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		94.10 5	94.77 5	93.31 6	98.31 1	98.34 3	95.43 4	92.54 3	94.41 13	83.05 27	91.38 15	90.97 4	92.24 9	95.05 5	94.02 5	98.31 1	99.20 7

gg-mvs-nofinetune		72.10 141	74.79 124	68.97 178	83.31 120	95.22 55	85.66 73	48.77 222	35.68 222	22.17 231	30.49 215	77.73 52	76.37 102	94.30 10	93.03 11	97.55 2	97.05 47

CHOSEN 1792x2688		80.23 78	79.16 88	81.48 70	91.97 51	96.56 31	86.18 69	75.40 95	76.17 87	61.32 93	37.43 206	61.08 108	76.52 99	92.35 25	91.64 27	97.46 3	98.86 13

CANet		89.98 25	90.42 29	89.47 28	94.13 42	98.05 8	91.76 34	83.27 43	90.87 31	81.90 34	72.32 51	84.82 26	88.42 28	94.52 9	93.78 8	97.34 4	98.58 20

QAPM		87.06 43	86.46 47	87.75 37	96.63 16	97.09 22	91.71 35	82.62 49	80.58 69	71.28 69	66.04 70	84.24 27	87.01 37	89.93 48	89.91 42	97.26 5	97.44 35

TSAR-MVS + GP.		91.29 18	93.11 16	89.18 29	87.81 85	96.21 41	92.51 28	83.83 40	94.24 14	83.77 21	91.87 14	89.62 8	90.07 15	90.40 43	90.31 39	97.09 6	99.10 9

HSP-MVS		94.69 2	95.39 2	93.88 4	96.78 14	98.11 5	94.75 7	90.91 9	96.89 2	89.12 10	96.98 1	89.47 9	94.76 1	95.24 2	93.29 10	96.98 7	97.73 30

MVS_0304		88.43 36	89.46 31	87.21 41	91.85 54	97.60 14	92.62 26	81.10 58	87.16 50	73.80 56	72.19 53	83.36 34	87.03 36	94.64 7	93.67 9	96.88 8	97.64 32

MVS_Test		84.60 57	85.13 55	83.99 59	88.17 80	95.27 54	88.21 54	73.15 107	84.31 59	70.55 72	68.67 63	68.78 88	86.99 38	91.71 32	91.90 23	96.84 9	95.27 84

MVSTER		87.68 41	89.12 32	86.01 52	88.11 82	90.05 106	89.28 48	77.05 80	91.37 26	79.97 37	76.70 42	85.25 24	84.89 50	93.53 15	91.41 30	96.73 10	95.55 79

CANet_DTU		83.33 59	86.59 45	79.53 87	88.88 69	94.87 57	86.63 62	68.85 144	85.45 56	50.54 155	77.86 40	69.94 84	85.62 44	92.63 23	90.88 36	96.63 11	94.46 92

GG-mvs-BLEND		62.08 205	88.31 37	31.46 230	0.16 238	98.10 6	91.57 36	0.09 236	85.07 58	0.21 241	73.90 49	83.74 33	0.19 238	88.98 58	89.39 51	96.58 12	99.02 11

gm-plane-assit		64.86 191	68.15 172	61.02 207	76.44 164	68.29 217	41.60 226	53.37 215	34.68 224	26.19 228	33.22 212	57.09 125	71.97 120	95.12 4	93.97 6	96.54 13	94.66 89

MSLP-MVS++		90.33 24	88.82 34	92.10 13	96.52 21	95.93 42	94.35 12	86.26 28	88.37 47	89.24 8	75.94 44	82.60 35	89.71 18	89.45 54	92.17 17	96.51 14	97.24 42

PVSNet_BlendedMVS		86.98 44	87.05 42	86.90 43	93.03 45	96.98 25	86.57 63	81.82 54	89.78 39	82.78 29	71.54 54	66.07 94	80.73 78	93.46 16	91.97 21	96.45 15	99.53 1

PVSNet_Blended		86.98 44	87.05 42	86.90 43	93.03 45	96.98 25	86.57 63	81.82 54	89.78 39	82.78 29	71.54 54	66.07 94	80.73 78	93.46 16	91.97 21	96.45 15	99.53 1

MVS_111021_HR		87.82 39	88.84 33	86.62 48	94.42 40	97.36 18	88.21 54	83.26 44	83.42 60	72.52 64	82.63 34	76.93 57	84.95 49	91.93 29	91.15 33	96.39 17	98.49 21

3Dnovator	80.58 8	88.20 37	86.53 46	90.15 21	96.86 13	96.46 35	91.97 32	83.06 46	85.16 57	83.66 23	62.28 91	82.15 37	88.98 23	90.99 38	92.65 14	96.38 18	96.03 72

ESAPD		95.11 1	95.65 1	94.48 1	97.96 3	98.62 1	96.45 1	92.82 2	96.24 3	90.25 5	96.16 2	93.09 1	93.32 3	93.93 13	92.02 19	96.07 19	99.50 3

PatchT		72.66 134	76.58 108	68.09 181	79.02 141	86.09 151	59.81 209	51.78 218	72.00 105	51.09 141	46.84 151	66.70 92	70.77 138	86.48 91	84.03 105	96.07 19	93.92 106

FMVSNet3		81.93 67	81.98 73	81.88 67	79.49 140	87.02 129	88.15 56	72.57 110	83.02 62	72.63 61	56.55 111	73.48 71	82.34 66	91.49 35	91.20 32	96.07 19	91.13 135

test-LLR		79.52 83	83.42 62	74.97 114	81.79 125	91.26 94	76.17 167	70.57 133	77.71 79	52.14 133	66.26 68	77.47 53	73.10 115	87.02 83	87.16 65	96.05 22	97.02 48

TESTMET0.1,1		79.15 85	83.42 62	74.18 124	79.81 138	91.26 94	76.17 167	67.83 155	77.71 79	52.14 133	66.26 68	77.47 53	73.10 115	87.02 83	87.16 65	96.05 22	97.02 48

CNVR-MVS		94.53 3	94.85 4	94.15 3	98.03 2	98.59 2	95.56 3	92.91 1	94.86 8	88.46 11	91.32 17	90.83 5	94.03 2	95.20 3	94.16 4	95.89 24	99.01 12

IB-MVS	74.10 12	78.52 93	78.51 92	78.52 94	90.15 61	95.39 52	71.95 186	77.53 75	74.95 91	77.25 45	58.93 101	55.92 128	58.37 188	79.01 169	87.89 62	95.88 25	97.47 34
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
MS-PatchMatch		77.47 104	76.48 109	78.63 93	89.89 62	90.42 100	85.42 74	69.53 139	70.79 109	60.43 102	50.05 140	70.62 83	70.66 141	86.71 87	82.54 123	95.86 26	84.23 185

DELS-MVS		87.75 40	86.92 44	88.71 31	94.69 37	97.34 21	92.78 24	84.50 37	77.87 77	81.94 33	67.17 65	75.49 62	82.84 63	95.38 1	95.93 1	95.55 27	99.27 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
APDe-MVS		94.31 4	94.30 7	94.33 2	97.57 7	98.06 7	95.79 2	91.98 5	95.50 6	92.19 1	95.25 3	87.97 14	92.93 4	93.01 20	91.02 35	95.52 28	99.29 5

OpenMVS	77.91 11	85.09 52	83.42 62	87.03 42	96.12 26	96.55 33	89.36 47	81.59 56	79.19 72	75.20 53	55.84 119	79.04 49	84.45 55	88.47 66	89.35 52	95.48 29	95.48 80

NCCC		93.59 7	94.00 9	93.10 7	97.90 5	97.93 9	95.40 5	92.39 4	94.47 12	84.94 18	91.21 18	89.32 10	92.53 6	93.90 14	92.98 12	95.44 30	98.22 24

DWT-MVSNet_training		82.66 61	83.34 65	81.87 68	88.71 70	92.63 77	82.07 100	72.21 116	86.37 53	72.64 59	64.51 79	71.44 78	80.35 81	84.43 110	87.73 63	95.27 31	96.25 69

HyFIR lowres test		78.08 100	76.81 104	79.56 86	90.77 60	94.64 58	82.97 93	69.85 137	69.81 116	59.53 105	33.52 211	64.66 98	78.97 88	88.77 62	88.38 58	95.27 31	97.86 28

test-mter		77.90 101	82.44 71	72.60 149	78.52 142	90.24 102	73.85 179	65.31 173	76.37 86	51.29 137	65.58 73	75.94 61	71.36 128	85.98 98	86.26 78	95.26 33	96.71 64

CLD-MVS		85.43 51	84.24 59	86.83 45	87.69 88	93.16 72	90.01 44	82.72 48	87.17 49	79.28 39	71.43 57	65.81 96	86.02 43	87.33 81	86.96 70	95.25 34	97.83 29
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
abl_6				89.54 27	95.55 33	97.59 15	89.01 50	85.00 34	94.67 11	83.04 28	84.70 31	91.47 3	89.46 20			95.20 35	98.63 17

EPMVS		77.16 110	79.08 89	74.92 115	86.73 94	91.98 88	78.62 139	55.44 209	79.43 70	56.59 115	61.24 95	70.73 82	76.97 96	80.59 145	81.43 151	95.15 36	88.17 169

PHI-MVS		89.88 27	92.75 18	86.52 50	94.97 35	97.57 16	89.99 45	84.56 36	92.52 23	69.72 75	90.35 20	87.11 15	84.89 50	91.82 30	92.37 15	95.02 37	97.51 33

DeepC-MVS_fast	86.59 2	91.69 16	91.39 22	92.05 14	97.43 8	96.92 27	94.05 15	90.23 12	93.31 20	83.19 25	77.91 39	84.23 28	92.42 7	94.62 8	94.83 2	95.00 38	97.88 27
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.		93.07 11	93.53 12	92.53 9	94.23 41	97.54 17	94.75 7	89.87 13	95.26 7	89.20 9	93.16 9	88.19 13	92.15 10	91.79 31	89.65 48	94.99 39	99.16 8

GBi-Net		80.72 73	80.49 80	81.00 79	78.18 144	86.19 147	86.73 59	72.57 110	83.02 62	72.63 61	56.55 111	73.48 71	80.99 74	86.57 88	86.83 72	94.89 40	90.77 137

test1		80.72 73	80.49 80	81.00 79	78.18 144	86.19 147	86.73 59	72.57 110	83.02 62	72.63 61	56.55 111	73.48 71	80.99 74	86.57 88	86.83 72	94.89 40	90.77 137

FMVSNet2		79.24 84	78.14 97	80.53 83	78.18 144	86.19 147	86.73 59	71.91 120	72.97 101	70.48 73	50.63 138	66.56 93	80.99 74	90.10 45	89.77 46	94.89 40	90.77 137

APD-MVS		93.47 8	93.44 13	93.50 5	97.06 10	97.09 22	95.27 6	91.47 6	95.71 5	89.57 7	93.66 6	86.28 19	92.81 5	92.06 27	90.70 37	94.83 43	98.60 18
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
MIMVSNet		68.66 178	69.43 159	67.76 182	64.92 216	84.68 166	74.16 177	54.10 214	60.85 144	51.27 138	39.47 199	49.48 146	67.48 158	84.86 108	85.57 87	94.63 44	81.10 200

SteuartSystems-ACMMP		92.31 12	93.31 14	91.15 18	96.88 12	97.36 18	93.95 16	89.44 15	92.62 22	83.20 24	94.34 5	85.55 21	88.95 24	93.07 19	91.90 23	94.51 45	98.30 23
Skip Steuart: Steuart Systems R&D Blog.
diffmvs		83.81 58	84.78 56	82.69 64	86.06 109	94.03 60	86.46 66	72.43 114	85.71 55	75.29 52	65.48 75	79.49 48	81.39 68	85.55 101	86.98 68	94.48 46	96.20 70

HPM-MVS++		94.04 6	94.96 3	92.96 8	97.93 4	97.71 13	94.65 9	91.01 8	95.91 4	87.43 13	93.52 8	92.63 2	92.29 8	94.22 12	92.34 16	94.47 47	98.37 22

Vis-MVSNet		77.24 107	79.99 86	74.02 128	84.62 116	93.92 61	80.33 109	72.55 113	62.58 140	55.25 118	64.45 80	69.49 86	57.00 190	88.78 61	88.21 60	94.36 48	92.54 125
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
thresconf0.02		78.87 88	80.50 79	76.96 105	87.88 84	91.71 92	82.90 97	78.51 64	67.91 119	50.85 148	74.56 45	69.93 85	67.32 160	86.86 86	85.65 86	94.32 49	86.89 176

CDPH-MVS		88.76 32	90.43 28	86.81 46	96.04 27	96.53 34	92.95 22	85.95 30	90.36 33	67.93 80	85.80 29	80.69 44	83.82 58	90.81 40	91.85 26	94.18 50	96.99 51

DI_MVS_plusplus_trai		83.32 60	82.53 70	84.25 58	86.26 105	93.66 66	90.23 43	77.16 79	77.05 84	74.06 55	53.74 128	74.33 65	83.61 60	91.40 36	89.82 44	94.17 51	97.73 30

tpmrst		76.27 115	77.65 101	74.66 117	86.13 107	89.53 114	79.31 131	54.91 210	77.19 83	56.27 116	55.87 118	64.58 99	77.25 93	80.85 143	80.21 171	94.07 52	95.32 82

FMVSNet5		72.83 133	73.89 131	71.59 165	67.42 211	76.28 207	75.88 171	63.74 183	77.27 82	54.59 121	53.32 130	71.48 77	73.85 111	81.95 134	81.69 137	94.06 53	75.20 211

zzz-MVS		91.59 17	91.12 23	92.13 12	96.76 15	96.68 30	93.39 19	88.00 20	93.63 16	90.76 4	83.97 32	85.33 23	89.89 16	91.60 33	89.65 48	94.00 54	96.97 52

train_agg		91.99 15	93.71 11	89.98 23	96.42 23	97.03 24	94.31 13	89.05 18	93.33 17	77.75 41	95.06 4	88.27 12	88.38 30	92.02 28	91.41 30	94.00 54	98.84 15

SMA-MVS		93.14 10	93.96 10	92.17 11	97.64 6	97.82 12	94.28 14	90.32 11	94.72 10	85.70 17	87.64 25	90.68 6	91.15 13	94.28 11	93.86 7	93.97 56	98.72 16

tfpnview11		74.85 118	75.06 121	74.61 118	86.58 97	89.54 113	79.98 111	75.81 89	64.95 132	47.47 172	64.85 76	54.72 129	63.86 170	84.54 109	82.20 130	93.97 56	84.64 181

EPNet		89.30 30	90.89 24	87.44 39	95.67 31	96.81 28	91.13 37	83.12 45	91.14 28	76.31 51	87.60 26	80.40 46	84.45 55	92.13 26	91.12 34	93.96 58	97.01 50
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
tfpn		77.45 105	76.23 113	78.87 91	87.15 92	91.90 90	82.17 99	76.59 82	62.98 139	56.93 113	53.08 132	57.31 124	76.41 101	87.26 82	85.20 89	93.95 59	95.89 73

PatchmatchNet		76.85 112	80.03 85	73.15 145	84.08 118	91.04 97	77.76 152	55.85 208	79.43 70	52.74 129	62.08 92	76.02 60	74.56 107	79.92 150	81.41 152	93.92 60	90.29 142
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
thres100view900		79.83 80	77.79 100	82.21 65	88.42 74	93.54 68	87.07 58	81.11 57	70.15 110	61.01 96	56.65 108	51.22 136	81.78 67	89.77 51	85.95 82	93.84 61	97.26 41

PVSNet_Blended_VisFu		82.55 62	83.70 61	81.21 74	89.66 63	95.15 56	82.41 98	77.36 77	72.53 104	73.64 57	61.15 96	77.19 55	70.35 148	91.31 37	89.72 47	93.84 61	98.85 14

conf0.002		80.80 72	80.30 82	81.38 72	88.59 71	93.19 71	85.12 79	78.10 65	70.15 110	61.55 91	63.30 86	62.66 104	81.11 69	88.74 63	86.94 71	93.79 63	97.15 44

view800		77.22 108	75.35 119	79.41 90	87.42 90	92.21 87	82.94 95	77.19 78	63.67 137	57.78 111	53.68 129	50.19 144	77.32 92	87.70 78	83.84 108	93.79 63	96.19 71

conf0.01		80.10 79	79.04 90	81.34 73	88.56 72	93.09 73	85.12 79	78.08 66	70.15 110	61.43 92	60.90 97	58.54 118	81.11 69	88.66 64	84.80 93	93.74 65	97.14 45

ACMMP_Plus		92.16 13	92.91 17	91.28 17	96.95 11	97.36 18	93.66 17	89.23 17	93.33 17	83.71 22	90.53 19	86.84 16	90.39 14	93.30 18	91.56 28	93.74 65	97.43 37

3Dnovator+	81.14 5	88.59 33	87.49 40	89.88 25	95.83 29	96.45 37	91.94 33	82.41 52	87.09 51	85.94 16	62.80 88	85.37 22	89.46 20	91.51 34	91.89 25	93.72 67	97.30 40

EPNet_dtu		78.49 94	81.96 74	74.45 121	92.57 49	88.74 118	82.98 92	78.83 61	83.28 61	44.64 193	77.40 41	67.73 90	53.98 198	85.44 103	84.91 91	93.71 68	86.22 178
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
PGM-MVS		89.97 26	90.64 27	89.18 29	96.53 20	95.90 46	93.06 21	82.48 51	90.04 36	80.37 36	92.75 11	80.96 43	88.93 25	89.88 49	89.08 54	93.69 69	95.86 74

OPM-MVS		81.34 68	78.18 96	85.02 55	91.27 58	91.78 91	90.66 40	83.62 41	62.39 141	65.91 81	63.35 85	64.33 101	85.03 48	87.77 77	85.88 84	93.66 70	91.75 133

HFP-MVS		92.02 14	92.13 19	91.89 15	97.16 9	96.46 35	93.57 18	87.60 21	93.79 15	88.17 12	93.15 10	83.94 32	91.19 12	90.81 40	89.83 43	93.66 70	96.94 54

MVS_111021_LR		87.58 42	88.67 36	86.31 51	92.58 48	95.89 48	86.20 68	82.49 50	89.08 44	77.47 43	86.20 28	74.22 66	85.49 46	90.03 46	88.52 55	93.66 70	96.74 61

DeepC-MVS	84.14 3	88.80 31	88.03 38	89.71 26	94.83 36	96.56 31	92.57 27	89.38 16	89.25 42	79.59 38	70.02 60	77.05 56	88.24 32	92.44 24	92.79 13	93.65 73	98.10 25
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
view600		77.68 102	75.68 117	80.01 84	87.72 86	92.57 78	83.79 88	77.95 68	64.41 133	58.72 109	54.32 126	50.54 142	78.25 89	88.23 68	83.13 117	93.64 74	96.59 66

ACMMPR		91.15 19	91.44 21	90.81 19	96.61 17	96.25 39	93.09 20	87.08 23	93.32 19	84.78 19	92.08 13	82.10 38	89.71 18	90.24 44	89.82 44	93.61 75	96.30 68

tfpn1000		75.39 117	76.18 115	74.47 120	86.71 95	90.10 105	77.57 153	74.78 97	68.76 118	53.33 125	63.57 84	58.37 120	60.84 182	83.80 116	81.24 156	93.58 76	87.42 172

thres400		78.39 95	76.39 111	80.73 82	88.02 83	92.94 74	84.77 84	78.88 60	65.20 130	59.70 104	55.20 121	50.85 141	79.45 86	88.81 60	84.81 92	93.57 77	96.91 56

ACMMP		88.48 34	88.71 35	88.22 36	94.61 39	95.53 50	90.64 41	85.60 32	90.97 29	78.62 40	89.88 21	74.20 67	86.29 42	88.16 74	86.37 77	93.57 77	95.86 74
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
CHOSEN 280x420		82.15 65	85.87 51	77.80 99	86.54 99	93.42 69	81.74 101	59.96 198	78.99 74	63.99 85	74.50 47	83.95 31	80.99 74	89.53 53	85.01 90	93.56 79	95.71 78

thres600view7		77.66 103	75.67 118	79.98 85	87.71 87	92.56 79	83.79 88	77.94 69	64.41 133	58.69 110	54.32 126	50.54 142	78.23 90	88.23 68	83.06 119	93.52 80	96.55 67

thres200		78.69 91	76.71 106	80.99 81	88.35 78	92.56 79	86.03 70	77.94 69	66.27 122	60.66 100	56.08 116	51.11 140	79.45 86	88.23 68	85.54 88	93.52 80	97.20 43

AdaColmap		88.46 35	85.75 53	91.62 16	96.25 24	95.35 53	90.71 39	91.08 7	90.22 35	86.17 14	74.33 48	73.67 70	92.00 11	86.31 94	85.82 85	93.52 80	94.53 91

XVS					89.65 64	95.93 42	85.97 71			76.32 47		82.05 39				93.51 83

X-MVStestdata					89.65 64	95.93 42	85.97 71			76.32 47		82.05 39				93.51 83

tfpn111		80.42 77	79.77 87	81.18 75	88.42 74	92.55 81	85.12 79	77.94 69	70.15 110	61.00 98	74.56 45	51.22 136	81.11 69	88.23 68	84.80 93	93.50 85	96.90 57

conf200view11		79.04 87	77.21 102	81.18 75	88.42 74	92.55 81	85.12 79	77.94 69	70.15 110	61.00 98	56.65 108	51.22 136	81.11 69	88.23 68	84.80 93	93.50 85	96.90 57

tfpn200view9		79.05 86	77.21 102	81.18 75	88.42 74	92.55 81	85.12 79	77.94 69	70.15 110	61.01 96	56.65 108	51.22 136	81.11 69	88.23 68	84.80 93	93.50 85	96.90 57

Effi-MVS+		79.80 81	80.04 83	79.52 88	85.53 111	93.31 70	85.28 75	70.68 132	74.15 94	58.79 108	62.03 93	60.51 111	83.37 61	88.41 67	86.09 81	93.49 88	95.80 76

UGNet		80.71 76	83.09 66	77.93 98	87.02 93	92.71 75	80.28 110	76.53 83	73.83 98	71.35 68	70.07 59	73.71 69	58.93 186	87.39 80	86.97 69	93.48 89	96.94 54
Wanjuan Su, Qingshan Xu, Wenbing Tao: Uncertainty-guided Multi-view Stereo Network for Depth Estimation. IEEE Transactions on Circuits and Systems for Video Technology, 2022
Fast-Effi-MVS+		77.37 106	76.68 107	78.17 97	82.84 122	89.94 107	81.47 103	68.01 152	72.99 100	60.26 103	55.07 122	53.20 134	82.99 62	86.47 93	86.12 80	93.46 90	92.98 123

V42		71.58 148	70.11 151	73.30 141	75.66 185	86.68 137	79.17 134	69.92 136	59.29 157	52.80 128	44.36 165	45.66 163	68.83 153	79.48 162	81.49 146	93.44 91	93.82 116

X-MVS		89.73 29	90.65 26	88.66 32	96.44 22	95.93 42	92.26 31	86.98 25	90.73 32	76.32 47	89.56 22	82.05 39	86.51 41	89.98 47	89.60 50	93.43 92	96.72 62

pmmvs5		70.01 165	69.31 162	70.82 174	75.80 182	86.26 143	72.94 181	67.91 153	53.84 185	47.22 175	47.31 150	41.47 195	67.61 156	83.93 115	81.93 133	93.42 93	90.42 141

MAR-MVS		85.65 49	86.30 48	84.88 56	95.51 34	95.89 48	86.50 65	76.71 81	89.23 43	68.59 77	70.93 58	74.49 64	88.55 26	89.40 55	90.30 40	93.42 93	93.88 110
Zhenyu Xu, Yiguang Liu, Xuelei Shi, Ying Wang, Yunan Zheng: MARMVS: Matching Ambiguity Reduced Multiple View Stereo for Efficient Large Scale Scene Reconstruction. CVPR 2020
TAPA-MVS	80.99 7	84.83 54	84.42 57	85.31 54	91.89 53	93.73 65	88.53 53	82.80 47	89.99 37	69.78 74	71.53 56	75.03 63	85.47 47	86.26 95	84.54 99	93.39 95	89.90 143
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
CostFormer		80.72 73	81.81 76	79.44 89	86.50 100	91.65 93	84.31 87	59.84 199	80.86 68	72.69 58	62.46 90	73.74 68	79.93 83	82.58 125	84.50 100	93.37 96	96.90 57

SD-MVS		93.36 9	94.33 6	92.22 10	94.68 38	97.89 11	94.56 10	90.89 10	94.80 9	90.04 6	93.53 7	90.14 7	89.78 17	92.74 22	92.17 17	93.35 97	99.07 10

IterMVS-LS		76.80 113	76.33 112	77.35 103	84.07 119	84.11 168	81.54 102	68.52 146	66.17 123	61.74 88	57.84 104	64.31 102	74.88 105	83.48 120	86.21 79	93.34 98	92.16 128
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
MP-MVS		90.81 22	91.45 20	90.06 22	96.59 18	96.33 38	92.46 29	87.19 22	90.27 34	82.54 31	91.38 15	84.88 25	88.27 31	90.58 42	89.30 53	93.30 99	97.44 35

v13		69.55 174	67.33 180	72.14 159	75.83 180	79.04 203	78.22 145	64.85 180	56.16 173	50.60 154	43.43 182	44.56 173	70.05 152	79.01 169	80.92 169	93.28 100	88.22 167

v16		70.93 158	68.76 167	73.47 138	76.60 158	79.66 196	79.57 125	65.81 170	60.85 144	54.44 122	44.50 163	45.90 158	71.15 130	79.50 160	81.39 153	93.27 101	89.51 148

v12		69.66 172	67.45 179	72.23 156	75.89 177	79.13 202	78.29 144	64.96 179	56.40 171	50.75 152	43.53 180	44.60 172	70.21 150	79.11 167	80.99 167	93.27 101	88.41 166

CP-MVS		90.57 23	90.68 25	90.44 20	96.13 25	95.90 46	92.77 25	86.86 27	92.12 25	84.19 20	89.18 23	82.37 36	89.43 22	89.65 52	88.43 57	93.27 101	97.13 46

v17		70.82 160	68.69 168	73.31 140	76.53 159	79.67 195	79.45 128	65.80 171	60.32 148	53.75 123	44.51 162	45.92 157	71.09 132	79.49 161	81.38 154	93.26 104	89.54 147

canonicalmvs		85.93 48	86.26 49	85.54 53	88.94 68	95.44 51	89.56 46	76.01 87	87.83 48	77.70 42	76.43 43	68.66 89	87.80 34	87.02 83	91.51 29	93.25 105	96.95 53

V9		69.79 171	67.57 178	72.38 155	75.95 172	79.21 200	78.72 138	65.06 176	56.51 170	51.06 145	43.66 178	44.70 169	70.28 149	79.22 165	81.06 166	93.24 106	88.67 165

v18		71.13 155	68.98 163	73.63 136	76.66 157	79.78 194	79.95 113	65.98 167	61.34 143	54.71 119	44.75 156	46.06 156	71.27 129	79.59 155	81.51 145	93.21 107	89.81 144

V14		69.91 167	67.71 177	72.47 153	76.01 170	79.30 199	78.92 136	65.17 175	56.74 168	51.08 144	43.82 177	44.73 168	70.44 147	79.31 164	81.14 163	93.20 108	88.91 161

v8		71.42 154	69.69 155	73.43 139	76.45 161	85.12 162	79.53 127	67.47 159	59.34 156	52.90 127	44.60 158	45.82 159	71.05 135	79.56 159	81.45 149	93.17 109	91.96 131

v11		69.84 170	67.85 174	72.17 157	75.78 183	79.15 201	78.20 146	64.76 181	56.10 174	49.50 160	43.54 179	43.36 183	71.62 124	82.21 128	81.52 142	93.17 109	89.05 158

v1neww		72.02 144	70.23 148	74.10 126	76.45 161	87.06 126	79.59 121	71.75 123	59.35 154	52.60 131	44.59 159	45.74 161	71.06 133	79.57 156	81.46 147	93.16 111	93.84 114

v7new		72.02 144	70.23 148	74.10 126	76.45 161	87.06 126	79.59 121	71.75 123	59.35 154	52.60 131	44.59 159	45.74 161	71.06 133	79.57 156	81.46 147	93.16 111	93.84 114

v6		72.04 143	70.26 146	74.11 125	76.46 160	87.06 126	79.60 118	71.75 123	59.48 153	52.69 130	44.61 157	45.79 160	71.01 136	79.57 156	81.45 149	93.16 111	93.85 113

test0.0.03 1		71.70 147	74.68 125	68.23 180	81.79 125	83.81 171	68.64 192	70.57 133	68.81 117	43.47 194	62.77 89	60.09 113	51.77 204	82.48 126	81.67 138	93.16 111	83.13 192

EG-PatchMatch MVS		66.23 186	65.20 186	67.43 184	77.74 148	86.20 146	72.51 184	63.68 184	43.95 214	43.44 195	36.22 208	45.43 165	54.04 196	81.00 141	80.95 168	93.15 115	82.67 197

v15		70.00 166	67.82 175	72.55 151	76.06 168	79.37 198	79.10 135	65.30 174	56.89 167	51.18 139	43.96 174	44.76 167	70.52 145	79.40 163	81.22 158	93.13 116	89.14 157

GA-MVS		73.62 126	74.52 127	72.58 150	79.93 136	89.29 116	78.02 149	71.67 126	60.79 146	42.68 197	54.41 125	49.07 148	70.07 151	89.39 56	86.55 75	93.13 116	92.12 129

v2v482		71.73 146	69.80 153	73.99 129	75.88 179	86.66 138	79.58 124	71.90 121	57.58 162	50.41 156	45.35 154	43.24 185	73.05 117	79.69 151	82.18 131	93.08 118	93.87 111

PCF-MVS	82.38 4	85.52 50	84.41 58	86.81 46	91.51 56	96.23 40	90.27 42	89.81 14	77.87 77	70.67 70	69.20 62	77.86 50	85.55 45	85.92 99	86.38 76	93.03 119	97.43 37
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
TSAR-MVS + COLMAP		84.93 53	85.79 52	83.92 60	90.90 59	93.57 67	89.25 49	82.00 53	91.29 27	61.66 89	88.25 24	59.46 115	86.71 39	89.79 50	87.09 67	93.01 120	91.09 136

FC-MVSNet-train		79.54 82	78.20 95	81.09 78	86.55 98	88.63 119	79.96 112	78.53 62	70.90 108	68.24 78	65.87 71	56.45 127	80.29 82	86.20 97	84.08 103	92.97 121	95.31 83

tfpn_n400		74.36 121	74.39 128	74.32 122	86.37 102	89.86 108	79.71 116	75.69 91	60.00 149	47.47 172	64.85 76	54.72 129	63.70 173	83.80 116	83.35 113	92.96 122	84.16 186

tfpnconf		74.36 121	74.39 128	74.32 122	86.37 102	89.86 108	79.71 116	75.69 91	60.00 149	47.47 172	64.85 76	54.72 129	63.70 173	83.80 116	83.35 113	92.96 122	84.16 186

v144192		70.10 164	68.55 169	71.90 163	74.55 189	85.67 154	77.81 150	68.22 150	54.65 181	46.91 177	42.76 185	41.27 196	70.95 137	80.48 146	81.11 165	92.96 122	93.90 109

v1921920		69.85 169	68.38 171	71.58 166	74.35 190	85.39 158	77.78 151	67.88 154	54.64 182	45.39 185	42.11 190	39.97 200	71.10 131	81.68 137	81.17 162	92.96 122	93.69 119

HQP-MVS		86.17 47	87.35 41	84.80 57	91.41 57	92.37 85	91.05 38	84.35 39	88.52 46	64.21 84	87.05 27	68.91 87	84.80 52	89.12 57	88.16 61	92.96 122	97.31 39

tfpn_ndepth		78.22 98	78.84 91	77.49 101	88.32 79	90.95 98	80.79 105	76.31 85	74.24 93	59.50 106	69.52 61	60.02 114	67.11 161	85.06 106	82.95 121	92.94 127	89.18 156

v1141		71.53 150	69.69 155	73.68 133	76.08 166	86.86 131	79.59 121	72.07 118	57.01 164	50.78 150	44.23 168	44.70 169	70.68 140	79.61 154	81.52 142	92.89 128	93.92 106

divwei89l23v2f112		71.53 150	69.69 155	73.68 133	76.09 165	86.86 131	79.60 118	72.08 117	56.96 166	50.78 150	44.24 167	44.70 169	70.65 142	79.62 152	81.53 140	92.89 128	93.93 104

v10		70.97 157	69.44 158	72.75 146	75.90 176	84.58 167	79.43 130	66.45 164	58.07 160	49.93 159	43.87 175	43.68 178	71.91 122	82.04 131	81.70 135	92.89 128	92.11 130

v1240		69.28 176	67.82 175	71.00 173	74.09 193	85.13 161	76.54 165	67.28 161	53.17 188	44.70 191	41.55 194	39.38 202	70.51 146	81.29 140	81.18 160	92.88 131	93.02 122

v7		71.49 152	69.98 152	73.25 142	75.89 177	86.45 141	79.44 129	69.29 142	58.07 160	50.08 158	43.87 175	43.67 179	71.94 121	82.03 133	81.70 135	92.88 131	94.04 100

v1		71.54 149	69.71 154	73.66 135	76.08 166	86.88 130	79.60 118	72.06 119	57.00 165	50.75 152	44.23 168	44.79 166	70.61 143	79.62 152	81.52 142	92.88 131	93.93 104

MSDG		78.11 99	73.17 134	83.86 61	91.78 55	86.83 134	85.25 76	86.02 29	72.84 102	69.69 76	51.43 135	54.00 133	77.61 91	81.95 134	82.27 128	92.83 134	82.91 194

tfpnnormal		69.29 175	65.58 183	73.62 137	79.87 137	84.82 164	76.97 160	75.12 96	45.29 212	49.03 163	35.57 209	37.20 210	68.02 155	82.70 123	81.24 156	92.69 135	92.20 127

tpm		73.50 128	74.85 123	71.93 161	83.19 121	86.84 133	78.61 140	55.91 207	65.64 125	48.90 164	56.30 114	61.09 107	72.31 119	79.10 168	80.61 170	92.68 136	94.35 95

IterMVS		72.43 136	74.05 130	70.55 175	80.34 135	81.17 187	77.44 155	61.00 193	63.57 138	46.82 178	55.88 117	59.09 117	65.03 167	83.15 121	83.83 109	92.67 137	91.65 134
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
TSAR-MVS + ACMM		90.98 21	93.18 15	88.42 34	95.69 30	96.73 29	94.52 11	86.97 26	92.99 21	76.32 47	92.31 12	86.64 17	84.40 57	92.97 21	92.02 19	92.62 138	98.59 19

v1144		70.93 158	69.42 160	72.70 147	75.48 186	86.26 143	79.22 133	69.39 141	55.61 177	48.05 170	43.47 181	42.55 190	71.51 127	82.11 129	81.74 134	92.56 139	94.17 98

conf0.05thres1000		74.20 125	71.44 142	77.43 102	86.09 108	89.85 110	80.82 104	75.79 90	53.51 187	54.71 119	44.37 164	49.78 145	74.67 106	85.02 107	83.47 111	92.49 140	94.10 99

CNLPA		84.72 56	82.14 72	87.73 38	92.85 47	93.83 63	84.70 85	85.07 33	90.90 30	83.16 26	56.28 115	71.53 76	88.14 33	84.19 112	84.00 107	92.48 141	94.26 96

Fast-Effi-MVS+-dtu		73.56 127	75.32 120	71.50 167	80.35 134	86.83 134	79.72 115	58.07 203	67.64 120	44.83 190	60.28 98	54.07 132	73.59 114	81.90 136	82.30 127	92.46 142	94.18 97

v1192		70.32 163	68.77 166	72.12 160	74.76 188	85.62 155	78.73 137	68.53 145	55.08 179	46.34 180	42.39 187	40.67 197	71.90 123	82.27 127	81.53 140	92.43 143	93.86 112

CSCG		89.81 28	89.69 30	89.96 24	96.55 19	97.90 10	92.89 23	87.06 24	88.74 45	86.17 14	78.24 38	86.53 18	84.75 53	87.82 76	90.59 38	92.32 144	98.01 26

LGP-MVS_train		82.12 66	82.57 69	81.59 69	89.26 67	90.23 103	88.76 52	78.05 67	81.26 67	61.64 90	79.52 36	62.11 105	79.59 85	85.20 105	84.68 98	92.27 145	95.02 86

PLC	81.02 6	84.81 55	81.81 76	88.31 35	93.77 43	90.35 101	88.80 51	84.47 38	86.76 52	82.17 32	66.56 67	71.01 80	88.41 29	85.48 102	84.28 102	92.26 146	88.21 168
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
UniMVSNet_NR-MVSNet		73.11 132	72.59 135	73.71 132	76.90 153	86.58 140	77.01 158	75.82 88	65.59 126	48.82 165	50.97 137	48.42 150	71.61 125	79.19 166	83.03 120	92.11 147	94.37 93

OMC-MVS		86.38 46	86.21 50	86.57 49	92.30 50	94.35 59	87.60 57	83.51 42	92.32 24	77.37 44	72.27 52	77.83 51	86.59 40	87.62 79	85.95 82	92.08 148	93.11 120

Baseline_NR-MVSNet		70.61 161	68.87 165	72.65 148	75.95 172	80.49 190	75.92 170	74.75 98	65.10 131	48.78 167	41.28 195	44.28 176	68.45 154	78.67 172	79.64 175	92.04 149	92.62 124

Vis-MVSNet (Re-imp)		78.28 97	82.68 68	73.16 144	86.64 96	92.68 76	78.07 148	74.48 101	74.05 95	53.47 124	64.22 82	76.52 59	54.28 194	88.96 59	88.29 59	92.03 150	94.00 101

NR-MVSNet		71.47 153	71.11 144	71.90 163	77.73 149	86.02 152	76.88 162	74.42 102	65.39 128	46.09 183	49.10 144	39.87 201	64.27 169	81.40 138	82.24 129	91.99 151	93.75 117

IS_MVSNet		80.92 69	84.14 60	77.16 104	87.43 89	93.90 62	80.44 106	74.64 99	75.05 90	61.10 95	65.59 72	76.89 58	67.39 159	90.88 39	90.05 41	91.95 152	96.62 65

ADS-MVSNet		72.11 140	73.72 132	70.24 176	81.24 130	86.59 139	74.75 176	50.56 219	72.58 103	49.17 162	55.40 120	61.46 106	73.80 112	76.01 188	78.14 184	91.93 153	85.86 179

USDC		73.43 130	72.31 137	74.73 116	80.86 133	86.21 145	80.42 107	71.83 122	71.69 106	46.94 176	59.60 100	42.58 189	76.47 100	82.66 124	81.22 158	91.88 154	82.24 199

v148		70.34 162	68.46 170	72.54 152	76.04 169	86.38 142	74.83 174	72.73 108	55.88 176	55.26 117	43.32 184	43.49 180	64.52 168	76.93 185	80.11 173	91.85 155	93.11 120

TranMVSNet+NR-MVSNet		71.12 156	70.24 147	72.15 158	76.01 170	84.80 165	76.55 164	75.65 93	61.99 142	45.29 186	48.42 148	43.07 186	67.55 157	78.28 175	82.83 122	91.85 155	92.29 126

FMVSNet1		74.26 124	71.95 139	76.95 106	74.28 192	83.94 170	83.61 91	69.99 135	57.08 163	65.08 82	42.39 187	57.41 123	76.98 95	86.57 88	86.83 72	91.77 157	89.42 149

pmmvs4		73.38 131	71.53 141	75.55 111	75.95 172	85.24 160	77.25 157	71.59 127	71.03 107	63.10 86	49.09 146	44.22 177	73.73 113	82.04 131	80.18 172	91.68 158	88.89 162

EPP-MVSNet		80.82 71	82.79 67	78.52 94	86.31 104	92.37 85	79.83 114	74.51 100	73.79 99	64.46 83	67.01 66	80.63 45	74.33 109	85.63 100	84.35 101	91.68 158	95.79 77

CDS-MVSNet		76.57 114	76.78 105	76.32 108	80.94 132	89.75 111	82.94 95	72.64 109	59.01 158	62.95 87	58.60 102	62.67 103	66.91 163	86.26 95	87.20 64	91.57 160	93.97 103
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
TransMVSNet (Re)		66.87 183	64.30 194	69.88 177	78.32 143	81.35 186	73.88 178	74.34 104	43.19 216	45.20 188	40.12 196	42.37 192	55.97 192	80.85 143	79.15 176	91.56 161	83.06 193

RPMNet		73.46 129	77.85 99	68.34 179	81.71 128	85.52 156	73.83 180	50.54 220	74.05 95	46.10 182	53.03 133	71.91 75	66.31 165	83.55 119	82.18 131	91.55 162	94.71 87

UniMVSNet (Re)		72.12 139	72.28 138	71.93 161	76.77 154	87.38 125	75.73 173	73.51 106	65.76 124	50.24 157	48.65 147	46.49 155	63.85 171	80.10 148	82.47 124	91.49 163	95.13 85

CR-MVSNet		74.84 119	77.91 98	71.26 171	81.77 127	85.52 156	78.32 141	54.14 212	74.05 95	51.09 141	50.00 141	71.38 79	70.77 138	86.48 91	84.03 105	91.46 164	93.92 106

tpmp4_e23		78.57 92	78.48 93	78.68 92	85.38 113	89.14 117	84.69 86	60.32 197	78.81 75	70.65 71	57.89 103	65.54 97	79.63 84	80.09 149	83.24 115	91.41 165	94.63 90

ACMP	79.58 9	82.23 64	81.82 75	82.71 63	88.15 81	90.95 98	85.23 77	78.52 63	81.70 66	72.52 64	78.41 37	60.63 110	80.48 80	82.88 122	83.44 112	91.37 166	94.70 88
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
pm-mvs1		69.62 173	68.07 173	71.44 168	77.21 151	85.32 159	76.11 169	71.05 128	46.55 210	51.17 140	41.83 192	48.20 151	61.81 179	84.00 114	81.14 163	91.28 167	89.42 149

DU-MVS		72.19 138	71.35 143	73.17 143	75.95 172	86.02 152	77.01 158	74.42 102	65.39 128	48.82 165	49.10 144	42.81 187	71.61 125	78.67 172	83.10 118	91.22 168	94.37 93

WR-MVS_H		64.14 200	65.36 185	62.71 202	72.47 197	82.33 180	65.13 200	66.99 162	51.81 192	36.47 207	43.33 183	42.77 188	43.99 209	72.41 200	75.99 198	91.20 169	88.86 163

TAMVS		72.06 142	71.76 140	72.41 154	76.68 156	88.12 121	74.82 175	68.09 151	53.52 186	56.91 114	52.94 134	56.93 126	66.91 163	81.37 139	82.44 125	91.07 170	86.99 174

PEN-MVS		64.35 196	64.29 195	64.42 197	72.67 195	79.83 193	66.97 194	68.24 149	51.21 194	35.29 211	44.09 170	38.51 205	52.36 202	71.06 205	77.65 187	90.99 171	87.68 171

tpm cat1		76.93 111	76.19 114	77.79 100	85.08 115	88.58 120	82.96 94	59.33 200	75.72 89	72.64 59	51.25 136	64.41 100	75.74 104	77.90 178	80.10 174	90.97 172	95.35 81

UA-Net		78.30 96	80.92 78	75.25 113	87.42 90	92.48 84	79.54 126	75.49 94	60.47 147	60.52 101	68.44 64	84.08 30	57.54 189	88.54 65	88.45 56	90.96 173	83.97 189

ACMM	78.09 10	80.91 70	78.39 94	83.86 61	89.61 66	87.71 123	85.16 78	80.67 59	79.04 73	74.18 54	63.82 83	60.84 109	82.59 64	84.33 111	83.59 110	90.96 173	89.39 151
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
CP-MVSNet		64.84 192	64.97 187	64.69 195	72.09 199	81.04 188	66.66 196	67.53 158	52.45 190	37.40 204	44.00 173	38.37 206	53.54 199	72.26 201	76.93 194	90.94 175	89.75 145

PS-CasMVS		64.22 199	64.19 196	64.25 198	71.86 201	80.67 189	66.42 198	67.43 160	50.64 196	36.48 206	42.60 186	37.46 209	52.56 201	71.98 202	76.69 196	90.76 176	89.29 155

DTE-MVSNet		63.26 203	63.41 201	63.08 201	72.59 196	78.56 204	65.03 202	68.28 148	50.53 198	32.38 215	44.03 171	37.79 208	49.48 205	70.83 208	76.73 195	90.73 177	85.42 180

CPTT-MVS		88.17 38	87.84 39	88.55 33	93.33 44	93.75 64	92.33 30	84.75 35	89.87 38	81.72 35	83.93 33	81.12 42	88.45 27	85.42 104	84.07 104	90.72 178	96.72 62

dps		75.76 116	75.02 122	76.63 107	84.51 117	88.12 121	77.51 154	58.33 202	75.91 88	71.98 66	57.37 106	57.85 121	76.81 98	77.89 179	78.40 183	90.63 179	89.63 146

ACMH	71.22 14	72.65 135	70.13 150	75.59 110	86.19 106	86.14 150	75.76 172	77.63 74	54.79 180	46.16 181	53.28 131	47.28 154	77.24 94	78.91 171	81.18 160	90.57 180	89.33 152
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
Anonymous20231206		62.05 206	61.83 206	62.30 204	72.09 199	77.84 206	63.10 204	67.62 157	50.20 199	36.68 205	29.59 219	37.05 212	43.90 210	77.33 184	77.31 189	90.41 181	83.49 190

MDTV_nov1_ep13_2view		64.72 193	64.94 188	64.46 196	71.14 206	81.94 182	67.53 193	54.54 211	55.92 175	43.29 196	44.02 172	43.27 184	59.87 185	71.85 203	74.77 201	90.36 182	82.82 195

MDTV_nov1_ep13		77.20 109	80.04 83	73.90 130	82.22 123	90.14 104	79.25 132	61.52 191	78.63 76	56.98 112	65.52 74	72.80 74	73.05 117	80.93 142	83.20 116	90.36 182	89.05 158

v7n		66.43 185	65.51 184	67.51 183	71.63 204	83.10 173	70.89 189	65.02 177	50.13 200	44.68 192	39.59 198	38.77 204	62.57 177	77.59 182	78.91 177	90.29 184	90.44 140

PMMVS		82.26 63	85.48 54	78.51 96	85.92 110	91.92 89	78.30 143	70.77 130	86.30 54	61.11 94	82.46 35	70.88 81	84.70 54	88.05 75	84.78 97	90.24 185	93.98 102

DeepPCF-MVS	86.71 1	91.00 20	94.05 8	87.43 40	95.58 32	98.17 4	86.22 67	88.59 19	97.01 1	76.77 46	85.11 30	88.90 11	87.29 35	95.02 6	94.69 3	90.15 186	99.48 4

MVS-HIRNet		64.63 194	64.03 197	65.33 193	75.01 187	82.84 174	58.54 213	52.10 217	55.42 178	49.29 161	29.83 218	43.48 181	66.97 162	78.28 175	78.81 178	90.07 187	79.52 202

PatchMatch-RL		78.75 89	76.47 110	81.41 71	88.53 73	91.10 96	78.09 147	77.51 76	77.33 81	71.98 66	64.38 81	48.10 152	82.55 65	84.06 113	82.35 126	89.78 188	87.97 170

V4		65.34 187	64.59 191	66.21 189	69.64 208	82.42 178	69.22 190	62.80 187	49.60 202	45.21 187	39.33 200	41.82 194	60.66 184	72.61 197	77.03 192	89.76 189	89.32 154

v52		65.34 187	64.59 191	66.21 189	69.63 209	82.41 179	69.22 190	62.80 187	49.63 201	45.15 189	39.31 201	41.85 193	60.68 183	72.61 197	77.02 193	89.75 190	89.33 152

anonymousdsp		67.61 180	68.94 164	66.04 191	71.44 205	83.97 169	66.45 197	63.53 185	50.54 197	42.42 198	49.39 142	45.63 164	62.84 176	77.99 177	81.34 155	89.59 191	93.75 117

v748		65.00 189	63.86 198	66.33 188	71.85 202	82.15 181	66.80 195	65.64 172	48.50 206	47.98 171	39.62 197	39.20 203	56.44 191	71.25 204	77.53 188	89.29 192	88.74 164

pmmvs-eth3d		64.24 197	61.96 205	66.90 186	66.35 213	76.04 209	66.09 199	66.31 165	52.59 189	50.94 146	37.61 204	32.79 221	62.43 178	75.78 189	75.48 199	89.27 193	83.39 191

WR-MVS		64.98 190	66.59 182	63.09 200	74.34 191	82.68 176	64.98 203	69.17 143	54.42 183	36.18 209	44.32 166	44.35 175	44.65 207	73.60 194	77.83 185	89.21 194	88.96 160

ACMH+	72.14 13	72.38 137	69.34 161	75.93 109	85.21 114	84.89 163	76.96 161	76.04 86	59.76 151	51.63 136	50.37 139	48.69 149	76.90 97	76.06 187	78.69 179	88.85 195	86.90 175

Effi-MVS+-dtu		74.57 120	74.60 126	74.53 119	81.38 129	86.74 136	80.39 108	67.70 156	67.36 121	53.06 126	59.86 99	57.50 122	75.84 103	80.19 147	78.62 181	88.79 196	91.95 132

LS3D		78.72 90	75.79 116	82.15 66	91.91 52	89.39 115	83.66 90	85.88 31	76.81 85	59.22 107	57.67 105	58.53 119	83.72 59	82.07 130	81.63 139	88.50 197	84.39 184

FC-MVSNet-test		67.04 182	72.47 136	60.70 208	76.92 152	81.41 184	61.52 205	69.45 140	65.58 127	26.74 226	61.79 94	60.40 112	41.17 217	77.60 181	77.78 186	88.41 198	82.70 196

test20.03		57.93 212	59.22 210	56.44 211	71.84 203	73.78 213	53.55 217	65.96 168	43.02 217	28.46 222	37.50 205	38.17 207	30.41 226	75.25 190	74.42 204	88.41 198	72.37 217

testgi		63.11 204	64.88 190	61.05 206	75.83 180	78.51 205	60.42 208	66.20 166	48.77 205	34.56 213	56.96 107	40.35 198	40.95 218	77.46 183	77.22 190	88.37 200	74.86 213

LP		59.72 208	58.23 212	61.44 205	75.67 184	74.97 211	61.05 207	48.34 223	54.02 184	40.82 200	31.61 213	36.92 213	54.69 193	67.52 213	71.18 209	88.08 201	71.42 219

TinyColmap		67.16 181	63.51 199	71.42 169	77.94 147	79.54 197	72.80 182	69.78 138	56.58 169	45.52 184	44.53 161	33.53 219	74.45 108	76.91 186	77.06 191	88.03 202	76.41 206

CVMVSNet		68.95 177	70.79 145	66.79 187	79.69 139	83.75 172	72.05 185	70.90 129	56.20 172	36.30 208	54.94 124	59.22 116	54.03 197	78.33 174	78.65 180	87.77 203	84.44 183

test2356		58.43 211	59.52 209	57.16 210	66.71 212	68.00 218	54.69 215	60.91 195	49.22 204	28.63 221	41.86 191	33.68 218	44.36 208	72.98 195	75.47 200	87.69 204	75.40 210

MIMVSNet1		52.76 216	53.95 214	51.38 219	41.96 233	70.79 216	53.56 216	63.03 186	39.36 219	27.83 224	22.73 228	33.07 220	34.47 223	70.49 209	72.69 206	87.41 205	68.51 220

testus		55.91 213	56.38 213	55.37 214	65.15 215	65.88 221	50.07 221	60.92 194	45.62 211	26.99 225	41.74 193	24.43 228	42.08 214	69.50 210	73.60 205	86.97 206	73.91 214

PM-MVS		63.52 202	62.51 204	64.70 194	64.79 218	76.08 208	65.07 201	62.08 189	58.13 159	46.56 179	44.98 155	31.31 222	62.89 175	72.58 199	69.93 214	86.81 207	84.55 182

N_pmnet		60.52 207	58.83 211	62.50 203	68.97 210	75.61 210	59.72 211	66.47 163	51.90 191	41.26 199	35.42 210	35.63 214	52.25 203	67.07 215	70.08 213	86.35 208	76.10 207

COLMAP_ROB	66.31 15	69.91 167	66.61 181	73.76 131	86.44 101	82.76 175	76.59 163	76.46 84	63.82 136	50.92 147	45.60 153	49.13 147	65.87 166	74.96 192	74.45 203	86.30 209	75.57 209
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
pmmvs6		64.24 197	61.77 207	67.12 185	72.39 198	81.39 185	71.33 187	65.95 169	36.05 221	48.48 168	30.55 214	43.45 182	58.75 187	77.88 180	76.36 197	85.83 210	86.70 177

new-patchmatchnet		53.91 215	52.69 215	55.33 215	64.83 217	70.90 215	52.24 219	61.75 190	41.09 218	30.82 216	29.90 217	28.22 224	36.69 221	61.52 221	65.08 220	85.64 211	72.14 218

SixPastTwentyTwo		63.75 201	63.42 200	64.13 199	72.91 194	80.34 191	61.29 206	63.90 182	49.58 203	40.42 201	54.99 123	37.13 211	60.90 180	68.46 211	70.80 210	85.37 212	82.65 198

TDRefinement		67.82 179	64.91 189	71.22 172	82.08 124	81.45 183	77.42 156	73.79 105	59.62 152	48.35 169	42.35 189	42.40 191	60.87 181	74.69 193	74.64 202	84.83 213	79.20 203

CMPMVS	50.59 17	66.74 184	62.72 203	71.42 169	85.40 112	89.72 112	72.69 183	70.72 131	51.24 193	51.75 135	38.91 202	44.40 174	63.74 172	70.84 207	71.52 207	84.19 214	72.45 216
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
EU-MVSNet		58.73 210	60.92 208	56.17 212	66.17 214	72.39 214	58.85 212	61.24 192	48.47 207	27.91 223	46.70 152	40.06 199	39.07 219	68.27 212	70.34 212	83.77 215	80.23 201

MDA-MVSNet-bldmvs		54.99 214	52.66 216	57.71 209	52.74 229	74.87 212	55.61 214	68.41 147	43.65 215	32.54 214	37.93 203	22.11 230	54.11 195	48.85 228	67.34 217	82.85 216	73.88 215

LTVRE_ROB	63.07 16	64.49 195	63.16 202	66.04 191	77.47 150	82.64 177	70.98 188	65.02 177	34.01 225	29.61 218	49.12 143	35.58 215	70.57 144	75.10 191	78.45 182	82.60 217	87.24 173
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
pmmvs3		52.59 217	52.43 217	52.78 217	54.53 227	64.49 223	50.07 221	46.89 226	35.31 223	30.19 217	27.27 221	26.96 226	53.02 200	67.28 214	70.54 211	81.96 218	75.20 211

1111		48.34 221	47.93 220	48.83 221	58.14 223	59.33 226	37.54 227	43.85 227	31.76 226	29.36 219	23.26 225	34.58 216	42.20 212	65.15 217	68.72 216	81.86 219	52.66 228

testmv		46.89 222	46.37 222	47.48 222	60.96 219	58.36 228	36.71 229	56.94 204	27.16 229	17.93 233	23.94 223	18.84 232	31.06 224	61.55 219	66.72 218	81.28 220	68.05 221

test1235678		46.88 223	46.36 223	47.48 222	60.96 219	58.35 229	36.71 229	56.94 204	27.15 230	17.93 233	23.93 224	18.82 233	31.06 224	61.55 219	66.71 219	81.27 221	68.04 222

RPSCF		74.27 123	73.24 133	75.48 112	81.01 131	80.18 192	76.24 166	72.37 115	74.84 92	68.24 78	72.47 50	67.39 91	73.89 110	71.05 206	69.38 215	81.14 222	77.37 205

Anonymous20231211		49.72 220	47.45 221	52.38 218	60.54 221	66.16 220	52.47 218	60.87 196	25.32 231	25.16 229	15.98 230	23.66 229	37.00 220	61.01 223	64.41 222	78.25 223	75.60 208

new_pmnet		50.32 218	51.36 218	49.11 220	49.19 230	64.89 222	48.66 224	47.99 225	47.55 208	26.27 227	29.51 220	28.66 223	44.89 206	61.12 222	62.74 224	77.66 224	65.03 223

testpf		59.38 209	64.51 193	53.40 216	76.71 155	66.40 219	50.18 220	38.98 233	64.13 135	35.10 212	47.91 149	51.41 135	43.16 211	66.37 216	71.23 208	76.25 225	84.14 188

ambc			50.35 219		55.61 226	59.93 225	48.73 223		44.08 213	35.81 210	24.01 222	10.64 237	41.57 216	72.83 196	63.35 223	74.99 226	77.61 204

test12356		41.15 224	41.46 225	40.78 225	53.10 228	49.87 230	33.37 232	52.25 216	25.12 232	15.64 235	22.76 227	15.01 234	15.81 231	52.97 225	64.54 221	74.50 227	59.96 224

FPMVS		50.25 219	45.67 224	55.58 213	70.48 207	60.12 224	59.78 210	59.33 200	46.66 209	37.94 202	30.22 216	27.51 225	35.94 222	50.98 227	47.90 227	70.02 228	56.31 225

PMMVS2		32.52 228	33.92 228	30.88 231	34.15 236	44.70 233	27.79 234	39.69 232	22.21 233	4.31 240	15.73 231	14.13 235	12.45 235	40.11 230	47.00 228	66.88 229	53.54 226

tmp_tt				39.78 226	56.31 225	31.71 236	35.84 231	15.08 235	82.57 65	50.83 149	63.07 87	47.51 153	15.28 232	52.23 226	44.24 229	65.35 230

PMVS	36.83 18	40.62 225	36.39 226	45.56 224	58.40 222	33.20 234	32.62 233	56.02 206	28.25 228	37.92 203	22.29 229	26.15 227	25.29 228	48.49 229	43.82 230	63.13 231	52.53 229
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
no-one		32.08 229	31.09 230	33.23 229	46.10 231	46.90 232	20.80 236	49.13 221	16.27 234	7.85 237	10.62 232	10.68 236	13.65 234	31.50 232	51.31 226	61.83 232	50.38 230

Gipuma		35.20 226	33.96 227	36.65 227	43.30 232	32.51 235	26.96 235	48.31 224	38.87 220	20.08 232	8.08 233	7.41 238	26.44 227	53.60 224	58.43 225	54.81 233	38.79 232
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
DeepMVS_CX						48.96 231	43.77 225	40.58 231	50.93 195	24.67 230	36.95 207	20.18 231	41.60 215	38.92 231		52.37 234	53.31 227

E-PMN		21.42 230	17.56 232	25.94 232	36.25 235	19.02 238	11.56 237	43.72 230	15.25 236	6.99 238	8.04 234	4.53 240	21.77 230	16.13 234	26.16 232	35.34 235	33.77 233

EMVS		20.61 232	16.32 233	25.62 233	36.41 234	18.93 239	11.51 238	43.75 229	15.65 235	6.53 239	7.56 236	4.68 239	22.03 229	14.56 235	23.10 233	33.51 236	29.77 234

MVE	25.07 19	21.25 231	23.51 231	18.62 234	15.07 237	29.77 237	10.67 239	34.60 234	12.51 237	9.46 236	7.84 235	3.82 241	14.38 233	27.45 233	42.42 231	27.56 237	40.74 231
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
.test1245		33.05 227	31.21 229	35.20 228	58.14 223	59.33 226	37.54 227	43.85 227	31.76 226	29.36 219	23.26 225	34.58 216	42.20 212	65.15 217	0.77 234	0.11 238	3.62 236

testmvs		0.76 233	1.23 234	0.21 235	0.05 239	0.21 240	0.38 241	0.09 236	0.94 238	0.05 242	2.13 238	0.08 242	0.60 237	0.82 236	0.77 234	0.11 238	3.62 236

test123		0.67 234	1.11 235	0.16 236	0.01 240	0.14 241	0.20 242	0.04 238	0.77 239	0.02 243	2.15 237	0.02 243	0.61 236	0.23 237	0.72 236	0.07 240	3.76 235

sosnet-low-res		0.00 235	0.00 236	0.00 237	0.00 241	0.00 242	0.00 243	0.00 239	0.00 240	0.00 244	0.00 239	0.00 244	0.00 239	0.00 238	0.00 237	0.00 241	0.00 238

sosnet		0.00 235	0.00 236	0.00 237	0.00 241	0.00 242	0.00 243	0.00 239	0.00 240	0.00 244	0.00 239	0.00 244	0.00 239	0.00 238	0.00 237	0.00 241	0.00 238

MTAPA										91.14 2		85.84 20

MTMP										90.95 3		84.13 29

Patchmatch-RL test							8.17 240

mPP-MVS					95.90 28							80.22 47

NP-MVS									89.55 41

Patchmtry						87.41 124	78.32 141	54.14 212		51.09 141

High-resolution multi-view benchmark