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.

DeepPCF-MVS	92.65 2	95.50 30	96.96 15	93.79 46	96.44 51	98.21 34	93.51 85	94.08 31	96.94 1	89.29 39	93.08 27	96.77 22	93.82 47	97.68 6	97.40 4	95.59 182	98.65 12

SD-MVS		97.35 5	97.73 4	96.90 11	97.35 39	98.66 11	97.85 23	96.25 8	96.86 2	94.54 5	96.75 8	99.13 3	96.99 5	96.94 20	96.58 19	98.39 33	99.20 1

APDe-MVS		97.79 1	97.96 2	97.60 1	99.20 1	99.10 3	98.88 1	96.68 2	96.81 3	94.64 4	97.84 1	98.02 8	97.24 2	97.74 5	97.02 10	98.97 1	99.16 2

TSAR-MVS + MP.		97.31 6	97.64 6	96.92 10	97.28 41	98.56 20	98.61 3	95.48 23	96.72 4	94.03 10	96.73 9	98.29 6	97.15 3	97.61 9	96.42 22	98.96 2	99.13 3

HFP-MVS		97.11 10	97.19 12	97.00 8	98.97 13	98.73 9	98.37 9	95.69 17	96.60 5	93.28 16	96.87 5	96.64 23	97.27 1	96.64 25	96.33 28	98.44 24	98.56 17

ACMMPR		96.92 14	96.96 15	96.87 12	98.99 12	98.78 8	98.38 8	95.52 20	96.57 6	92.81 21	96.06 15	95.90 30	97.07 4	96.60 27	96.34 27	98.46 20	98.42 28

TSAR-MVS + ACMM		96.19 20	97.39 10	94.78 33	97.70 35	98.41 30	97.72 24	95.49 22	96.47 7	86.66 58	96.35 10	97.85 10	93.99 43	97.19 15	96.37 24	97.12 122	99.13 3

NCCC		96.75 16	96.67 21	96.85 13	99.03 9	98.44 29	98.15 14	96.28 7	96.32 8	92.39 22	92.16 31	97.55 15	96.68 17	97.32 10	96.65 18	98.55 12	98.26 32

DeepC-MVS_fast	93.32 1	96.48 19	96.42 24	96.56 17	98.70 21	98.31 33	97.97 20	95.76 16	96.31 9	92.01 24	91.43 36	95.42 34	96.46 20	97.65 8	97.69 1	98.49 17	98.12 41
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
SMA-MVS		97.42 4	97.82 3	96.95 9	99.18 2	99.05 5	98.10 17	96.31 6	96.28 10	92.94 19	95.50 21	99.21 2	96.69 16	97.96 2	97.67 2	98.50 15	99.06 7

OMC-MVS		94.49 39	94.36 39	94.64 36	97.17 43	97.73 50	95.49 51	92.25 39	96.18 11	90.34 36	88.51 47	92.88 44	94.90 34	94.92 56	94.17 58	97.69 101	96.15 107

CNVR-MVS		97.30 7	97.41 8	97.18 6	99.02 10	98.60 18	98.15 14	96.24 9	96.12 12	94.10 8	95.54 20	97.99 9	96.99 5	97.97 1	97.17 6	98.57 11	98.50 23

zzz-MVS		96.98 12	96.68 20	97.33 2	99.09 4	98.71 10	98.43 6	96.01 12	96.11 13	95.19 3	92.89 29	97.32 18	96.84 9	97.20 14	96.09 33	98.44 24	98.46 27

HPM-MVS++		97.22 8	97.40 9	97.01 7	99.08 5	98.55 21	98.19 12	96.48 4	96.02 14	93.28 16	96.26 12	98.71 5	96.76 13	97.30 12	96.25 30	98.30 47	98.68 11

HSP-MVS		97.51 3	97.70 5	97.29 3	99.00 11	99.17 1	98.61 3	96.41 5	95.88 15	94.34 7	97.72 2	99.04 4	96.93 8	97.29 13	95.90 36	98.45 23	98.94 9

ESAPD		97.65 2	97.98 1	97.27 4	99.12 3	99.14 2	98.66 2	96.80 1	95.74 16	93.46 13	97.72 2	99.48 1	96.76 13	97.77 3	96.92 13	98.83 4	99.07 6

TSAR-MVS + GP.		95.86 25	96.95 17	94.60 38	94.07 79	98.11 38	96.30 39	91.76 45	95.67 17	91.07 28	96.82 7	97.69 13	95.71 27	95.96 41	95.75 39	98.68 6	98.63 13

ACMMP_Plus		96.93 13	97.27 11	96.53 20	99.06 6	98.95 6	98.24 11	96.06 11	95.66 18	90.96 30	95.63 19	97.71 12	96.53 18	97.66 7	96.68 16	98.30 47	98.61 16

CNLPA		93.69 45	92.50 54	95.06 32	97.11 44	97.36 56	93.88 77	93.30 33	95.64 19	93.44 14	80.32 97	90.73 55	94.99 33	93.58 95	93.33 82	97.67 103	96.57 92

MCST-MVS		96.83 15	97.06 13	96.57 16	98.88 17	98.47 27	98.02 19	96.16 10	95.58 20	90.96 30	95.78 18	97.84 11	96.46 20	97.00 19	96.17 32	98.94 3	98.55 22

CSCG		95.68 27	95.46 32	95.93 24	98.71 20	99.07 4	97.13 31	93.55 32	95.48 21	93.35 15	90.61 40	93.82 40	95.16 31	94.60 71	95.57 41	97.70 100	99.08 5

CP-MVS		96.68 17	96.59 23	96.77 14	98.85 18	98.58 19	98.18 13	95.51 21	95.34 22	92.94 19	95.21 24	96.25 25	96.79 12	96.44 32	95.77 38	98.35 36	98.56 17

TSAR-MVS + COLMAP		92.39 54	92.31 59	92.47 59	95.35 67	96.46 85	96.13 41	92.04 42	95.33 23	80.11 98	94.95 25	77.35 112	94.05 42	94.49 75	93.08 94	97.15 119	94.53 149

MVS_111021_LR		94.84 35	95.57 29	94.00 40	97.11 44	97.72 52	94.88 57	91.16 51	95.24 24	88.74 43	96.03 16	91.52 51	94.33 39	95.96 41	95.01 48	97.79 92	97.49 60

X-MVS		96.07 23	96.33 25	95.77 26	98.94 15	98.66 11	97.94 21	95.41 25	95.12 25	88.03 46	93.00 28	96.06 26	95.85 24	96.65 24	96.35 25	98.47 18	98.48 24

SteuartSystems-ACMMP		97.10 11	97.49 7	96.65 15	98.97 13	98.95 6	98.43 6	95.96 13	95.12 25	91.46 25	96.85 6	97.60 14	96.37 22	97.76 4	97.16 7	98.68 6	98.97 8
Skip Steuart: Steuart Systems R&D Blog.
3Dnovator+	90.56 5	95.06 32	94.56 37	95.65 28	98.11 27	98.15 37	97.19 29	91.59 47	95.11 27	93.23 18	81.99 90	94.71 37	95.43 30	96.48 29	96.88 14	98.35 36	98.63 13

DeepC-MVS	92.10 3	95.22 31	94.77 35	95.75 27	97.77 33	98.54 22	97.63 25	95.96 13	95.07 28	88.85 42	85.35 61	91.85 47	95.82 25	96.88 22	97.10 8	98.44 24	98.63 13
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
PLC	90.69 4	94.32 40	92.99 49	95.87 25	97.91 29	96.49 84	95.95 46	94.12 30	94.94 29	94.09 9	85.90 57	90.77 54	95.58 28	94.52 73	93.32 83	97.55 108	95.00 146
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
TAPA-MVS	90.35 6	93.69 45	93.52 43	93.90 43	96.89 47	97.62 53	96.15 40	91.67 46	94.94 29	85.97 61	87.72 50	91.96 46	94.40 36	93.76 89	93.06 96	98.30 47	95.58 125
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
ACMMP		95.54 28	95.49 31	95.61 29	98.27 26	98.53 23	97.16 30	94.86 27	94.88 31	89.34 38	95.36 23	91.74 48	95.50 29	95.51 47	94.16 59	98.50 15	98.22 34
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
APD-MVS		97.12 9	97.05 14	97.19 5	99.04 7	98.63 16	98.45 5	96.54 3	94.81 32	93.50 11	96.10 14	97.40 17	96.81 10	97.05 17	96.82 15	98.80 5	98.56 17
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
3Dnovator	90.28 7	94.70 38	94.34 40	95.11 31	98.06 28	98.21 34	96.89 33	91.03 53	94.72 33	91.45 26	82.87 81	93.10 43	94.61 35	96.24 38	97.08 9	98.63 9	98.16 37

MSLP-MVS++		96.05 24	95.63 28	96.55 18	98.33 25	98.17 36	96.94 32	94.61 29	94.70 34	94.37 6	89.20 45	95.96 29	96.81 10	95.57 46	97.33 5	98.24 59	98.47 25

CPTT-MVS		95.54 28	95.07 33	96.10 22	97.88 31	97.98 43	97.92 22	94.86 27	94.56 35	92.16 23	91.01 38	95.71 31	96.97 7	94.56 72	93.50 78	96.81 157	98.14 39

MP-MVS		96.56 18	96.72 19	96.37 21	98.93 16	98.48 25	98.04 18	95.55 19	94.32 36	90.95 32	95.88 17	97.02 20	96.29 23	96.77 23	96.01 35	98.47 18	98.56 17

MVS_111021_HR		94.84 35	95.91 27	93.60 47	97.35 39	98.46 28	95.08 54	91.19 50	94.18 37	85.97 61	95.38 22	92.56 45	93.61 48	96.61 26	96.25 30	98.40 31	97.92 47

PHI-MVS		95.86 25	96.93 18	94.61 37	97.60 37	98.65 15	96.49 36	93.13 35	94.07 38	87.91 49	97.12 4	97.17 19	93.90 46	96.46 30	96.93 12	98.64 8	98.10 43

canonicalmvs		93.08 47	93.09 47	93.07 56	94.24 72	97.86 45	95.45 52	87.86 99	94.00 39	87.47 52	88.32 48	82.37 88	95.13 32	93.96 88	96.41 23	98.27 54	98.73 10

train_agg		96.15 22	96.64 22	95.58 30	98.44 23	98.03 40	98.14 16	95.40 26	93.90 40	87.72 50	96.26 12	98.10 7	95.75 26	96.25 37	95.45 43	98.01 83	98.47 25

abl_6				94.78 33	97.46 38	97.99 42	95.76 47	91.80 44	93.72 41	91.25 27	91.33 37	96.47 24	94.28 40			98.14 67	97.39 64

AdaColmap		95.02 33	93.71 42	96.54 19	98.51 22	97.76 49	96.69 35	95.94 15	93.72 41	93.50 11	89.01 46	90.53 57	96.49 19	94.51 74	93.76 69	98.07 77	96.69 82

PGM-MVS		96.16 21	96.33 25	95.95 23	99.04 7	98.63 16	98.32 10	92.76 37	93.42 43	90.49 35	96.30 11	95.31 35	96.71 15	96.46 30	96.02 34	98.38 34	98.19 36

CANet		94.85 34	94.92 34	94.78 33	97.25 42	98.52 24	97.20 28	91.81 43	93.25 44	91.06 29	86.29 53	94.46 38	92.99 54	97.02 18	96.68 16	98.34 38	98.20 35

diffmvs		91.35 66	91.81 65	90.82 82	92.80 108	95.62 98	93.74 80	86.04 108	93.17 45	85.82 65	84.48 71	89.74 62	90.23 85	90.49 145	92.45 103	96.29 168	96.72 80

CLD-MVS		92.50 53	91.96 64	93.13 53	93.93 84	96.24 92	95.69 48	88.77 83	92.92 46	89.01 41	88.19 49	81.74 92	93.13 53	93.63 93	93.08 94	98.23 60	97.91 49
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
CDPH-MVS		94.80 37	95.50 30	93.98 42	98.34 24	98.06 39	97.41 26	93.23 34	92.81 47	82.98 82	92.51 30	94.82 36	93.53 49	96.08 40	96.30 29	98.42 27	97.94 45

MVS_0304		94.30 41	94.68 36	93.86 45	96.33 53	98.48 25	97.41 26	91.20 49	92.75 48	86.96 56	86.03 56	93.81 41	92.64 58	96.89 21	96.54 21	98.61 10	98.24 33

QAPM		94.13 42	94.33 41	93.90 43	97.82 32	98.37 32	96.47 37	90.89 54	92.73 49	85.63 67	85.35 61	93.87 39	94.17 41	95.71 45	95.90 36	98.40 31	98.42 28

LS3D		91.97 58	90.98 71	93.12 54	97.03 46	97.09 71	95.33 53	95.59 18	92.47 50	79.26 103	81.60 93	82.77 84	94.39 37	94.28 77	94.23 57	97.14 121	94.45 151

HQP-MVS		92.39 54	92.49 55	92.29 61	95.65 59	95.94 96	95.64 50	92.12 41	92.46 51	79.65 101	91.97 33	82.68 85	92.92 56	93.47 100	92.77 99	97.74 96	98.12 41

ACMM	88.76 10	91.70 64	90.43 73	93.19 52	95.56 60	95.14 101	93.35 88	91.48 48	92.26 52	87.12 54	84.02 76	79.34 99	93.99 43	94.07 83	92.68 101	97.62 107	95.50 126
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
PVSNet_BlendedMVS		92.80 49	92.44 56	93.23 50	96.02 55	97.83 47	93.74 80	90.58 55	91.86 53	90.69 33	85.87 59	82.04 89	90.01 86	96.39 33	95.26 45	98.34 38	97.81 52

PVSNet_Blended		92.80 49	92.44 56	93.23 50	96.02 55	97.83 47	93.74 80	90.58 55	91.86 53	90.69 33	85.87 59	82.04 89	90.01 86	96.39 33	95.26 45	98.34 38	97.81 52

ACMP	89.13 9	92.03 57	91.70 66	92.41 60	94.92 68	96.44 87	93.95 75	89.96 60	91.81 55	85.48 71	90.97 39	79.12 100	92.42 60	93.28 105	92.55 102	97.76 94	97.74 55
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
EPNet		93.92 43	94.40 38	93.36 49	97.89 30	96.55 81	96.08 42	92.14 40	91.65 56	89.16 40	94.07 26	90.17 61	87.78 107	95.24 49	94.97 49	97.09 124	98.15 38
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
NP-MVS									91.63 57

MVS_Test		91.81 62	92.19 60	91.37 76	93.24 103	96.95 73	94.43 59	86.25 107	91.45 58	83.45 80	86.31 52	85.15 74	92.93 55	93.99 84	94.71 52	97.92 88	96.77 79

tfpn1000		89.30 92	89.72 82	88.81 101	93.83 92	96.50 83	91.53 118	88.74 84	91.20 59	76.74 113	84.96 67	75.44 120	83.50 169	93.63 93	92.42 105	98.51 13	93.88 160

tfpn_ndepth		89.72 82	89.91 80	89.49 93	93.56 101	96.67 78	92.34 100	89.25 76	90.85 60	78.68 106	84.25 75	77.39 111	84.84 154	93.58 95	92.76 100	98.30 47	93.90 159

MAR-MVS		92.71 51	92.63 52	92.79 58	97.70 35	97.15 68	93.75 79	87.98 93	90.71 61	85.76 66	86.28 54	86.38 67	94.35 38	94.95 54	95.49 42	97.22 116	97.44 62
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
CANet_DTU		90.74 71	92.93 50	88.19 107	94.36 71	96.61 79	94.34 62	84.66 127	90.66 62	68.75 174	90.41 41	86.89 65	89.78 88	95.46 48	94.87 50	97.25 115	95.62 123

UGNet		91.52 65	93.41 45	89.32 96	94.13 74	97.15 68	91.83 114	89.01 80	90.62 63	85.86 64	86.83 51	91.73 49	77.40 196	94.68 68	94.43 54	97.71 98	98.40 30
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
PMMVS		89.88 80	91.19 70	88.35 105	89.73 138	91.97 180	90.62 122	81.92 160	90.57 64	80.58 96	92.16 31	86.85 66	91.17 74	92.31 114	91.35 127	96.11 171	93.11 170

LGP-MVS_train		91.83 61	92.04 63	91.58 67	95.46 63	96.18 94	95.97 45	89.85 61	90.45 65	77.76 107	91.92 34	80.07 97	92.34 61	94.27 78	93.47 79	98.11 71	97.90 50

RPSCF		89.68 83	89.24 84	90.20 87	92.97 106	92.93 152	92.30 101	87.69 101	90.44 66	85.12 74	91.68 35	85.84 73	90.69 80	87.34 191	86.07 195	92.46 205	90.37 195

DI_MVS_plusplus_trai		91.05 68	90.15 76	92.11 62	92.67 111	96.61 79	96.03 43	88.44 87	90.25 67	85.92 63	73.73 134	84.89 76	91.92 62	94.17 81	94.07 63	97.68 102	97.31 66

EPP-MVSNet		92.13 56	93.06 48	91.05 80	93.66 97	97.30 59	92.18 106	87.90 95	90.24 68	83.63 78	86.14 55	90.52 59	90.76 79	94.82 65	94.38 55	98.18 65	97.98 44

CHOSEN 280x420		90.77 70	92.14 61	89.17 98	93.86 91	92.81 156	93.16 89	80.22 181	90.21 69	84.67 77	89.89 42	91.38 52	90.57 82	94.94 55	92.11 112	92.52 204	93.65 163

EPMVS		85.77 129	86.24 123	85.23 152	92.76 109	93.78 119	89.91 150	73.60 209	90.19 70	74.22 121	82.18 89	78.06 105	87.55 110	85.61 200	85.38 201	93.32 197	88.48 205

PatchmatchNet		85.70 130	86.65 117	84.60 164	91.79 117	93.40 134	89.27 161	73.62 208	90.19 70	72.63 129	82.74 84	81.93 91	87.64 108	84.99 202	84.29 207	92.64 202	89.00 201
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
thresconf0.02		88.86 97	88.70 93	89.04 99	93.59 100	96.40 88	92.97 92	89.75 65	90.16 72	74.34 120	84.41 72	71.00 132	85.16 150	93.32 103	93.12 93	98.41 29	92.52 180

MVSTER		91.73 63	91.61 67	91.86 64	93.18 104	94.56 103	94.37 60	87.90 95	90.16 72	88.69 44	89.23 44	81.28 94	88.92 101	95.75 44	93.95 66	98.12 69	96.37 96

GBi-Net		90.21 76	90.11 77	90.32 84	88.66 147	93.65 124	94.25 69	85.78 112	90.03 74	85.56 68	77.38 110	86.13 68	89.38 92	93.97 85	94.16 59	98.31 44	95.47 127

test1		90.21 76	90.11 77	90.32 84	88.66 147	93.65 124	94.25 69	85.78 112	90.03 74	85.56 68	77.38 110	86.13 68	89.38 92	93.97 85	94.16 59	98.31 44	95.47 127

FMVSNet3		90.19 78	90.06 79	90.34 83	88.69 146	93.85 116	94.58 58	85.78 112	90.03 74	85.56 68	77.38 110	86.13 68	89.22 95	93.29 104	94.36 56	98.20 63	95.40 131

ADS-MVSNet		84.08 161	84.95 137	83.05 184	91.53 123	91.75 183	88.16 174	70.70 218	89.96 77	69.51 167	78.83 103	76.97 114	86.29 126	84.08 209	84.60 205	92.13 210	88.48 205

tfpnview11		88.80 98	89.21 85	88.31 106	93.70 96	96.24 92	92.35 99	89.11 77	89.90 78	72.14 133	85.12 64	73.93 121	84.20 160	93.75 90	92.85 97	98.38 34	92.68 178

PatchMatch-RL		90.30 75	88.93 89	91.89 63	95.41 66	95.68 97	90.94 119	88.67 85	89.80 79	86.95 57	85.90 57	72.51 125	92.46 59	93.56 98	92.18 110	96.93 142	92.89 171

CHOSEN 1792x2688		88.57 103	87.82 104	89.44 95	95.46 63	96.89 75	93.74 80	85.87 111	89.63 80	77.42 110	61.38 207	83.31 80	88.80 104	93.44 101	93.16 91	95.37 187	96.95 73

MSDG		90.42 74	88.25 98	92.94 57	96.67 50	94.41 109	93.96 73	92.91 36	89.59 81	86.26 60	76.74 116	80.92 95	90.43 83	92.60 111	92.08 114	97.44 112	91.41 184

DELS-MVS		93.71 44	93.47 44	94.00 40	96.82 48	98.39 31	96.80 34	91.07 52	89.51 82	89.94 37	83.80 77	89.29 63	90.95 77	97.32 10	97.65 3	98.42 27	98.32 31
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
MDTV_nov1_ep13		86.64 121	87.50 111	85.65 144	90.73 131	93.69 123	89.96 148	78.03 193	89.48 83	76.85 112	84.92 68	82.42 87	86.14 132	86.85 196	86.15 194	92.17 208	88.97 202

OpenMVS	88.18 11	92.51 52	91.61 67	93.55 48	97.74 34	98.02 41	95.66 49	90.46 57	89.14 84	86.50 59	75.80 124	90.38 60	92.69 57	94.99 52	95.30 44	98.27 54	97.63 56

EPNet_dtu		88.32 106	90.61 72	85.64 145	96.79 49	92.27 170	92.03 112	90.31 58	89.05 85	65.44 196	89.43 43	85.90 72	74.22 204	92.76 107	92.09 113	95.02 192	92.76 175
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
Vis-MVSNet (Re-imp)		90.54 73	92.76 51	87.94 113	93.73 95	96.94 74	92.17 108	87.91 94	88.77 86	76.12 116	83.68 78	90.80 53	79.49 191	96.34 35	96.35 25	98.21 62	96.46 93

GG-mvs-BLEND		62.84 223	90.21 74	30.91 234	0.57 239	94.45 107	86.99 184	0.34 238	88.71 87	0.98 241	81.55 95	91.58 50	0.86 238	92.66 109	91.43 126	95.73 177	91.11 188

IS_MVSNet		91.87 60	93.35 46	90.14 89	94.09 78	97.73 50	93.09 90	88.12 91	88.71 87	79.98 99	84.49 70	90.63 56	87.49 112	97.07 16	96.96 11	98.07 77	97.88 51

DWT-MVSNet_training		86.83 117	84.44 142	89.61 91	92.75 110	93.82 117	91.66 116	82.85 147	88.57 89	87.48 51	79.00 102	64.24 192	88.82 103	85.18 201	87.50 191	94.07 195	92.79 172

FMVSNet2		89.61 84	89.14 86	90.16 88	88.66 147	93.65 124	94.25 69	85.44 119	88.57 89	84.96 76	73.53 136	83.82 78	89.38 92	94.23 79	94.68 53	98.31 44	95.47 127

PVSNet_Blended_VisFu		91.92 59	92.39 58	91.36 77	95.45 65	97.85 46	92.25 103	89.54 73	88.53 91	87.47 52	79.82 99	90.53 57	85.47 148	96.31 36	95.16 47	97.99 85	98.56 17

USDC		86.73 120	85.96 128	87.63 118	91.64 119	93.97 115	92.76 93	84.58 129	88.19 92	70.67 150	80.10 98	67.86 153	89.43 90	91.81 121	89.77 176	96.69 161	90.05 198

tpmrst		83.72 167	83.45 151	84.03 172	92.21 112	91.66 184	88.74 170	73.58 210	88.14 93	72.67 128	77.37 113	72.11 128	86.34 125	82.94 212	82.05 214	90.63 217	89.86 199

CostFormer		86.78 119	86.05 124	87.62 119	92.15 113	93.20 142	91.55 117	75.83 200	88.11 94	85.29 73	81.76 91	76.22 117	87.80 106	84.45 206	85.21 202	93.12 198	93.42 166

tpmp4_e23		85.67 131	84.28 144	87.30 121	91.96 116	92.00 179	92.06 111	76.27 198	87.95 95	83.59 79	76.97 115	70.88 133	87.52 111	84.80 205	84.73 204	92.40 206	92.61 179

FC-MVSNet-train		90.55 72	90.19 75	90.97 81	93.78 93	95.16 100	92.11 110	88.85 82	87.64 96	83.38 81	84.36 74	78.41 103	89.53 89	94.69 67	93.15 92	98.15 66	97.92 47

Effi-MVS+		89.79 81	89.83 81	89.74 90	92.98 105	96.45 86	93.48 87	84.24 131	87.62 97	76.45 114	81.76 91	77.56 110	93.48 50	94.61 70	93.59 71	97.82 91	97.22 67

tfpn_n400		88.58 101	88.91 90	88.19 107	93.63 98	96.34 90	92.22 104	89.04 78	87.37 98	72.14 133	85.12 64	73.93 121	84.04 165	93.65 91	93.20 88	98.09 73	92.77 173

tfpnconf		88.58 101	88.91 90	88.19 107	93.63 98	96.34 90	92.22 104	89.04 78	87.37 98	72.14 133	85.12 64	73.93 121	84.04 165	93.65 91	93.20 88	98.09 73	92.77 173

PCF-MVS	90.19 8	92.98 48	92.07 62	94.04 39	96.39 52	97.87 44	96.03 43	95.47 24	87.16 100	85.09 75	84.81 69	93.21 42	93.46 51	91.98 120	91.98 117	97.78 93	97.51 59
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
COLMAP_ROB	84.39 15	87.61 111	86.03 125	89.46 94	95.54 62	94.48 106	91.77 115	90.14 59	87.16 100	75.50 117	73.41 139	76.86 115	87.33 114	90.05 153	89.76 177	96.48 163	90.46 194
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
Fast-Effi-MVS+		88.56 104	87.99 100	89.22 97	91.56 121	95.21 99	92.29 102	82.69 149	86.82 102	77.73 108	76.24 122	73.39 124	93.36 52	94.22 80	93.64 70	97.65 104	96.43 94

pmmvs4		86.00 128	84.28 144	88.00 111	87.80 158	92.01 178	89.94 149	84.91 125	86.79 103	80.98 93	73.41 139	66.34 169	88.12 105	89.31 169	88.90 187	96.24 170	93.20 169

MS-PatchMatch		87.63 110	87.61 108	87.65 117	93.95 82	94.09 113	92.60 96	81.52 165	86.64 104	76.41 115	73.46 138	85.94 71	85.01 153	92.23 117	90.00 170	96.43 165	90.93 190

HyFIR lowres test		87.87 109	86.42 121	89.57 92	95.56 60	96.99 72	92.37 98	84.15 133	86.64 104	77.17 111	57.65 212	83.97 77	91.08 76	92.09 119	92.44 104	97.09 124	95.16 143

FC-MVSNet-test		86.15 124	89.10 87	82.71 190	89.83 136	93.18 144	87.88 177	84.69 126	86.54 106	62.18 206	82.39 88	83.31 80	74.18 205	92.52 112	91.86 118	97.50 110	93.88 160

IterMVS-LS		88.60 100	88.45 94	88.78 102	92.02 115	92.44 168	92.00 113	83.57 140	86.52 107	78.90 105	78.61 106	81.34 93	89.12 96	90.68 141	93.18 90	97.10 123	96.35 97
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
tmp_tt				50.24 230	68.55 230	46.86 237	48.90 235	18.28 235	86.51 108	68.32 177	70.19 152	65.33 181	26.69 235	74.37 225	66.80 229	70.72 234

IB-MVS	85.10 14	87.98 107	87.97 101	87.99 112	94.55 70	96.86 76	84.52 199	88.21 90	86.48 109	88.54 45	74.41 133	77.74 108	74.10 206	89.65 159	92.85 97	98.06 79	97.80 54
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
tpm cat1		84.13 159	81.99 186	86.63 129	91.74 118	91.50 187	90.68 121	75.69 201	86.12 110	85.44 72	72.39 143	70.72 134	85.16 150	80.89 218	81.56 217	91.07 214	90.71 192

tfpn111		90.16 79	88.99 88	91.52 71	93.90 85	97.26 60	94.31 64	89.75 65	85.87 111	81.10 90	84.41 72	70.38 136	91.76 65	94.92 56	93.51 72	98.29 51	96.61 85

conf0.01		89.34 91	87.39 115	91.61 66	93.88 89	97.34 58	94.31 64	89.82 64	85.87 111	81.53 86	77.93 109	66.15 176	91.76 65	94.90 63	93.51 72	98.32 43	96.05 111

conf0.002		89.25 95	87.21 116	91.62 65	93.87 90	97.35 57	94.31 64	89.83 62	85.87 111	81.62 85	78.72 105	63.89 193	91.76 65	94.90 63	93.98 65	98.33 42	95.77 118

conf200view11		89.55 85	87.86 102	91.52 71	93.90 85	97.26 60	94.31 64	89.75 65	85.87 111	81.10 90	76.46 118	70.38 136	91.76 65	94.92 56	93.51 72	98.29 51	96.61 85

thres100view900		89.36 89	87.61 108	91.39 74	93.90 85	96.86 76	94.35 61	89.66 72	85.87 111	81.15 88	76.46 118	70.38 136	91.17 74	94.09 82	93.43 80	98.13 68	96.16 106

tfpn200view9		89.55 85	87.86 102	91.53 69	93.90 85	97.26 60	94.31 64	89.74 68	85.87 111	81.15 88	76.46 118	70.38 136	91.76 65	94.92 56	93.51 72	98.28 53	96.61 85

thres400		89.40 88	87.58 110	91.53 69	94.06 80	97.21 65	94.19 72	89.83 62	85.69 117	81.08 92	75.50 126	69.76 142	91.80 63	94.79 66	93.51 72	98.20 63	96.60 90

view600		89.29 93	87.48 112	91.41 73	94.10 77	97.21 65	93.96 73	89.70 71	85.67 118	80.75 94	75.29 128	69.35 143	91.70 70	94.92 56	93.23 85	98.26 58	96.69 82

thres200		89.49 87	87.72 105	91.55 68	93.95 82	97.25 63	94.34 62	89.74 68	85.66 119	81.18 87	76.12 123	70.19 141	91.80 63	94.92 56	93.51 72	98.27 54	96.40 95

test0.0.03 1		85.58 132	87.69 107	83.11 181	91.22 124	92.54 163	85.60 198	83.62 138	85.66 119	67.84 181	82.79 83	79.70 98	73.51 208	91.15 132	90.79 131	96.88 153	91.23 187

thres600view7		89.28 94	87.47 113	91.39 74	94.12 76	97.25 63	93.94 76	89.74 68	85.62 121	80.63 95	75.24 131	69.33 144	91.66 71	94.92 56	93.23 85	98.27 54	96.72 80

view800		89.21 96	87.44 114	91.27 78	94.13 74	97.18 67	93.74 80	89.53 74	85.60 122	80.34 97	75.29 128	68.89 145	91.57 72	94.97 53	93.36 81	98.34 38	96.79 78

tfpn		88.67 99	86.57 119	91.12 79	94.14 73	97.15 68	93.51 85	89.37 75	85.49 123	79.91 100	75.26 130	62.24 199	91.39 73	95.00 51	93.95 66	98.41 29	96.88 76

dps		85.00 139	83.21 162	87.08 123	90.73 131	92.55 162	89.34 160	75.29 202	84.94 124	87.01 55	79.27 101	67.69 154	87.27 115	84.22 208	83.56 208	92.83 200	90.25 196

CR-MVSNet		85.48 134	86.29 122	84.53 166	91.08 128	92.10 172	89.18 163	73.30 213	84.75 125	71.08 145	73.12 142	77.91 107	86.27 127	91.48 125	90.75 134	96.27 169	93.94 157

RPMNet		84.82 142	85.90 129	83.56 176	91.10 126	92.10 172	88.73 171	71.11 217	84.75 125	68.79 173	73.56 135	77.62 109	85.33 149	90.08 152	89.43 182	96.32 167	93.77 162

test-LLR		86.88 116	88.28 96	85.24 151	91.22 124	92.07 174	87.41 180	83.62 138	84.58 127	69.33 168	83.00 79	82.79 82	84.24 158	92.26 115	89.81 174	95.64 180	93.44 164

TESTMET0.1,1		86.11 126	88.28 96	83.59 175	87.80 158	92.07 174	87.41 180	77.12 195	84.58 127	69.33 168	83.00 79	82.79 82	84.24 158	92.26 115	89.81 174	95.64 180	93.44 164

DU-MVS		86.12 125	84.81 139	87.66 116	87.77 160	93.78 119	90.15 141	87.87 97	84.40 129	73.45 126	70.59 148	64.82 187	88.95 99	90.14 148	92.33 106	97.76 94	97.62 57

NR-MVSNet		85.46 135	84.54 141	86.52 130	88.33 152	93.78 119	90.45 125	87.87 97	84.40 129	71.61 138	70.59 148	62.09 202	82.79 172	91.75 122	91.75 120	98.10 72	97.44 62

UniMVSNet (Re)		86.22 123	85.46 136	87.11 122	88.34 151	94.42 108	89.65 158	87.10 106	84.39 131	74.61 119	70.41 151	68.10 151	85.10 152	91.17 131	91.79 119	97.84 90	97.94 45

test-mter		86.09 127	88.38 95	83.43 178	87.89 157	92.61 160	86.89 185	77.11 196	84.30 132	68.62 176	82.57 86	82.45 86	84.34 157	92.40 113	90.11 168	95.74 176	94.21 155

FMVSNet5		84.47 150	84.72 140	84.18 170	83.30 212	88.43 207	88.09 175	79.42 187	84.25 133	74.14 123	73.15 141	78.74 101	83.65 167	91.19 130	91.19 128	96.46 164	86.07 211

Vis-MVSNet		89.36 89	91.49 69	86.88 125	92.10 114	97.60 54	92.16 109	85.89 110	84.21 134	75.20 118	82.58 85	87.13 64	77.40 196	95.90 43	95.63 40	98.51 13	97.36 65
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
UniMVSNet_NR-MVSNet		86.80 118	85.86 131	87.89 115	88.17 153	94.07 114	90.15 141	88.51 86	84.20 135	73.45 126	72.38 144	70.30 140	88.95 99	90.25 147	92.21 109	98.12 69	97.62 57

IterMVS		85.25 137	86.49 120	83.80 173	90.42 134	90.77 195	90.02 146	78.04 192	84.10 136	66.27 192	77.28 114	78.41 103	83.01 170	90.88 134	89.72 178	95.04 191	94.24 153
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
PatchT		83.86 164	85.51 135	81.94 197	88.41 150	91.56 186	78.79 212	71.57 216	84.08 137	71.08 145	70.62 147	76.13 118	86.27 127	91.48 125	90.75 134	95.52 185	93.94 157

Baseline_NR-MVSNet		85.28 136	83.42 153	87.46 120	87.77 160	90.80 194	89.90 152	87.69 101	83.93 138	74.16 122	64.72 197	66.43 166	87.48 113	90.14 148	90.83 130	97.73 97	97.11 70

Fast-Effi-MVS+-dtu		86.25 122	87.70 106	84.56 165	90.37 135	93.70 122	90.54 124	78.14 191	83.50 139	65.37 197	81.59 94	75.83 119	86.09 137	91.70 123	91.70 121	96.88 153	95.84 117

Effi-MVS+-dtu		87.51 112	88.13 99	86.77 127	91.10 126	94.90 102	90.91 120	82.67 150	83.47 140	71.55 139	81.11 96	77.04 113	89.41 91	92.65 110	91.68 123	95.00 193	96.09 109

ACMH+	85.75 12	87.19 115	86.02 126	88.56 104	93.42 102	94.41 109	89.91 150	87.66 103	83.45 141	72.25 131	76.42 121	71.99 129	90.78 78	89.86 154	90.94 129	97.32 113	95.11 145

TranMVSNet+NR-MVSNet		85.57 133	84.41 143	86.92 124	87.67 163	93.34 136	90.31 132	88.43 88	83.07 142	70.11 160	69.99 155	65.28 182	86.96 118	89.73 156	92.27 107	98.06 79	97.17 69

OPM-MVS		91.08 67	89.34 83	93.11 55	96.18 54	96.13 95	96.39 38	92.39 38	82.97 143	81.74 84	82.55 87	80.20 96	93.97 45	94.62 69	93.23 85	98.00 84	95.73 120

TDRefinement		84.97 140	83.39 154	86.81 126	92.97 106	94.12 112	92.18 106	87.77 100	82.78 144	71.31 142	68.43 160	68.07 152	81.10 186	89.70 158	89.03 186	95.55 184	91.62 182

tpm		83.16 179	83.64 148	82.60 192	90.75 130	91.05 190	88.49 172	73.99 206	82.36 145	67.08 187	78.10 108	68.79 146	84.17 161	85.95 199	85.96 197	91.09 213	93.23 168

UA-Net		90.81 69	92.58 53	88.74 103	94.87 69	97.44 55	92.61 95	88.22 89	82.35 146	78.93 104	85.20 63	95.61 32	79.56 190	96.52 28	96.57 20	98.23 60	94.37 152

TinyColmap		84.04 162	82.01 185	86.42 131	90.87 129	91.84 181	88.89 169	84.07 134	82.11 147	69.89 164	71.08 146	60.81 208	89.04 97	90.52 143	89.19 184	95.76 175	88.50 204

ACMH	85.51 13	87.31 114	86.59 118	88.14 110	93.96 81	94.51 105	89.00 167	87.99 92	81.58 148	70.15 157	78.41 107	71.78 130	90.60 81	91.30 129	91.99 116	97.17 118	96.58 91
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
MIMVSNet		82.97 183	84.00 147	81.77 199	82.23 213	92.25 171	87.40 182	72.73 215	81.48 149	69.55 166	68.79 159	72.42 126	81.82 181	92.23 117	92.25 108	96.89 150	88.61 203

v18		84.21 156	82.90 168	85.74 142	87.63 164	89.75 196	90.56 123	80.82 171	81.42 150	72.24 132	67.16 165	67.23 156	86.27 127	89.25 173	90.24 153	96.92 146	95.27 136

v16		84.14 158	82.86 170	85.64 145	87.61 165	89.71 198	90.36 126	80.70 173	81.36 151	71.99 136	66.91 172	67.19 157	86.23 130	89.32 167	90.25 150	96.94 139	95.29 134

FMVSNet1		87.33 113	86.00 127	88.89 100	87.13 190	92.83 155	93.08 91	84.46 130	81.35 152	82.20 83	66.33 182	77.96 106	88.96 98	93.97 85	94.16 59	97.54 109	95.38 132

GA-MVS		85.08 138	85.65 133	84.42 167	89.77 137	94.25 111	89.26 162	84.62 128	81.19 153	62.25 205	75.72 125	68.44 149	84.14 162	93.57 97	91.68 123	96.49 162	94.71 148

v17		84.10 160	82.83 171	85.57 147	87.58 166	89.72 197	90.30 135	80.70 173	81.00 154	71.72 137	67.01 167	67.24 155	86.19 131	89.32 167	90.25 150	96.95 137	95.29 134

conf0.05thres1000		87.90 108	85.88 130	90.26 86	93.74 94	96.39 89	92.67 94	88.94 81	80.97 155	77.71 109	70.15 153	68.40 150	90.42 84	94.46 76	93.29 84	98.09 73	97.49 60

LP		77.28 208	76.57 210	78.12 206	88.17 153	88.06 209	80.85 209	68.35 225	80.78 156	61.49 208	57.59 213	61.80 203	77.59 195	81.45 217	82.34 213	92.25 207	83.96 217

testgi		81.94 193	84.09 146	79.43 203	89.53 141	90.83 193	82.49 205	81.75 163	80.59 157	59.46 212	82.82 82	65.75 177	67.97 210	90.10 151	89.52 181	95.39 186	89.03 200

v6		84.67 143	83.36 156	86.20 132	87.53 168	93.49 128	90.34 127	85.16 124	80.58 158	71.13 144	66.97 169	67.10 160	86.11 136	89.25 173	90.22 159	96.93 142	95.89 114

v1neww		84.65 144	83.34 158	86.18 134	87.53 168	93.49 128	90.32 128	85.17 122	80.57 159	71.02 148	66.93 170	67.04 162	86.13 134	89.26 170	90.23 156	96.93 142	95.88 115

v7new		84.65 144	83.34 158	86.18 134	87.53 168	93.49 128	90.32 128	85.17 122	80.57 159	71.02 148	66.93 170	67.04 162	86.13 134	89.26 170	90.23 156	96.93 142	95.88 115

v8		84.45 151	83.30 160	85.80 140	87.53 168	92.95 150	90.31 132	82.46 154	80.46 161	71.43 140	66.99 168	67.16 159	86.14 132	89.26 170	90.22 159	96.94 139	96.06 110

CDS-MVSNet		88.34 105	88.71 92	87.90 114	90.70 133	94.54 104	92.38 97	86.02 109	80.37 162	79.42 102	79.30 100	83.43 79	82.04 178	93.39 102	94.01 64	96.86 155	95.93 113
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
V42		84.48 149	83.36 156	85.79 141	87.14 189	93.28 139	90.03 145	83.98 135	80.30 163	71.20 143	66.90 174	67.17 158	85.55 146	89.35 164	90.27 147	96.82 156	96.27 103

MDTV_nov1_ep13_2view		80.43 199	80.94 197	79.84 201	84.82 207	90.87 192	84.23 200	73.80 207	80.28 164	64.33 200	70.05 154	68.77 147	79.67 188	84.83 204	83.50 209	92.17 208	88.25 207

CVMVSNet		83.83 165	85.53 134	81.85 198	89.60 139	90.92 191	87.81 178	83.21 144	80.11 165	60.16 210	76.47 117	78.57 102	76.79 198	89.76 155	90.13 164	93.51 196	92.75 176

WR-MVS		83.14 180	83.38 155	82.87 186	87.55 167	93.29 138	86.36 190	84.21 132	80.05 166	66.41 191	66.91 172	66.92 164	75.66 202	88.96 181	90.56 139	97.05 126	96.96 72

PM-MVS		80.29 200	79.30 201	81.45 200	81.91 214	88.23 208	82.61 204	79.01 188	79.99 167	67.15 186	69.07 158	51.39 220	82.92 171	87.55 190	85.59 198	95.08 190	93.28 167

v2v482		84.51 147	83.05 164	86.20 132	87.25 186	93.28 139	90.22 138	85.40 120	79.94 168	69.78 165	67.74 163	65.15 184	87.57 109	89.12 179	90.55 140	96.97 133	95.60 124

v10		84.18 157	83.17 163	85.37 148	87.34 178	92.68 158	90.32 128	81.33 166	79.93 169	69.23 171	66.33 182	65.74 179	87.03 117	90.84 135	90.38 143	96.97 133	96.29 102

v7		84.37 155	83.23 161	85.69 143	87.34 178	93.19 143	90.32 128	83.10 146	79.88 170	69.33 168	66.33 182	65.75 177	87.06 116	90.83 136	90.38 143	96.97 133	96.26 104

v1141		84.40 152	83.00 166	86.03 136	87.41 174	93.42 131	90.28 136	85.53 116	79.58 171	70.12 159	66.62 178	66.27 173	85.94 138	89.16 176	90.19 161	96.89 150	95.73 120

divwei89l23v2f112		84.40 152	83.00 166	86.02 138	87.42 173	93.42 131	90.28 136	85.52 117	79.57 172	70.11 160	66.64 177	66.29 172	85.91 139	89.16 176	90.19 161	96.90 148	95.73 120

v11		83.72 167	82.61 174	85.02 157	87.34 178	89.56 203	89.89 153	79.92 184	79.55 173	69.21 172	66.36 181	65.48 180	86.84 120	91.43 128	90.51 142	96.92 146	95.37 133

V14		83.66 170	82.38 175	85.16 154	87.37 177	89.62 200	90.15 141	80.33 177	79.51 174	70.26 154	66.30 185	66.37 168	85.87 142	89.38 163	90.24 153	96.98 132	95.22 141

v1		84.40 152	83.01 165	86.03 136	87.41 174	93.42 131	90.31 132	85.52 117	79.51 174	70.13 158	66.66 176	66.40 167	85.89 140	89.15 178	90.19 161	96.89 150	95.74 119

v15		83.67 169	82.37 176	85.19 153	87.39 176	89.63 199	90.19 139	80.43 175	79.49 176	70.27 153	66.37 179	66.33 170	85.88 141	89.34 166	90.23 156	96.96 136	95.22 141

v12		83.59 173	82.32 179	85.07 156	87.32 184	89.57 202	89.87 155	80.19 182	79.46 177	70.19 155	66.05 187	66.23 175	85.84 143	89.44 161	90.26 149	97.01 129	95.26 137

V9		83.61 171	82.33 178	85.11 155	87.34 178	89.59 201	90.10 144	80.25 178	79.38 178	70.17 156	66.15 186	66.33 170	85.82 144	89.41 162	90.24 153	96.99 131	95.23 140

v13		83.55 175	82.29 180	85.01 158	87.31 185	89.55 204	89.89 153	80.13 183	79.34 179	69.93 163	65.92 190	66.25 174	85.80 145	89.45 160	90.27 147	97.01 129	95.25 139

SixPastTwentyTwo		83.12 181	83.44 152	82.74 189	87.71 162	93.11 148	82.30 206	82.33 155	79.24 180	64.33 200	78.77 104	62.75 196	84.11 163	88.11 184	87.89 189	95.70 178	94.21 155

v148		83.61 171	82.10 183	85.37 148	87.34 178	92.94 151	87.48 179	85.72 115	78.92 181	73.87 124	65.71 192	64.69 188	81.78 182	87.82 185	89.35 183	96.01 172	95.26 137

v1144		84.03 163	82.88 169	85.37 148	87.17 188	93.15 147	90.18 140	83.31 143	78.83 182	67.85 180	65.99 188	64.99 185	86.79 121	90.75 138	90.33 146	96.90 148	96.15 107

testpf		74.66 210	76.34 211	72.71 215	87.34 178	80.91 221	73.15 221	60.30 232	78.73 183	61.68 207	69.83 156	62.22 200	67.48 211	76.83 223	78.17 224	86.28 226	87.68 208

v1192		83.56 174	82.35 177	84.98 159	86.84 194	92.84 153	90.01 147	82.70 148	78.54 184	66.48 190	64.88 196	62.91 195	86.91 119	90.72 139	90.25 150	96.94 139	96.32 99

v1921920		83.30 178	82.09 184	84.70 162	86.59 197	92.67 159	89.82 156	82.23 157	78.32 185	65.76 194	64.64 198	62.35 198	86.78 122	90.34 146	90.02 169	97.02 128	96.31 101

N_pmnet		77.55 207	76.68 209	78.56 205	85.43 205	87.30 213	78.84 211	81.88 161	78.30 186	60.61 209	61.46 206	62.15 201	74.03 207	82.04 213	80.69 220	90.59 218	84.81 215

anonymousdsp		84.51 147	85.85 132	82.95 185	86.30 199	93.51 127	85.77 196	80.38 176	78.25 187	63.42 203	73.51 137	72.20 127	84.64 156	93.21 106	92.16 111	97.19 117	98.14 39

v144192		83.48 176	82.23 181	84.94 160	86.65 195	92.84 153	89.63 159	82.48 153	77.87 188	67.36 184	65.33 194	63.50 194	86.51 123	89.72 157	89.99 171	97.03 127	96.35 97

CP-MVSNet		83.11 182	82.15 182	84.23 169	87.20 187	92.70 157	86.42 189	83.53 141	77.83 189	67.67 182	66.89 175	60.53 210	82.47 175	89.23 175	90.65 138	98.08 76	97.20 68

DeepMVS_CX						71.82 230	68.37 225	48.05 234	77.38 190	46.88 230	65.77 191	47.03 228	67.48 211	64.27 231		76.89 233	76.72 224

WR-MVS_H		82.86 185	82.66 173	83.10 182	87.44 172	93.33 137	85.71 197	83.20 145	77.36 191	68.20 179	66.37 179	65.23 183	76.05 201	89.35 164	90.13 164	97.99 85	96.89 75

v1240		82.88 184	81.66 188	84.29 168	86.46 198	92.52 166	89.06 165	81.82 162	77.16 192	65.09 198	64.17 199	61.50 204	86.36 124	90.12 150	90.13 164	96.95 137	96.04 112

TAMVS		84.94 141	84.95 137	84.93 161	88.82 143	93.18 144	88.44 173	81.28 167	77.16 192	73.76 125	75.43 127	76.57 116	82.04 178	90.59 142	90.79 131	95.22 189	90.94 189

PEN-MVS		82.49 188	81.58 189	83.56 176	86.93 193	92.05 177	86.71 187	83.84 136	76.94 194	64.68 199	67.24 164	60.11 211	81.17 185	87.78 186	90.70 137	98.02 82	96.21 105

v7n		82.25 190	81.54 190	83.07 183	85.55 203	92.58 161	86.68 188	81.10 170	76.54 195	65.97 193	62.91 204	60.56 209	82.36 176	91.07 133	90.35 145	96.77 158	96.80 77

pmmvs5		83.37 177	82.68 172	84.18 170	87.13 190	93.18 144	86.74 186	82.08 158	76.48 196	67.28 185	71.26 145	62.70 197	84.71 155	90.77 137	90.12 167	97.15 119	94.24 153

PS-CasMVS		82.53 187	81.54 190	83.68 174	87.08 192	92.54 163	86.20 191	83.46 142	76.46 197	65.73 195	65.71 192	59.41 215	81.61 183	89.06 180	90.55 140	98.03 81	97.07 71

DTE-MVSNet		81.76 195	81.04 196	82.60 192	86.63 196	91.48 189	85.97 193	83.70 137	76.45 198	62.44 204	67.16 165	59.98 212	78.98 192	87.15 193	89.93 172	97.88 89	95.12 144

v52		82.11 191	81.50 192	82.82 188	84.59 209	92.51 167	85.96 195	80.24 179	76.38 199	66.83 189	63.12 202	64.62 190	82.56 173	87.70 187	89.55 179	96.73 159	96.61 85

V4		82.11 191	81.49 193	82.83 187	84.60 208	92.53 165	85.97 193	80.24 179	76.35 200	66.87 188	63.17 201	64.55 191	82.54 174	87.70 187	89.55 179	96.73 159	96.61 85

EU-MVSNet		78.43 203	80.25 200	76.30 209	83.81 211	87.27 214	80.99 207	79.52 186	76.01 201	54.12 219	70.44 150	64.87 186	67.40 213	86.23 198	85.54 200	91.95 211	91.41 184

new_pmnet		72.29 216	73.25 215	71.16 220	75.35 225	81.38 219	73.72 219	69.27 221	75.97 202	49.84 225	56.27 214	56.12 218	69.08 209	81.73 214	80.86 219	89.72 223	80.44 221

MVS-HIRNet		78.16 204	77.57 207	78.83 204	85.83 201	87.76 210	76.67 213	70.22 219	75.82 203	67.39 183	55.61 215	70.52 135	81.96 180	86.67 197	85.06 203	90.93 216	81.58 220

Anonymous20231206		78.09 205	78.11 205	78.07 207	85.19 206	89.17 205	80.99 207	81.24 169	75.46 204	58.25 214	54.78 219	59.90 213	66.73 214	88.94 182	88.26 188	96.01 172	90.25 196

pmmvs-eth3d		79.78 202	77.58 206	82.34 195	81.57 215	87.46 212	82.92 203	81.28 167	75.33 205	71.34 141	61.88 205	52.41 219	81.59 184	87.56 189	86.90 193	95.36 188	91.48 183

v748		81.57 197	80.68 199	82.60 192	85.55 203	92.07 174	83.57 201	82.06 159	74.64 206	69.97 162	63.11 203	61.46 205	78.09 194	87.30 192	89.88 173	96.37 166	96.32 99

pm-mvs1		84.55 146	83.46 150	85.82 139	88.16 155	93.39 135	89.05 166	85.36 121	74.03 207	72.43 130	65.08 195	71.11 131	82.30 177	93.48 99	91.70 121	97.64 105	95.43 130

testus		73.65 213	74.92 212	72.17 218	80.93 218	81.11 220	73.02 222	75.23 203	73.23 208	48.77 226	69.38 157	46.10 229	62.28 220	84.84 203	86.01 196	92.77 201	83.75 218

tfpnnormal		83.80 166	81.26 195	86.77 127	89.60 139	93.26 141	89.72 157	87.60 104	72.78 209	70.44 151	60.53 210	61.15 207	85.55 146	92.72 108	91.44 125	97.71 98	96.92 74

test2356		73.82 211	74.82 213	72.66 216	81.25 216	80.70 222	73.47 220	75.91 199	72.55 210	48.73 227	68.14 161	50.74 221	63.96 216	84.44 207	85.57 199	92.63 203	81.60 219

FPMVS		69.87 218	67.10 221	73.10 213	84.09 210	78.35 225	79.40 210	76.41 197	71.92 211	57.71 215	54.06 221	50.04 222	56.72 223	71.19 227	68.70 228	84.25 228	75.43 225

ambc			67.96 219		73.69 226	79.79 224	73.82 218		71.61 212	59.80 211	46.00 225	20.79 237	66.15 215	86.92 195	80.11 222	89.13 224	90.50 193

MDA-MVSNet-bldmvs		73.81 212	72.56 216	75.28 210	72.52 228	88.87 206	74.95 216	82.67 150	71.57 213	55.02 217	65.96 189	42.84 231	76.11 200	70.61 228	81.47 218	90.38 219	86.59 209

EG-PatchMatch MVS		81.70 196	81.31 194	82.15 196	88.75 144	93.81 118	87.14 183	78.89 189	71.57 213	64.12 202	61.20 209	68.46 148	76.73 199	91.48 125	90.77 133	97.28 114	91.90 181

TransMVSNet (Re)		82.67 186	80.93 198	84.69 163	88.71 145	91.50 187	87.90 176	87.15 105	71.54 215	68.24 178	63.69 200	64.67 189	78.51 193	91.65 124	90.73 136	97.64 105	92.73 177

test20.03		76.41 209	78.49 204	73.98 211	85.64 202	87.50 211	75.89 214	80.71 172	70.84 216	51.07 224	68.06 162	61.40 206	54.99 225	88.28 183	87.20 192	95.58 183	86.15 210

CMPMVS	61.19 17	79.86 201	77.46 208	82.66 191	91.54 122	91.82 182	83.25 202	81.57 164	70.51 217	68.64 175	59.89 211	66.77 165	79.63 189	84.00 210	84.30 206	91.34 212	84.89 214
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
Gipuma		58.52 226	56.17 227	61.27 225	67.14 231	58.06 234	52.16 234	68.40 224	69.00 218	45.02 231	22.79 233	20.57 238	55.11 224	76.27 224	79.33 223	79.80 231	67.16 231
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
new-patchmatchnet		72.32 215	71.09 217	73.74 212	81.17 217	84.86 217	72.21 223	77.48 194	68.32 219	54.89 218	55.10 217	49.31 224	63.68 218	79.30 220	76.46 225	93.03 199	84.32 216

MIMVSNet1		73.19 214	73.70 214	72.60 217	65.42 232	86.69 215	75.56 215	79.65 185	67.87 220	55.30 216	45.24 228	56.41 217	63.79 217	86.98 194	87.66 190	95.85 174	85.04 213

LTVRE_ROB	81.71 16	82.44 189	81.84 187	83.13 180	89.01 142	92.99 149	88.90 168	82.32 156	66.26 221	54.02 220	74.68 132	59.62 214	88.87 102	90.71 140	92.02 115	95.68 179	96.62 84
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
pmmvs6		80.90 198	78.77 202	83.38 179	85.84 200	91.61 185	86.01 192	82.54 152	64.17 222	70.43 152	54.14 220	67.06 161	80.73 187	90.50 144	89.17 185	94.74 194	94.75 147

gm-plane-assit		77.65 206	78.50 203	76.66 208	87.96 156	85.43 216	64.70 226	74.50 204	64.15 223	51.26 223	61.32 208	58.17 216	84.11 163	95.16 50	93.83 68	97.45 111	91.41 184

gg-mvs-nofinetune		81.83 194	83.58 149	79.80 202	91.57 120	96.54 82	93.79 78	68.80 222	62.71 224	43.01 232	55.28 216	85.06 75	83.65 167	96.13 39	94.86 51	97.98 87	94.46 150

pmmvs3		71.13 217	71.06 218	71.21 219	73.54 227	80.19 223	71.69 224	64.86 226	62.04 225	52.10 222	54.92 218	48.00 227	75.03 203	83.75 211	83.24 210	90.04 222	85.27 212

1111		66.22 220	66.42 222	65.98 221	75.69 222	76.42 226	58.90 227	63.25 227	57.86 226	48.33 228	45.46 226	49.13 225	61.32 221	81.57 215	82.80 211	88.38 225	71.69 230

.test1245		48.95 229	46.78 230	51.48 227	75.69 222	76.42 226	58.90 227	63.25 227	57.86 226	48.33 228	45.46 226	49.13 225	61.32 221	81.57 215	5.58 234	1.40 238	11.42 236

PMVS	56.77 18	61.27 224	58.64 226	64.35 224	75.66 224	54.60 235	53.62 233	74.23 205	53.69 228	58.37 213	44.27 229	49.38 223	44.16 230	69.51 229	65.35 230	80.07 230	73.66 226
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
testmv		65.29 221	65.25 223	65.34 222	77.73 220	75.55 228	58.75 229	73.56 211	53.22 229	38.47 233	49.33 222	38.30 232	53.38 226	79.13 221	81.65 215	90.15 220	79.58 222

test1235678		65.29 221	65.24 224	65.34 222	77.73 220	75.54 229	58.75 229	73.56 211	53.19 230	38.47 233	49.32 223	38.28 233	53.38 226	79.13 221	81.65 215	90.15 220	79.57 223

Anonymous20231211		69.76 219	67.18 220	72.76 214	78.31 219	83.47 218	74.12 217	78.37 190	51.44 231	52.48 221	36.04 230	45.46 230	62.33 219	80.49 219	82.43 212	90.96 215	90.93 190

test12356		60.37 225	61.08 225	59.53 226	72.42 229	70.09 231	57.72 231	69.53 220	51.31 232	36.05 235	47.32 224	32.04 234	36.19 231	74.15 226	80.35 221	85.27 227	72.29 228

PMMVS2		53.68 227	55.72 228	51.30 228	58.84 234	67.02 233	54.23 232	60.97 231	47.50 233	19.42 238	34.81 231	31.97 235	30.88 233	65.84 230	69.99 227	83.47 229	72.92 227

EMVS		39.04 232	34.32 233	44.54 232	58.25 235	39.35 238	27.61 238	62.55 230	35.99 234	16.40 240	20.04 236	14.77 239	44.80 228	33.12 235	44.10 233	57.61 236	52.89 234

E-PMN		40.00 230	35.74 232	44.98 231	57.69 236	39.15 239	28.05 237	62.70 229	35.52 235	17.78 239	20.90 234	14.36 240	44.47 229	35.89 234	47.86 232	59.15 235	56.47 233

no-one		49.70 228	49.06 229	50.46 229	65.32 233	67.46 232	38.16 236	68.73 223	34.38 236	22.88 237	24.40 232	22.99 236	28.55 234	51.41 232	70.93 226	79.08 232	71.81 229

MVE	39.81 19	39.52 231	41.58 231	37.11 233	33.93 237	49.06 236	26.45 239	54.22 233	29.46 237	24.15 236	20.77 235	10.60 241	34.42 232	51.12 233	65.27 231	49.49 237	64.81 232
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
testmvs		4.35 233	6.54 234	1.79 235	0.60 238	1.82 240	3.06 241	0.95 236	7.22 238	0.88 242	12.38 237	1.25 242	3.87 237	6.09 236	5.58 234	1.40 238	11.42 236

test123		3.48 234	5.31 235	1.34 236	0.20 240	1.52 241	2.17 242	0.58 237	6.13 239	0.31 243	9.85 238	0.31 243	3.90 236	2.65 237	5.28 236	0.87 240	11.46 235

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										95.36 2		97.46 16

MTMP										95.70 1		96.90 21

Patchmatch-RL test							18.47 240

XVS					95.68 57	98.66 11	94.96 55			88.03 46		96.06 26				98.46 20

X-MVStestdata					95.68 57	98.66 11	94.96 55			88.03 46		96.06 26				98.46 20

mPP-MVS					98.76 19							95.49 33

Patchmtry						92.39 169	89.18 163	73.30 213		71.08 145

High-resolution multi-view benchmark