74	90.76 71

PHI-MVS		79.43 30	84.06 22	74.04 53	86.15 45	91.57 15	80.85 43	68.90 37	82.22 29	51.81 87	78.10 24	74.28 30	70.39 55	84.01 22	84.00 20	86.14 76	94.24 33

ACMMPR		80.62 26	82.98 25	77.87 31	88.41 30	87.05 56	83.02 26	69.18 34	83.91 23	68.35 34	82.89 16	73.64 32	72.16 43	80.78 55	81.13 51	86.10 77	91.43 59

PatchmatchNet		65.43 109	67.71 103	62.78 124	73.49 114	82.83 88	66.42 138	45.40 181	60.40 90	45.27 111	49.22 97	57.60 98	60.01 113	70.61 147	71.38 151	86.08 78	81.91 147
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
v2v482		63.68 122	62.85 134	64.65 107	68.01 143	80.46 111	71.90 93	57.60 122	44.26 153	42.82 127	39.80 145	38.62 159	61.56 103	73.06 123	74.86 107	86.03 79	88.90 92

zzz-MVS		81.65 22	83.10 24	79.97 20	88.14 35	87.62 49	83.96 23	69.90 28	86.92 14	77.67 11	72.47 33	78.74 22	74.13 31	81.59 48	81.15 50	86.01 80	93.19 45

FC-MVSNet-train		68.83 86	68.29 99	69.47 78	78.35 79	79.94 114	64.72 141	66.38 50	54.96 114	54.51 81	56.75 70	47.91 131	66.91 78	75.57 98	75.75 98	85.92 81	87.12 105

v8		63.44 124	62.58 136	64.43 109	68.28 142	78.07 130	71.82 94	54.85 150	46.70 144	45.20 112	39.40 146	40.91 148	60.54 109	72.85 127	74.39 115	85.92 81	85.76 118

3Dnovator+	70.16 6	77.87 39	77.29 49	78.55 27	89.25 26	88.32 40	80.09 45	67.95 41	74.89 52	71.83 23	52.05 89	70.68 45	76.27 21	82.27 39	82.04 33	85.92 81	90.77 70

MP-MVS		80.94 24	83.49 23	77.96 29	88.48 29	88.16 42	82.82 29	69.34 33	80.79 36	69.67 31	82.35 17	77.13 25	71.60 48	80.97 54	80.96 54	85.87 84	94.06 36
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
v10		63.00 127	62.22 139	63.90 116	67.88 145	77.78 134	71.59 98	54.34 154	45.37 150	42.76 128	38.53 148	38.93 157	61.05 107	74.39 108	74.52 113	85.75 85	86.04 114

ACMMP		77.61 41	79.59 38	75.30 46	85.87 46	85.58 69	81.42 36	67.38 45	79.38 41	62.61 47	78.53 23	65.79 59	68.80 69	78.56 70	78.50 73	85.75 85	90.80 68
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
OPM-MVS		72.74 66	70.93 85	74.85 50	85.30 48	84.34 79	82.82 29	69.79 29	49.96 130	55.39 79	54.09 84	60.14 84	70.04 57	80.38 58	79.43 63	85.74 87	88.20 99

thres100view900		67.14 101	66.09 115	68.38 88	77.70 85	83.84 84	74.52 81	66.33 52	49.16 134	43.40 122	43.24 121	41.34 143	62.59 97	79.31 64	75.92 97	85.73 88	89.81 79

CP-MVS		79.44 29	81.51 32	77.02 35	86.95 40	85.96 68	82.00 31	68.44 39	81.82 31	67.39 36	77.43 26	73.68 31	71.62 47	79.56 63	79.58 62	85.73 88	92.51 51

pmmvs5		59.72 150	60.24 154	59.11 149	62.77 174	77.33 140	63.17 155	54.00 155	40.21 169	37.23 152	40.41 141	35.99 172	51.75 147	72.55 132	72.74 138	85.72 90	82.45 143

EPNet_dtu		66.17 103	70.13 90	61.54 134	81.04 64	77.39 139	68.87 120	62.50 82	69.78 60	33.51 169	63.77 51	56.22 102	37.65 182	72.20 133	72.18 143	85.69 91	79.38 155
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
XVS					82.43 53	86.27 64	75.70 65			61.07 56		72.27 36				85.67 92

X-MVStestdata					82.43 53	86.27 64	75.70 65			61.07 56		72.27 36				85.67 92

PatchT		60.46 147	63.85 125	56.51 164	65.95 159	75.68 152	47.34 188	41.39 196	53.89 121	41.40 131	37.84 154	50.30 126	57.29 133	72.76 128	73.27 130	85.67 92	83.23 137

X-MVS		78.16 38	80.55 35	75.38 45	87.99 37	86.27 64	81.05 41	68.98 35	78.33 42	61.07 56	75.25 30	72.27 36	67.52 77	80.03 59	80.52 60	85.66 95	91.20 63

v1144		63.00 127	62.39 138	63.70 117	67.72 146	80.27 112	71.23 102	56.40 131	42.51 158	40.81 136	38.12 153	37.73 160	60.42 111	74.46 106	74.55 112	85.64 96	89.12 88

v144192		62.05 137	61.46 145	62.73 127	66.59 155	79.87 115	69.30 117	55.88 136	41.50 165	39.41 141	37.23 156	36.45 168	59.62 115	72.69 130	73.51 123	85.61 97	88.93 90

SD-MVS		84.31 12	86.96 11	81.22 13	88.98 28	88.68 35	85.65 14	73.85 12	89.09 10	79.63 5	87.34 9	84.84 4	73.71 32	82.66 31	81.60 42	85.48 98	94.51 29
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
v1921920		61.66 140	61.10 148	62.31 129	66.32 156	79.57 118	68.41 122	55.49 143	41.03 166	38.69 145	36.64 164	35.27 176	59.60 116	73.23 121	73.41 125	85.37 99	88.51 97

UniMVSNet_NR-MVSNet		62.30 132	63.51 127	60.89 136	69.48 137	77.83 133	64.07 145	63.94 66	50.03 129	31.17 173	44.82 119	41.12 146	51.37 149	71.02 143	74.81 108	85.30 100	84.95 122

thres400		65.18 111	64.44 123	66.04 98	76.40 97	82.63 89	71.52 99	64.27 64	44.93 152	40.69 137	41.86 133	40.79 149	58.12 125	77.67 75	74.64 109	85.26 101	88.56 95

v1192		62.25 133	61.64 143	62.96 121	66.88 151	79.72 116	69.96 113	55.77 138	41.58 163	39.42 140	37.05 158	35.96 173	60.50 110	74.30 111	74.09 117	85.24 102	88.76 93

NR-MVSNet		61.08 144	62.09 141	59.90 141	71.96 121	75.87 148	63.60 151	61.96 84	49.31 132	27.95 178	42.76 123	33.85 183	48.82 156	74.35 109	74.05 119	85.13 103	84.45 125

v1240		61.09 143	60.55 152	61.72 133	65.92 160	79.28 121	67.16 132	54.91 149	39.79 171	38.10 148	36.08 166	34.64 178	59.15 120	72.86 126	73.36 127	85.10 104	87.84 101

test0.0.03 1		57.35 165	59.89 157	54.38 172	71.37 124	73.45 164	52.71 180	61.03 95	46.11 147	26.33 182	41.73 134	44.08 138	29.72 186	71.43 142	70.90 154	85.10 104	71.56 181

V42		62.86 129	62.97 131	62.74 126	60.84 179	78.99 123	71.46 100	57.13 129	46.85 142	44.28 118	38.87 147	40.73 151	57.63 132	72.60 131	74.14 116	85.09 106	88.63 94

v148		62.00 138	61.19 147	62.96 121	67.46 149	79.49 119	67.87 124	57.66 121	42.30 159	45.02 114	38.20 152	38.89 158	54.77 140	69.83 157	72.60 139	84.96 107	87.01 106

Baseline_NR-MVSNet		59.47 152	60.28 153	58.54 152	66.69 152	73.90 162	61.63 161	62.90 79	49.15 136	26.87 180	35.18 171	37.62 161	48.20 157	69.67 158	73.61 122	84.92 108	82.82 140

FMVSNet5		58.86 156	60.24 154	57.25 160	52.66 197	66.25 185	63.77 150	52.86 163	57.85 99	37.92 150	36.12 165	52.22 121	51.37 149	70.88 145	71.43 149	84.92 108	66.91 189

TranMVSNet+NR-MVSNet		60.38 148	61.30 146	59.30 147	68.34 141	75.57 154	63.38 154	63.78 68	46.74 143	27.73 179	42.56 127	36.84 166	47.66 159	70.36 152	74.59 111	84.91 110	82.46 142

thres600view7		63.77 121	63.14 129	64.51 108	75.49 105	81.61 95	69.59 115	62.95 76	43.96 155	38.90 144	41.09 137	40.24 154	55.25 139	76.24 89	71.54 145	84.89 111	87.30 104

thres200		65.58 106	64.74 121	66.56 97	77.52 90	81.61 95	73.44 88	62.95 76	46.23 146	42.45 129	42.76 123	41.18 145	58.12 125	76.24 89	75.59 101	84.89 111	89.58 82

tfpn200view9		65.90 105	64.96 119	67.00 96	77.70 85	81.58 97	71.71 97	62.94 78	49.16 134	43.40 122	43.24 121	41.34 143	61.42 104	76.24 89	74.63 110	84.84 113	88.52 96

Fast-Effi-MVS+-dtu		63.05 126	64.72 122	61.11 135	71.21 127	76.81 143	70.72 110	43.13 191	52.51 124	35.34 163	46.55 117	46.36 133	61.40 105	71.57 141	71.44 148	84.84 113	87.79 102

EG-PatchMatch MVS		58.73 158	58.03 165	59.55 144	72.32 118	80.49 110	63.44 153	55.55 142	32.49 190	38.31 147	28.87 186	37.22 164	42.84 175	74.30 111	75.70 99	84.84 113	77.14 162

IterMVS-LS		66.08 104	66.56 111	65.51 100	73.67 112	74.88 156	70.89 109	53.55 158	50.42 128	48.32 103	50.59 93	55.66 106	61.83 101	73.93 113	74.42 114	84.82 116	86.01 115
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
LGP-MVS_train		72.02 70	73.18 72	70.67 73	82.13 58	80.26 113	79.58 48	63.04 74	70.09 59	51.98 85	65.06 48	55.62 107	62.49 99	75.97 93	76.32 93	84.80 117	88.93 90

PCF-MVS	70.85 4	75.73 51	76.55 56	74.78 51	83.67 51	88.04 46	81.47 35	70.62 26	69.24 65	57.52 70	60.59 63	69.18 49	70.65 53	77.11 81	77.65 80	84.75 118	94.01 37
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
IS_MVSNet		67.29 99	71.98 76	61.82 132	76.92 93	84.32 81	65.90 140	58.22 112	55.75 110	39.22 142	54.51 80	62.47 68	45.99 167	78.83 68	78.52 72	84.70 119	89.47 84

AdaColmap		76.23 49	73.55 69	79.35 22	89.38 24	85.00 73	79.99 47	73.04 17	76.60 48	71.17 24	55.18 75	57.99 96	77.87 13	76.82 85	76.82 86	84.67 120	86.45 110

thisisatest0530		68.38 90	70.98 84	65.35 101	72.61 117	84.42 77	68.21 123	57.98 115	59.77 91	50.80 92	54.63 78	58.48 90	57.92 127	76.99 83	77.47 81	84.60 121	85.07 121

FMVSNet1		63.48 123	63.07 130	63.97 114	65.31 162	76.37 146	71.77 96	57.90 118	43.32 157	45.66 109	35.06 172	49.43 127	58.57 123	77.49 76	78.22 75	84.59 122	81.60 149

tfpnnormal		58.97 155	56.48 170	61.89 131	71.27 126	76.21 147	66.65 136	61.76 90	32.90 189	36.41 157	27.83 188	29.14 194	50.64 153	73.06 123	73.05 134	84.58 123	83.15 139

Vis-MVSNet		65.53 108	69.83 91	60.52 138	70.80 130	84.59 75	66.37 139	55.47 144	48.40 137	40.62 138	57.67 68	58.43 92	45.37 170	77.49 76	76.24 94	84.47 124	85.99 116
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
ACMP	68.86 7	72.15 69	72.25 74	72.03 64	80.96 65	80.87 107	77.93 56	64.13 65	69.29 63	60.79 59	64.04 50	53.54 117	63.91 89	73.74 117	75.27 104	84.45 125	88.98 89
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
TransMVSNet (Re)		57.83 161	56.90 168	58.91 150	72.26 119	74.69 159	63.57 152	61.42 92	32.30 191	32.65 170	33.97 173	35.96 173	39.17 180	73.84 116	72.84 137	84.37 126	74.69 168

IterMVS		61.87 139	63.55 126	59.90 141	67.29 150	72.20 168	67.34 131	48.56 172	47.48 140	37.86 151	47.07 111	48.27 128	54.08 142	72.12 134	73.71 121	84.30 127	83.99 129
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
tttt0517		67.99 93	70.61 86	64.94 104	71.94 122	83.96 83	67.62 127	57.98 115	59.30 93	49.90 97	54.50 81	57.98 97	57.92 127	76.48 87	77.47 81	84.24 128	84.58 124

UGNet		67.57 96	71.69 79	62.76 125	69.88 132	82.58 90	66.43 137	58.64 110	54.71 118	51.87 86	61.74 56	62.01 72	45.46 169	74.78 104	74.99 105	84.24 128	91.02 65
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
MDTV_nov1_ep13_2view		54.47 174	54.61 173	54.30 173	60.50 180	73.82 163	57.92 172	43.38 188	39.43 173	32.51 171	33.23 174	34.05 181	47.26 161	62.36 181	66.21 176	84.24 128	73.19 176

MDTV_nov1_ep13		65.21 110	67.28 106	62.79 123	70.91 128	81.72 94	69.28 118	49.50 170	58.08 97	43.94 119	50.50 94	56.02 103	58.86 122	70.72 146	73.37 126	84.24 128	80.52 152

ACMM	66.70 10	70.42 77	68.49 98	72.67 61	82.85 52	77.76 135	77.70 58	64.76 62	64.61 79	60.74 60	49.29 96	53.97 115	65.86 81	74.97 101	75.57 102	84.13 132	83.29 135
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
TSAR-MVS + ACMM		81.59 23	85.84 17	76.63 36	89.82 19	86.53 60	86.32 12	66.72 49	85.96 19	65.43 40	88.98 8	82.29 10	67.57 76	82.06 42	81.33 46	83.93 133	93.75 40

UniMVSNet (Re)		60.62 146	62.93 133	57.92 154	67.64 147	77.90 132	61.75 160	61.24 94	49.83 131	29.80 177	42.57 126	40.62 152	43.36 173	70.49 151	73.27 130	83.76 134	85.81 117

DeepPCF-MVS	76.94 1	83.08 16	87.77 7	77.60 32	90.11 16	90.96 16	78.48 51	72.63 19	93.10 3	65.84 39	80.67 20	81.55 15	74.80 27	85.94 11	85.39 8	83.75 135	96.77 11

SCA		63.90 120	66.67 108	60.66 137	73.75 110	71.78 171	59.87 167	43.66 187	61.13 87	45.03 113	51.64 90	59.45 86	57.92 127	70.96 144	70.80 155	83.71 136	80.92 151

pmmvs4		63.14 125	62.46 137	63.94 115	66.03 158	76.40 145	66.82 134	57.60 122	56.74 101	50.26 96	40.81 140	37.51 162	59.26 119	71.75 139	71.48 147	83.68 137	82.53 141

MSDG		65.57 107	61.57 144	70.24 75	82.02 59	76.47 144	74.46 84	68.73 38	56.52 103	50.33 95	38.47 149	41.10 147	62.42 100	72.12 134	72.94 135	83.47 138	73.37 175

DCV-MVSNet		69.13 84	69.07 93	69.21 80	77.65 87	77.52 137	74.68 76	57.85 119	54.92 115	55.34 80	55.74 72	55.56 108	66.35 79	75.05 100	76.56 89	83.35 139	88.13 100

EPP-MVSNet		67.58 95	71.10 83	63.48 118	75.71 103	83.35 86	66.85 133	57.83 120	53.02 122	41.15 134	55.82 71	67.89 52	56.01 136	74.40 107	72.92 136	83.33 140	90.30 75

DU-MVS		60.87 145	61.82 142	59.76 143	66.69 152	75.87 148	64.07 145	61.96 84	49.31 132	31.17 173	42.76 123	36.95 165	51.37 149	69.67 158	73.20 133	83.30 141	84.95 122

tpm cat1		67.47 97	67.05 107	67.98 89	76.63 95	81.51 99	74.49 83	47.65 176	61.18 86	61.12 55	42.51 128	53.02 120	64.74 87	70.11 155	71.50 146	83.22 142	89.49 83

MVS_111021_LR		74.26 58	75.95 58	72.27 63	79.43 74	85.04 72	72.71 89	65.27 60	70.92 56	63.58 45	69.32 36	60.31 83	69.43 63	77.01 82	77.15 83	83.22 142	91.93 57

Anonymous202405211			66.35 113		78.00 83	84.41 78	74.85 75	63.18 73	51.00 126		31.37 180	53.73 116	69.67 60	76.28 88	76.84 85	83.21 144	90.85 67

CPTT-MVS		75.43 52	77.13 51	73.44 55	81.43 62	82.55 91	80.96 42	64.35 63	77.95 45	61.39 53	69.20 37	70.94 44	69.38 65	73.89 114	73.32 128	83.14 145	92.06 55

pm-mvs1		59.21 154	59.58 159	58.77 151	67.97 144	77.07 142	64.12 143	57.20 127	34.73 186	36.86 153	35.34 169	40.54 153	43.34 174	74.32 110	73.30 129	83.13 146	81.77 148

MVS-HIRNet		53.86 176	53.02 178	54.85 169	60.30 181	72.36 167	44.63 196	42.20 194	39.45 172	43.47 121	21.66 199	34.00 182	55.47 137	65.42 171	67.16 171	83.02 147	71.08 183

Anonymous20231211		68.44 88	66.37 112	70.86 70	77.58 88	83.49 85	75.15 74	61.89 86	52.54 123	58.50 65	28.89 185	56.78 100	69.29 66	74.96 103	76.61 87	82.73 148	91.36 62

CDS-MVSNet		64.22 117	65.89 117	62.28 130	70.05 131	80.59 109	69.91 114	57.98 115	43.53 156	46.58 107	48.22 100	50.76 124	46.45 164	75.68 95	76.08 95	82.70 149	86.34 112
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
TSAR-MVS + COLMAP		73.09 62	76.86 52	68.71 84	74.97 108	82.49 92	74.51 82	61.83 87	83.16 25	49.31 99	82.22 18	51.62 122	68.94 68	78.76 69	75.52 103	82.67 150	84.23 128

v7n		57.04 166	56.64 169	57.52 158	62.85 173	74.75 158	61.76 159	51.80 166	35.58 185	36.02 160	32.33 177	33.61 184	50.16 154	67.73 165	70.34 160	82.51 151	82.12 145

Vis-MVSNet (Re-imp)		62.25 133	68.74 96	54.68 170	73.70 111	78.74 124	56.51 175	57.49 124	55.22 112	26.86 181	54.56 79	61.35 75	31.06 184	73.10 122	74.90 106	82.49 152	83.31 134

Effi-MVS+-dtu		64.58 114	64.08 124	65.16 102	73.04 116	75.17 155	70.68 111	56.23 134	54.12 120	44.71 116	47.42 105	51.10 123	63.82 90	68.08 164	66.32 175	82.47 153	86.38 111

TAPA-MVS	67.10 9	71.45 73	73.47 71	69.10 82	77.04 92	80.78 108	73.81 87	62.10 83	80.80 35	51.28 88	60.91 60	63.80 66	67.98 71	74.59 105	72.42 140	82.37 154	80.97 150
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
OMC-MVS		74.03 59	75.82 59	71.95 65	79.56 72	80.98 105	75.35 73	63.21 72	84.48 22	61.83 51	61.54 58	66.89 55	69.41 64	76.60 86	74.07 118	82.34 155	86.15 113

dps		64.08 118	63.22 128	65.08 103	75.27 106	79.65 117	66.68 135	46.63 180	56.94 100	55.67 76	43.96 120	43.63 140	64.00 88	69.50 160	69.82 161	82.25 156	79.02 157

Anonymous20231206		52.23 179	52.80 181	51.56 177	64.70 165	69.41 177	51.01 182	58.60 111	36.63 178	22.44 188	21.80 198	31.42 189	30.52 185	66.79 167	67.83 167	82.10 157	75.73 164

PMMVS		70.37 80	75.06 62	64.90 105	71.46 123	81.88 93	64.10 144	55.64 140	71.31 55	46.69 106	70.69 35	58.56 88	69.53 61	79.03 66	75.63 100	81.96 158	88.32 98

TAMVS		58.86 156	60.91 149	56.47 165	62.38 176	77.57 136	58.97 170	52.98 161	38.76 174	36.17 158	42.26 131	47.94 130	46.45 164	70.23 154	70.79 156	81.86 159	78.82 158

UniMVSNet_ETH3D		57.83 161	56.46 171	59.43 146	63.24 171	73.22 165	67.70 125	55.58 141	36.17 181	36.84 154	32.64 175	35.14 177	51.50 148	65.81 169	69.81 162	81.73 160	82.44 144

PLC	64.00 12	68.54 87	66.66 109	70.74 72	80.28 71	74.88 156	72.64 90	63.70 69	69.26 64	55.71 75	47.24 109	55.31 109	70.42 54	72.05 136	70.67 157	81.66 161	77.19 161
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
ADS-MVSNet		58.40 160	59.16 161	57.52 158	65.80 161	74.57 160	60.26 164	40.17 200	50.51 127	38.01 149	40.11 144	44.72 137	59.36 118	64.91 173	66.55 173	81.53 162	72.72 178

PEN-MVS		51.04 180	52.94 179	48.82 182	61.45 178	66.00 186	48.68 185	57.20 127	36.87 177	15.36 196	36.98 159	32.72 185	28.77 190	57.63 188	66.37 174	81.44 163	74.00 171

pmmvs-eth3d		55.20 167	53.95 176	56.65 163	57.34 191	67.77 181	57.54 173	53.74 157	40.93 167	41.09 135	31.19 181	29.10 195	49.07 155	65.54 170	67.28 169	81.14 164	75.81 163

WR-MVS_H		49.62 186	52.63 182	46.11 191	58.80 186	67.58 182	46.14 193	54.94 147	36.51 179	13.63 201	36.75 162	35.67 175	22.10 196	56.43 192	62.76 186	81.06 165	72.73 177

thisisatest0515		59.37 153	60.68 151	57.84 156	64.39 166	75.65 153	58.56 171	53.86 156	41.55 164	42.12 130	40.40 142	39.59 155	47.09 162	71.69 140	73.79 120	81.02 166	82.08 146

CNLPA		71.37 75	70.27 89	72.66 62	80.79 68	81.33 101	71.07 107	65.75 55	82.36 28	64.80 42	42.46 129	56.49 101	72.70 40	73.00 125	70.52 159	80.84 167	85.76 118

UA-Net		64.62 113	68.23 101	60.42 139	77.53 89	81.38 100	60.08 166	57.47 125	47.01 141	44.75 115	60.68 61	71.32 43	41.84 177	73.27 120	72.25 142	80.83 168	71.68 180

DTE-MVSNet		49.82 185	51.92 185	47.37 187	61.75 177	64.38 191	45.89 194	57.33 126	36.11 182	12.79 203	36.87 160	31.93 188	25.73 193	58.01 186	65.22 179	80.75 169	70.93 184

CP-MVSNet		50.57 182	52.60 183	48.21 185	58.77 187	65.82 187	48.17 186	56.29 133	37.41 176	16.59 193	37.14 157	31.95 187	29.21 187	56.60 191	63.71 183	80.22 170	75.56 165

PS-CasMVS		50.17 183	52.02 184	48.02 186	58.60 188	65.54 188	48.04 187	56.19 135	36.42 180	16.42 195	35.68 168	31.33 190	28.85 189	56.42 193	63.54 184	80.01 171	75.18 166

USDC		59.69 151	60.03 156	59.28 148	64.04 167	71.84 169	63.15 156	55.36 145	54.90 116	35.02 164	48.34 99	29.79 193	58.16 124	70.60 148	71.33 152	79.99 172	73.42 174

WR-MVS		51.02 181	54.56 174	46.90 188	63.84 168	69.23 178	44.78 195	56.38 132	38.19 175	14.19 198	37.38 155	36.82 167	22.39 195	60.14 184	66.20 177	79.81 173	73.95 172

ACMH	59.42 14	61.59 141	59.22 160	64.36 111	78.92 78	78.26 129	67.65 126	67.48 44	39.81 170	30.98 175	38.25 151	34.59 179	61.37 106	70.55 150	73.47 124	79.74 174	79.59 154
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
ACMH+	60.36 13	61.16 142	58.38 162	64.42 110	77.37 91	74.35 161	68.45 121	62.81 80	45.86 148	38.48 146	35.71 167	37.35 163	59.81 114	67.24 166	69.80 163	79.58 175	78.32 159

anonymousdsp		54.99 170	57.24 167	52.36 175	53.82 195	71.75 172	51.49 181	48.14 173	33.74 187	33.66 168	38.34 150	36.13 171	47.54 160	64.53 177	70.60 158	79.53 176	85.59 120

pmmvs6		54.20 175	53.54 177	54.97 168	63.22 172	72.98 166	60.17 165	52.32 165	26.77 200	34.30 166	23.29 195	36.23 170	40.33 179	68.77 162	68.76 165	79.47 177	78.00 160

test20.03		47.23 192	48.69 192	45.53 192	63.28 170	64.39 190	41.01 199	56.93 130	29.16 195	15.21 197	23.90 192	30.76 192	17.51 202	64.63 176	65.26 178	79.21 178	62.71 197

CR-MVSNet		62.31 131	64.75 120	59.47 145	68.63 140	71.29 173	67.53 128	43.18 189	55.83 108	41.40 131	41.04 138	55.85 104	57.29 133	72.76 128	73.27 130	78.77 179	83.23 137

testgi		48.51 189	50.53 187	46.16 190	64.78 163	67.15 184	41.54 198	54.81 151	29.12 196	17.03 192	32.07 178	31.98 186	20.15 199	65.26 172	67.00 172	78.67 180	61.10 200

RPMNet		58.63 159	62.80 135	53.76 174	67.59 148	71.29 173	54.60 178	38.13 201	55.83 108	35.70 161	41.58 135	53.04 119	47.89 158	66.10 168	67.38 168	78.65 181	84.40 126

LS3D		64.54 116	62.14 140	67.34 95	80.85 66	75.79 150	69.99 112	65.87 54	60.77 88	44.35 117	42.43 130	45.95 135	65.01 83	69.88 156	68.69 166	77.97 182	71.43 182

CHOSEN 280x420		62.23 135	66.57 110	57.17 161	59.88 182	68.92 179	61.20 163	42.28 193	54.17 119	39.57 139	47.78 103	64.97 61	62.68 96	73.85 115	69.52 164	77.43 183	86.75 107

PatchMatch-RL		62.22 136	60.69 150	64.01 113	68.74 139	75.75 151	59.27 168	60.35 104	56.09 107	53.80 82	47.06 112	36.45 168	64.80 86	68.22 163	67.22 170	77.10 184	74.02 170

CVMVSNet		54.92 172	58.16 163	51.13 179	62.61 175	68.44 180	55.45 177	52.38 164	42.28 160	21.45 189	47.10 110	46.10 134	37.96 181	64.42 178	63.81 182	76.92 185	75.01 167

N_pmnet		47.67 190	47.00 194	48.45 184	54.72 194	62.78 194	46.95 190	51.25 167	36.01 183	26.09 183	26.59 191	25.93 201	35.50 183	55.67 195	59.01 191	76.22 186	63.04 195

COLMAP_ROB	51.17 15	55.13 168	52.90 180	57.73 157	73.47 115	67.21 183	62.13 158	55.82 137	47.83 138	34.39 165	31.60 179	34.24 180	44.90 171	63.88 180	62.52 187	75.67 187	63.02 196
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
MIMVSNet1		40.84 197	43.46 195	37.79 198	32.14 206	58.92 199	39.24 201	50.83 168	27.00 199	11.29 205	16.76 205	26.53 199	17.75 201	57.14 190	61.12 190	75.46 188	56.78 201

CMPMVS	43.63 17	57.67 164	55.43 172	60.28 140	72.01 120	79.00 122	62.77 157	53.23 160	41.77 162	45.42 110	30.74 182	39.03 156	53.01 144	64.81 175	64.65 181	75.26 189	68.03 187
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
TinyColmap		52.66 178	50.09 189	55.65 166	59.72 183	64.02 193	57.15 174	52.96 162	40.28 168	32.51 171	32.42 176	20.97 204	56.65 135	63.95 179	65.15 180	74.91 190	63.87 194

MDA-MVSNet-bldmvs		44.15 194	42.27 199	46.34 189	38.34 205	62.31 195	46.28 191	55.74 139	29.83 194	20.98 190	27.11 190	16.45 209	41.98 176	41.11 203	57.47 194	74.72 191	61.65 199

new-patchmatchnet		42.21 195	42.97 196	41.33 195	53.05 196	59.89 197	39.38 200	49.61 169	28.26 198	12.10 204	22.17 197	21.54 203	19.22 200	50.96 198	56.04 196	74.61 192	61.92 198

TDRefinement		52.70 177	51.02 186	54.66 171	57.41 190	65.06 189	61.47 162	54.94 147	44.03 154	33.93 167	30.13 184	27.57 196	46.17 166	61.86 182	62.48 188	74.01 193	66.06 190

PM-MVS		50.11 184	50.38 188	49.80 180	47.23 203	62.08 196	50.91 183	44.84 185	41.90 161	36.10 159	35.22 170	26.05 200	46.83 163	57.64 187	55.42 198	72.90 194	74.32 169

pmmvs3		41.86 196	42.29 198	41.36 194	39.80 204	52.66 203	38.93 202	35.85 205	23.40 204	20.22 191	19.30 200	20.84 205	40.56 178	55.98 194	58.79 192	72.80 195	65.03 192

SixPastTwentyTwo		49.11 188	49.22 191	48.99 181	58.54 189	64.14 192	47.18 189	47.75 175	31.15 193	24.42 184	41.01 139	26.55 198	44.04 172	54.76 196	58.70 193	71.99 196	68.21 185

IterMVS-SCA-FT		60.21 149	62.97 131	57.00 162	66.64 154	71.84 169	67.53 128	46.93 179	47.56 139	36.77 156	46.85 115	48.21 129	52.51 145	70.36 152	72.40 141	71.63 197	83.53 133

EU-MVSNet		44.84 193	47.85 193	41.32 196	49.26 200	56.59 201	43.07 197	47.64 177	33.03 188	13.82 199	36.78 161	30.99 191	24.37 194	53.80 197	55.57 197	69.78 198	68.21 185

RPSCF		55.07 169	58.06 164	51.57 176	48.87 201	58.95 198	53.68 179	41.26 198	62.42 82	45.88 108	54.38 82	54.26 113	53.75 143	57.15 189	53.53 199	66.01 199	65.75 191

LTVRE_ROB	47.26 16	49.41 187	49.91 190	48.82 182	64.76 164	69.79 176	49.05 184	47.12 178	20.36 207	16.52 194	36.65 163	26.96 197	50.76 152	60.47 183	63.16 185	64.73 200	72.00 179
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
FC-MVSNet-test		47.24 191	54.37 175	38.93 197	59.49 184	58.25 200	34.48 204	53.36 159	45.66 149	6.66 208	50.62 92	42.02 141	16.62 203	58.39 185	61.21 189	62.99 201	64.40 193

ambc			42.30 197		50.36 199	49.51 204	35.47 203		32.04 192	23.53 185	17.36 202	8.95 211	29.06 188	64.88 174	56.26 195	61.29 202	67.12 188

FPMVS		39.11 198	36.39 200	42.28 193	55.97 192	45.94 205	46.23 192	41.57 195	35.73 184	22.61 186	23.46 194	19.82 206	28.32 191	43.57 200	40.67 202	58.96 203	45.54 203

new_pmnet		33.19 199	35.52 201	30.47 200	27.55 210	45.31 206	29.29 206	30.92 206	29.00 197	9.88 207	18.77 201	17.64 208	26.77 192	44.07 199	45.98 201	58.41 204	47.87 202

PMVS	27.44 18	32.08 200	29.07 202	35.60 199	48.33 202	24.79 208	26.97 207	41.34 197	20.45 206	22.50 187	17.11 204	18.64 207	20.44 198	41.99 202	38.06 203	54.02 205	42.44 204
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
PMMVS2		20.45 202	22.31 204	18.27 205	20.52 211	26.73 207	14.85 210	28.43 208	13.69 208	0.79 213	10.35 206	9.10 210	3.83 209	27.64 205	32.87 204	41.17 206	35.81 205

Gipuma		24.91 201	24.61 203	25.26 202	31.47 207	21.59 209	18.06 208	37.53 202	25.43 202	10.03 206	4.18 210	4.25 213	14.85 204	43.20 201	47.03 200	39.62 207	26.55 208
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
tmp_tt				16.09 206	13.07 212	8.12 214	13.61 211	2.08 210	55.09 113	30.10 176	40.26 143	22.83 202	5.35 208	29.91 204	25.25 206	32.33 208

E-PMN		15.08 203	11.65 206	19.08 203	28.73 208	12.31 212	6.95 213	36.87 204	10.71 210	3.63 211	5.13 207	2.22 216	13.81 206	11.34 208	18.50 207	24.49 209	21.32 209

EMVS		14.40 204	10.71 207	18.70 204	28.15 209	12.09 213	7.06 212	36.89 203	11.00 209	3.56 212	4.95 208	2.27 215	13.91 205	10.13 209	16.06 208	22.63 210	18.51 210

DeepMVS_CX						19.81 211	17.01 209	10.02 209	23.61 203	5.85 209	17.21 203	8.03 212	21.13 197	22.60 206		21.42 211	30.01 206

MVE	15.98 19	14.37 205	16.36 205	12.04 207	7.72 213	20.24 210	5.90 214	29.05 207	8.28 211	3.92 210	4.72 209	2.42 214	9.57 207	18.89 207	31.46 205	16.07 212	28.53 207
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
testmvs		0.05 206	0.08 208	0.01 208	0.00 215	0.01 215	0.03 216	0.01 212	0.05 212	0.00 215	0.14 212	0.01 217	0.03 212	0.05 210	0.05 209	0.01 213	0.24 212

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

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

test123		0.05 206	0.08 208	0.01 208	0.00 215	0.01 215	0.01 217	0.00 213	0.05 212	0.00 215	0.16 211	0.00 218	0.04 210	0.02 211	0.05 209	0.00 214	0.26 211

SR-MVS					86.33 44			67.54 43				80.78 16

our_test_3					63.32 169	71.07 175	55.90 176

test_part1																	97.29 4

MTAPA										78.32 8		79.42 21

MTMP										76.04 13		76.65 26

Patchmatch-RL test							2.17 215

mPP-MVS					86.96 39							70.61 46

NP-MVS									81.60 33

Patchmtry						78.06 131	67.53 128	43.18 189		41.40 131

High-resolution multi-view benchmark