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	88.51 2	92.64 24	94.42 12	90.56 35	94.84 37	96.92 13	91.31 55	89.61 25	95.16 1	84.55 40	89.91 24	91.45 15	90.15 28	95.12 7	94.81 4	92.90 159	97.58 9

APDe-MVS		95.23 2	95.69 2	94.70 1	97.12 10	97.81 3	97.19 1	92.83 2	95.06 2	90.98 5	96.47 1	92.77 8	93.38 1	95.34 5	94.21 11	96.68 4	98.17 2

SD-MVS		94.53 5	95.22 4	93.73 10	95.69 30	97.03 9	95.77 16	91.95 8	94.41 3	91.35 4	94.97 4	93.34 5	91.80 15	94.72 15	93.99 15	95.82 26	98.07 3

TSAR-MVS + MP.		94.48 6	94.97 5	93.90 8	95.53 31	97.01 10	96.69 3	90.71 17	94.24 4	90.92 6	94.97 4	92.19 10	93.03 2	94.83 11	93.60 21	96.51 7	97.97 4

HFP-MVS		94.02 10	94.22 13	93.78 9	97.25 7	96.85 15	95.81 14	90.94 16	94.12 5	90.29 9	94.09 10	89.98 24	92.52 8	93.94 25	93.49 26	95.87 21	97.10 18

MPTG		93.80 13	93.45 21	94.20 6	97.53 2	96.43 31	95.88 13	91.12 15	94.09 6	92.74 3	87.68 28	90.77 19	92.04 11	94.74 14	93.56 23	95.91 19	96.85 22

HSP-MVS		94.83 3	95.37 3	94.21 5	96.82 19	97.94 2	96.69 3	92.37 7	93.97 7	90.29 9	96.16 3	93.71 3	92.70 5	94.80 12	93.13 31	96.37 8	97.90 5

ESAPD		95.35 1	95.97 1	94.63 2	97.35 5	97.95 1	97.09 2	93.48 1	93.91 8	90.13 11	96.41 2	95.14 1	92.88 4	95.64 2	94.53 7	96.86 2	98.21 1

ACMMPR		93.72 14	93.94 15	93.48 13	97.07 11	96.93 12	95.78 15	90.66 19	93.88 9	89.24 16	93.53 12	89.08 31	92.24 9	93.89 27	93.50 24	95.88 20	96.73 26

HPM-MVS++		94.60 4	94.91 6	94.24 4	97.86 1	96.53 27	96.14 6	92.51 4	93.87 10	90.76 7	93.45 13	93.84 2	92.62 6	95.11 8	94.08 14	95.58 39	97.48 10

CNVR-MVS		94.37 7	94.65 7	94.04 7	97.29 6	97.11 7	96.00 8	92.43 6	93.45 11	89.85 14	90.92 20	93.04 6	92.59 7	95.77 1	94.82 3	96.11 15	97.42 12

TSAR-MVS + ACMM		92.97 19	94.51 9	91.16 31	95.88 28	96.59 25	95.09 23	90.45 23	93.42 12	83.01 47	94.68 6	90.74 20	88.74 34	94.75 13	93.78 19	93.82 138	97.63 8

ACMMP_Plus		93.94 11	94.49 10	93.30 15	97.03 13	97.31 6	95.96 9	91.30 13	93.41 13	88.55 19	93.00 14	90.33 21	91.43 21	95.53 3	94.41 9	95.53 41	97.47 11

NCCC		93.69 15	93.66 18	93.72 11	97.37 4	96.66 24	95.93 12	92.50 5	93.40 14	88.35 20	87.36 30	92.33 9	92.18 10	94.89 10	94.09 13	96.00 16	96.91 21

OMC-MVS		90.23 38	90.40 37	90.03 39	93.45 50	95.29 44	91.89 50	86.34 45	93.25 15	84.94 39	81.72 45	86.65 42	88.90 32	91.69 52	90.27 63	94.65 89	93.95 65

DeepC-MVS_fast	88.76 1	93.10 18	93.02 25	93.19 17	97.13 9	96.51 28	95.35 20	91.19 14	93.14 16	88.14 21	85.26 36	89.49 28	91.45 18	95.17 6	95.07 1	95.85 24	96.48 29
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
TSAR-MVS + GP.		92.71 23	93.91 16	91.30 29	91.96 65	96.00 36	93.43 35	87.94 34	92.53 17	86.27 35	93.57 11	91.94 12	91.44 20	93.29 32	92.89 36	96.78 3	97.15 16

CSCG		92.76 21	93.16 23	92.29 24	96.30 22	97.74 4	94.67 27	88.98 29	92.46 18	89.73 15	86.67 32	92.15 11	88.69 35	92.26 45	92.92 35	95.40 45	97.89 6

APD-MVS		94.37 7	94.47 11	94.26 3	97.18 8	96.99 11	96.53 5	92.68 3	92.45 19	89.96 12	94.53 7	91.63 14	92.89 3	94.58 17	93.82 18	96.31 11	97.26 13
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
CP-MVS		93.25 17	93.26 22	93.24 16	96.84 18	96.51 28	95.52 18	90.61 20	92.37 20	88.88 17	90.91 21	89.52 27	91.91 13	93.64 29	92.78 37	95.69 32	97.09 19

MCST-MVS		93.81 12	94.06 14	93.53 12	96.79 20	96.85 15	95.95 10	91.69 11	92.20 21	87.17 27	90.83 22	93.41 4	91.96 12	94.49 19	93.50 24	97.61 1	97.12 17

DeepC-MVS	87.86 3	92.26 26	91.86 29	92.73 20	96.18 23	96.87 14	95.19 22	91.76 10	92.17 22	86.58 30	81.79 43	85.85 43	90.88 24	94.57 18	94.61 5	95.80 27	97.18 14
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
SteuartSystems-ACMMP		94.06 9	94.65 7	93.38 14	96.97 15	97.36 5	96.12 7	91.78 9	92.05 23	87.34 25	94.42 8	90.87 18	91.87 14	95.47 4	94.59 6	96.21 13	97.77 7
Skip Steuart: Steuart Systems R&D Blog.
CNLPA		88.40 49	87.00 61	90.03 39	93.73 48	94.28 56	89.56 68	85.81 47	91.87 24	87.55 24	69.53 107	81.49 60	89.23 30	89.45 94	88.59 100	94.31 110	93.82 67

TSAR-MVS + COLMAP		88.40 49	89.09 44	87.60 60	92.72 60	93.92 62	92.21 44	85.57 49	91.73 25	73.72 84	91.75 18	73.22 97	87.64 46	91.49 53	89.71 77	93.73 144	91.82 116

ACMMP		92.03 28	92.16 27	91.87 28	95.88 28	96.55 26	94.47 29	89.49 26	91.71 26	85.26 36	91.52 19	84.48 48	90.21 27	92.82 40	91.63 44	95.92 18	96.42 31
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
TAPA-MVS	84.37 7	88.91 45	88.93 45	88.89 49	93.00 56	94.85 51	92.00 47	84.84 54	91.68 27	80.05 60	79.77 52	84.56 47	88.17 40	90.11 86	89.00 96	95.30 52	92.57 94
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
X-MVS		92.36 25	92.75 26	91.90 27	96.89 16	96.70 20	95.25 21	90.48 22	91.50 28	83.95 42	88.20 26	88.82 33	89.11 31	93.75 28	93.43 27	95.75 31	96.83 24

CPTT-MVS		91.39 31	90.95 34	91.91 26	95.06 32	95.24 45	95.02 24	88.98 29	91.02 29	86.71 29	84.89 38	88.58 36	91.60 17	90.82 75	89.67 78	94.08 116	96.45 30

MP-MVS		93.35 16	93.59 19	93.08 18	97.39 3	96.82 17	95.38 19	90.71 17	90.82 30	88.07 22	92.83 16	90.29 22	91.32 22	94.03 22	93.19 30	95.61 37	97.16 15

PHI-MVS		92.05 27	93.74 17	90.08 38	94.96 34	97.06 8	93.11 39	87.71 37	90.71 31	80.78 57	92.40 17	91.03 16	87.68 45	94.32 21	94.48 8	96.21 13	96.16 35

train_agg		92.87 20	93.53 20	92.09 25	96.88 17	95.38 43	95.94 11	90.59 21	90.65 32	83.65 45	94.31 9	91.87 13	90.30 26	93.38 31	92.42 38	95.17 57	96.73 26

3Dnovator+	86.06 4	91.60 30	90.86 36	92.47 22	96.00 27	96.50 30	94.70 26	87.83 36	90.49 33	89.92 13	74.68 73	89.35 29	90.66 25	94.02 23	94.14 12	95.67 34	96.85 22

canonicalmvs		89.36 43	89.92 38	88.70 52	91.38 66	95.92 38	91.81 51	82.61 85	90.37 34	82.73 50	82.09 41	79.28 74	88.30 39	91.17 59	93.59 22	95.36 48	97.04 20

abl_6				90.66 34	94.65 40	96.27 32	92.21 44	86.94 41	90.23 35	86.38 32	85.50 35	92.96 7	88.37 38			95.40 45	95.46 48

MVS_111021_LR		90.14 39	90.89 35	89.26 47	93.23 52	94.05 60	90.43 59	84.65 55	90.16 36	84.52 41	90.14 23	83.80 52	87.99 41	92.50 43	90.92 52	94.74 82	94.70 57

PGM-MVS		92.76 21	93.03 24	92.45 23	97.03 13	96.67 23	95.73 17	87.92 35	90.15 37	86.53 31	92.97 15	88.33 37	91.69 16	93.62 30	93.03 32	95.83 25	96.41 32

MSLP-MVS++		92.02 29	91.40 31	92.75 19	96.01 26	95.88 39	93.73 34	89.00 27	89.89 38	90.31 8	81.28 48	88.85 32	91.45 18	92.88 39	94.24 10	96.00 16	96.76 25

3Dnovator	85.17 5	90.48 36	89.90 40	91.16 31	94.88 36	95.74 40	93.82 31	85.36 50	89.28 39	87.81 23	74.34 75	87.40 41	88.56 36	93.07 35	93.74 20	96.53 6	95.71 42

PLC	83.76 9	88.61 48	86.83 63	90.70 33	94.22 42	92.63 85	91.50 53	87.19 40	89.16 40	86.87 28	75.51 70	80.87 62	89.98 29	90.01 87	89.20 90	94.41 106	90.45 147
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
AdaColmap		90.29 37	88.38 49	92.53 21	96.10 25	95.19 46	92.98 40	91.40 12	89.08 41	88.65 18	78.35 59	81.44 61	91.30 23	90.81 76	90.21 64	94.72 84	93.59 69

CDPH-MVS		91.14 33	92.01 28	90.11 37	96.18 23	96.18 34	94.89 25	88.80 31	88.76 42	77.88 72	89.18 25	87.71 40	87.29 51	93.13 34	93.31 29	95.62 36	95.84 40

HQP-MVS		89.13 44	89.58 42	88.60 54	93.53 49	93.67 63	93.29 37	87.58 38	88.53 43	75.50 76	87.60 29	80.32 65	87.07 52	90.66 81	89.95 71	94.62 92	96.35 34

CLD-MVS		88.66 46	88.52 47	88.82 50	91.37 67	94.22 57	92.82 42	82.08 93	88.27 44	85.14 37	81.86 42	78.53 76	85.93 58	91.17 59	90.61 58	95.55 40	95.00 50
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
CANet		91.33 32	91.46 30	91.18 30	95.01 33	96.71 19	93.77 32	87.39 39	87.72 45	87.26 26	81.77 44	89.73 25	87.32 50	94.43 20	93.86 17	96.31 11	96.02 38

MVS_111021_HR		90.56 35	91.29 33	89.70 43	94.71 39	95.63 41	91.81 51	86.38 44	87.53 46	81.29 54	87.96 27	85.43 45	87.69 44	93.90 26	92.93 34	96.33 9	95.69 43

NP-MVS									87.47 47

ACMP	83.90 8	88.32 52	88.06 52	88.62 53	92.18 63	93.98 61	91.28 56	85.24 51	86.69 48	81.23 55	85.62 34	75.13 86	87.01 53	89.83 89	89.77 76	94.79 78	95.43 49
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
MVS_0304		90.88 34	91.35 32	90.34 36	93.91 45	96.79 18	94.49 28	86.54 43	86.57 49	82.85 48	81.68 46	89.70 26	87.57 47	94.64 16	93.93 16	96.67 5	96.15 36

diffmvs		85.70 68	86.35 66	84.95 70	87.75 110	90.96 101	89.09 73	78.56 135	86.50 50	80.44 59	77.86 61	83.93 51	81.64 73	85.52 151	86.79 124	92.21 166	92.87 81

ACMM	83.27 10	87.68 57	86.09 68	89.54 45	93.26 51	92.19 88	91.43 54	86.74 42	86.02 51	82.85 48	75.63 69	75.14 85	88.41 37	90.68 80	89.99 68	94.59 93	92.97 78
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
LGP-MVS_train		88.25 53	88.55 46	87.89 58	92.84 59	93.66 64	93.35 36	85.22 52	85.77 52	74.03 83	86.60 33	76.29 83	86.62 55	91.20 57	90.58 60	95.29 53	95.75 41

EPNet		89.60 41	89.91 39	89.24 48	96.45 21	93.61 65	92.95 41	88.03 33	85.74 53	83.36 46	87.29 31	83.05 55	80.98 78	92.22 46	91.85 42	93.69 145	95.58 46
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
QAPM		89.49 42	89.58 42	89.38 46	94.73 38	95.94 37	92.35 43	85.00 53	85.69 54	80.03 61	76.97 65	87.81 39	87.87 42	92.18 49	92.10 40	96.33 9	96.40 33

MAR-MVS		88.39 51	88.44 48	88.33 57	94.90 35	95.06 48	90.51 58	83.59 65	85.27 55	79.07 65	77.13 63	82.89 56	87.70 43	92.19 48	92.32 39	94.23 111	94.20 63
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
PVSNet_BlendedMVS		88.19 54	88.00 53	88.42 55	92.71 61	94.82 52	89.08 74	83.81 61	84.91 56	86.38 32	79.14 54	78.11 77	82.66 67	93.05 36	91.10 47	95.86 22	94.86 53

PVSNet_Blended		88.19 54	88.00 53	88.42 55	92.71 61	94.82 52	89.08 74	83.81 61	84.91 56	86.38 32	79.14 54	78.11 77	82.66 67	93.05 36	91.10 47	95.86 22	94.86 53

LS3D		85.96 66	84.37 76	87.81 59	94.13 43	93.27 70	90.26 61	89.00 27	84.91 56	72.84 89	71.74 93	72.47 99	87.45 48	89.53 93	89.09 93	93.20 154	89.60 149

RPSCF		83.46 81	83.36 80	83.59 94	87.75 110	87.35 161	84.82 153	79.46 126	83.84 59	78.12 68	82.69 40	79.87 67	82.60 69	82.47 189	81.13 193	88.78 195	86.13 181

MVS_Test		86.93 60	87.24 60	86.56 64	90.10 94	93.47 67	90.31 60	80.12 112	83.55 60	78.12 68	79.58 53	79.80 69	85.45 60	90.17 85	90.59 59	95.29 53	93.53 70

CANet_DTU		85.43 69	87.72 59	82.76 102	90.95 73	93.01 79	89.99 62	75.46 171	82.67 61	64.91 146	83.14 39	80.09 66	80.68 83	92.03 51	91.03 49	94.57 95	92.08 108

DELS-MVS		89.71 40	89.68 41	89.74 41	93.75 47	96.22 33	93.76 33	85.84 46	82.53 62	85.05 38	78.96 56	84.24 49	84.25 64	94.91 9	94.91 2	95.78 30	96.02 38
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
OpenMVS	82.53 11	87.71 56	86.84 62	88.73 51	94.42 41	95.06 48	91.02 57	83.49 68	82.50 63	82.24 52	67.62 118	85.48 44	85.56 59	91.19 58	91.30 46	95.67 34	94.75 55

PCF-MVS	84.60 6	88.66 46	87.75 58	89.73 42	93.06 55	96.02 35	93.22 38	90.00 24	82.44 64	80.02 62	77.96 60	85.16 46	87.36 49	88.54 105	88.54 101	94.72 84	95.61 45
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
DI_MVS_plusplus_trai		86.41 62	85.54 70	87.42 61	89.24 102	93.13 73	92.16 46	82.65 83	82.30 65	80.75 58	68.30 114	80.41 64	85.01 61	90.56 82	90.07 66	94.70 86	94.01 64

MVSTER		86.03 65	86.12 67	85.93 65	88.62 105	89.93 125	89.33 70	79.91 115	81.87 66	81.35 53	81.07 49	74.91 87	80.66 84	92.13 50	90.10 65	95.68 33	92.80 84

EPP-MVSNet		86.55 61	87.76 57	85.15 68	90.52 79	94.41 55	87.24 105	82.32 90	81.79 67	73.60 85	78.57 58	82.41 57	82.07 71	91.23 55	90.39 62	95.14 60	95.48 47

UGNet		85.90 67	88.23 50	83.18 98	88.96 103	94.10 58	87.52 96	83.60 64	81.66 68	77.90 71	80.76 50	83.19 54	66.70 195	91.13 70	90.71 57	94.39 107	96.06 37
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		81.65 104	84.05 77	78.86 160	78.56 206	82.63 200	83.10 164	67.22 203	81.39 69	70.11 97	84.91 37	79.74 70	82.12 70	87.31 114	85.70 161	92.03 169	86.67 177

GBi-Net		84.51 72	84.80 73	84.17 83	84.20 149	89.95 122	89.70 65	80.37 106	81.17 70	75.50 76	69.63 102	79.69 71	79.75 113	90.73 77	90.72 54	95.52 42	91.71 121

test1		84.51 72	84.80 73	84.17 83	84.20 149	89.95 122	89.70 65	80.37 106	81.17 70	75.50 76	69.63 102	79.69 71	79.75 113	90.73 77	90.72 54	95.52 42	91.71 121

FMVSNet3		84.44 74	84.64 75	84.21 82	84.32 148	90.13 120	89.85 64	80.37 106	81.17 70	75.50 76	69.63 102	79.69 71	79.62 116	89.72 91	90.52 61	95.59 38	91.58 127

USDC		80.69 115	79.89 121	81.62 121	86.48 124	89.11 144	86.53 129	78.86 131	81.15 73	63.48 156	72.98 89	59.12 184	81.16 76	87.10 117	85.01 169	93.23 153	84.77 191

IS_MVSNet		86.18 63	88.18 51	83.85 90	91.02 70	94.72 54	87.48 97	82.46 87	81.05 74	70.28 95	76.98 64	82.20 59	76.65 143	93.97 24	93.38 28	95.18 56	94.97 51

EPMVS		77.53 166	78.07 145	76.90 177	86.89 122	84.91 191	82.18 174	66.64 206	81.00 75	64.11 152	72.75 91	69.68 107	74.42 158	79.36 202	78.13 201	87.14 204	80.68 205

Vis-MVSNet (Re-imp)		83.65 79	86.81 64	79.96 153	90.46 82	92.71 84	84.84 152	82.00 94	80.93 76	62.44 163	76.29 67	82.32 58	65.54 198	92.29 44	91.66 43	94.49 101	91.47 128

PatchmatchNet		78.67 156	78.85 132	78.46 164	86.85 123	86.03 178	83.77 161	68.11 200	80.88 77	66.19 128	72.90 90	73.40 95	78.06 135	79.25 203	77.71 205	87.75 199	81.75 200
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
PatchMatch-RL		83.34 82	81.36 95	85.65 66	90.33 85	89.52 135	84.36 156	81.82 96	80.87 78	79.29 63	74.04 78	62.85 139	86.05 57	88.40 107	87.04 121	92.04 168	86.77 174

EPNet_dtu		81.98 97	83.82 78	79.83 155	94.10 44	85.97 179	87.29 102	84.08 60	80.61 79	59.96 184	81.62 47	77.19 81	62.91 202	87.21 115	86.38 135	90.66 183	87.77 165
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
MSDG		83.87 76	81.02 101	87.19 62	92.17 64	89.80 128	89.15 71	85.72 48	80.61 79	79.24 64	66.66 122	68.75 111	82.69 66	87.95 111	87.44 114	94.19 112	85.92 183

CostFormer		80.94 113	80.21 116	81.79 113	87.69 113	88.58 150	87.47 98	70.66 189	80.02 81	77.88 72	73.03 88	71.40 102	78.24 134	79.96 199	79.63 195	88.82 194	88.84 153

tpmp4_e23		79.82 131	77.96 153	82.00 109	87.59 115	86.93 165	87.81 92	72.21 184	79.99 82	78.02 70	67.83 117	64.77 123	78.74 131	79.99 198	78.90 198	87.65 200	87.29 169

CHOSEN 1792x2688		82.16 95	80.91 108	83.61 93	91.14 68	92.01 90	89.55 69	79.15 130	79.87 83	70.29 94	52.51 206	72.56 98	81.39 74	88.87 100	88.17 106	90.15 187	92.37 106

FC-MVSNet-train		85.18 71	85.31 71	85.03 69	90.67 74	91.62 94	87.66 94	83.61 63	79.75 84	74.37 82	78.69 57	71.21 103	78.91 130	91.23 55	89.96 70	94.96 67	94.69 58

DWT-MVSNet_training		80.51 119	78.05 148	83.39 96	88.64 104	88.33 155	86.11 135	76.33 156	79.65 85	78.64 67	69.62 105	58.89 186	80.82 79	80.50 196	82.03 191	89.77 190	87.36 168

GG-mvs-BLEND		57.56 219	82.61 84	28.34 232	0.22 237	90.10 121	79.37 191	0.14 236	79.56 86	0.40 240	71.25 96	83.40 53	0.30 237	86.27 135	83.87 177	89.59 191	83.83 193

PVSNet_Blended_VisFu		87.40 59	87.80 55	86.92 63	92.86 57	95.40 42	88.56 86	83.45 71	79.55 87	82.26 51	74.49 74	84.03 50	79.24 128	92.97 38	91.53 45	95.15 59	96.65 28

Effi-MVS+		85.33 70	85.08 72	85.63 67	89.69 101	93.42 68	89.90 63	80.31 110	79.32 88	72.48 91	73.52 85	74.03 90	86.55 56	90.99 72	89.98 69	94.83 76	94.27 62

ADS-MVSNet		74.53 196	75.69 187	73.17 199	81.57 196	80.71 208	79.27 192	63.03 218	79.27 89	59.94 185	67.86 116	68.32 115	71.08 177	77.33 207	76.83 208	84.12 218	79.53 206

tfpn_ndepth		81.77 103	82.29 85	81.15 133	89.79 100	91.71 93	85.49 144	81.63 99	79.17 90	64.76 147	73.04 87	68.14 116	70.62 179	88.72 101	87.88 110	94.63 91	87.38 167

FMVSNet2		83.87 76	83.73 79	84.05 88	84.20 149	89.95 122	89.70 65	80.21 111	79.17 90	74.89 80	65.91 124	77.49 80	79.75 113	90.87 74	91.00 51	95.52 42	91.71 121

MDTV_nov1_ep13		79.14 144	79.49 127	78.74 161	85.40 134	86.89 166	84.32 158	70.29 191	78.85 92	69.42 105	75.37 71	73.29 96	75.64 148	80.61 195	79.48 197	87.36 201	81.91 199

pmmvs4		79.99 126	78.08 144	82.22 106	83.04 180	87.16 164	84.95 150	78.80 133	78.64 93	74.53 81	64.61 141	59.41 179	79.45 126	84.13 175	84.54 175	92.53 163	88.08 160

tpmrst		76.55 175	75.99 181	77.20 173	87.32 118	83.05 196	82.86 165	65.62 210	78.61 94	67.22 122	69.19 108	65.71 120	75.87 147	76.75 210	75.33 212	84.31 216	83.28 196

CHOSEN 280x420		80.28 121	81.66 90	78.67 162	82.92 183	79.24 212	85.36 146	66.79 205	78.11 95	70.32 93	75.03 72	79.87 67	81.09 77	89.07 97	83.16 182	85.54 211	87.17 171

Fast-Effi-MVS+		83.77 78	82.98 81	84.69 71	87.98 108	91.87 91	88.10 90	77.70 145	78.10 96	73.04 88	69.13 109	68.51 112	86.66 54	90.49 83	89.85 73	94.67 88	92.88 80

tfpn1000		81.03 111	81.70 89	80.25 151	90.18 92	91.35 95	83.96 159	81.15 103	78.00 97	62.11 166	73.37 86	65.75 119	69.17 184	88.68 103	87.44 114	94.93 68	87.29 169

MS-PatchMatch		81.79 101	81.44 94	82.19 107	90.35 84	89.29 139	88.08 91	75.36 172	77.60 98	69.00 108	64.37 143	78.87 75	77.14 142	88.03 110	85.70 161	93.19 155	86.24 180

tpm cat1		77.78 164	75.28 191	80.70 141	87.14 120	85.84 181	85.81 137	70.40 190	77.44 99	78.80 66	63.72 144	64.01 131	76.55 144	75.60 213	75.21 213	85.51 212	85.12 188

COLMAP_ROB	76.78 15	80.50 120	78.49 134	82.85 100	90.96 72	89.65 133	86.20 134	83.40 72	77.15 100	66.54 124	62.27 148	65.62 121	77.89 137	85.23 163	84.70 173	92.11 167	84.83 190
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
FC-MVSNet-test		76.53 178	81.62 91	70.58 203	84.99 142	85.73 182	74.81 204	78.85 132	77.00 101	39.13 227	75.90 68	73.50 94	54.08 210	86.54 131	85.99 157	91.65 171	86.68 175

thresconf0.02		81.14 110	80.93 106	81.39 125	90.01 98	91.31 96	86.79 125	82.28 91	76.97 102	61.46 175	74.24 76	62.08 153	72.98 168	88.70 102	87.90 108	94.81 77	85.28 186

IterMVS-LS		83.28 83	82.95 82	83.65 92	88.39 107	88.63 149	86.80 124	78.64 134	76.56 103	73.43 86	72.52 92	75.35 84	80.81 81	86.43 134	88.51 102	93.84 137	92.66 88
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
IB-MVS	79.09 12	82.60 89	82.19 86	83.07 99	91.08 69	93.55 66	80.90 184	81.35 100	76.56 103	80.87 56	64.81 140	69.97 106	68.87 185	85.64 144	90.06 67	95.36 48	94.74 56
Christian Sormann, Mattia Rossi, Andreas Kuhn and Friedrich Fraundorfer: IB-MVS: An Iterative Algorithm for Deep Multi-View Stereo based on Binary Decisions. BMVC 2021
HyFIR lowres test		81.62 107	79.45 128	84.14 85	91.00 71	93.38 69	88.27 87	78.19 139	76.28 105	70.18 96	48.78 210	73.69 93	83.52 65	87.05 119	87.83 112	93.68 146	89.15 152

UniMVSNet (Re)		81.22 108	81.08 100	81.39 125	85.35 135	91.76 92	84.93 151	82.88 73	76.13 106	65.02 145	64.94 138	63.09 136	75.17 150	87.71 112	89.04 94	94.97 66	94.88 52

UniMVSNet_NR-MVSNet		81.87 98	81.33 96	82.50 103	85.31 136	91.30 97	85.70 138	84.25 57	75.89 107	64.21 149	66.95 121	64.65 125	80.22 98	87.07 118	89.18 91	95.27 55	94.29 60

tfpnview11		80.84 114	81.10 99	80.54 145	90.10 94	90.96 101	85.44 145	81.84 95	75.77 108	59.27 188	73.54 82	64.40 126	71.69 173	89.16 96	87.97 107	94.91 69	85.92 183

DU-MVS		81.20 109	80.30 115	82.25 105	84.98 143	90.94 103	85.70 138	83.58 66	75.74 109	64.21 149	65.30 132	59.60 178	80.22 98	86.89 122	89.31 86	94.77 80	94.29 60

NR-MVSNet		80.25 122	79.98 120	80.56 144	85.20 138	90.94 103	85.65 140	83.58 66	75.74 109	61.36 177	65.30 132	56.75 194	72.38 169	88.46 106	88.80 98	95.16 58	93.87 66

Baseline_NR-MVSNet		79.84 129	78.37 141	81.55 123	84.98 143	86.66 168	85.06 149	83.49 68	75.57 111	63.31 157	58.22 190	60.97 166	78.00 136	86.89 122	87.13 119	94.47 102	93.15 74

OPM-MVS		87.56 58	85.80 69	89.62 44	93.90 46	94.09 59	94.12 30	88.18 32	75.40 112	77.30 75	76.41 66	77.93 79	88.79 33	92.20 47	90.82 53	95.40 45	93.72 68

tfpn111		83.51 80	82.68 83	84.47 76	90.30 86	93.09 74	89.05 76	82.72 77	75.14 113	69.49 103	74.24 76	63.13 132	80.38 91	91.15 64	89.51 80	94.91 69	92.50 100

conf0.01		82.64 88	81.02 101	84.53 74	90.30 86	93.22 72	89.05 76	82.75 75	75.14 113	69.69 99	67.15 120	59.19 181	80.38 91	91.16 62	89.51 80	95.00 65	91.76 119

conf0.002		82.54 91	80.83 109	84.54 72	90.28 91	93.24 71	89.05 76	82.75 75	75.14 113	69.75 98	67.99 115	57.12 192	80.38 91	91.16 62	89.79 74	95.02 63	91.36 130

conf200view11		82.85 85	81.46 92	84.47 76	90.30 86	93.09 74	89.05 76	82.72 77	75.14 113	69.49 103	65.72 126	63.13 132	80.38 91	91.15 64	89.51 80	94.91 69	92.50 100

thres100view900		82.55 90	81.01 104	84.34 79	90.30 86	92.27 86	89.04 81	82.77 74	75.14 113	69.56 100	65.72 126	63.13 132	79.62 116	89.97 88	89.26 88	94.73 83	91.61 126

tfpn200view9		82.86 84	81.46 92	84.48 75	90.30 86	93.09 74	89.05 76	82.71 79	75.14 113	69.56 100	65.72 126	63.13 132	80.38 91	91.15 64	89.51 80	94.91 69	92.50 100

dps		78.02 161	75.94 182	80.44 147	86.06 127	86.62 171	82.58 166	69.98 193	75.14 113	77.76 74	69.08 110	59.93 174	78.47 132	79.47 201	77.96 202	87.78 198	83.40 195

CR-MVSNet		78.71 155	78.86 131	78.55 163	85.85 130	85.15 188	82.30 171	68.23 198	74.71 120	65.37 141	64.39 142	69.59 108	77.18 140	85.10 167	84.87 170	92.34 165	88.21 158

RPMNet		77.07 169	77.63 159	76.42 180	85.56 133	85.15 188	81.37 176	65.27 212	74.71 120	60.29 183	63.71 145	66.59 118	73.64 162	82.71 186	82.12 189	92.38 164	88.39 156

Vis-MVSNet		84.38 75	86.68 65	81.70 118	87.65 114	94.89 50	88.14 88	80.90 104	74.48 122	68.23 114	77.53 62	80.72 63	69.98 181	92.68 41	91.90 41	95.33 51	94.58 59
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
thres200		82.77 86	81.25 97	84.54 72	90.38 83	93.05 77	89.13 72	82.67 81	74.40 123	69.53 102	65.69 129	63.03 137	80.63 85	91.15 64	89.42 85	94.88 73	92.04 110

thres400		82.68 87	81.15 98	84.47 76	90.52 79	92.89 83	88.95 82	82.71 79	74.33 124	69.22 107	65.31 131	62.61 141	80.63 85	90.96 73	89.50 84	94.79 78	92.45 105

TranMVSNet+NR-MVSNet		80.52 118	79.84 122	81.33 128	84.92 145	90.39 110	85.53 143	84.22 59	74.27 125	60.68 182	64.93 139	59.96 173	77.48 139	86.75 127	89.28 87	95.12 62	93.29 71

TDRefinement		79.05 148	77.05 170	81.39 125	88.45 106	89.00 146	86.92 120	82.65 83	74.21 126	64.41 148	59.17 183	59.16 182	74.52 156	85.23 163	85.09 168	91.37 175	87.51 166

view600		82.51 93	81.00 105	84.27 81	90.56 78	92.95 81	88.57 84	82.57 86	74.16 127	68.70 111	65.13 134	62.15 151	80.36 96	91.15 64	88.98 97	94.87 75	92.48 103

thres600view7		82.53 92	81.02 101	84.28 80	90.61 75	93.05 77	88.57 84	82.67 81	74.12 128	68.56 112	65.09 135	62.13 152	80.40 90	91.15 64	89.02 95	94.88 73	92.59 91

view800		82.38 94	80.93 106	84.06 86	90.59 77	92.96 80	88.11 89	82.44 88	73.92 129	68.10 115	65.07 136	61.64 154	80.10 102	91.17 59	89.24 89	95.01 64	92.56 95

PatchT		76.42 179	77.81 157	74.80 192	78.46 207	84.30 193	71.82 210	65.03 214	73.89 130	65.37 141	61.58 150	66.70 117	77.18 140	85.10 167	84.87 170	90.94 182	88.21 158

ACMH+	79.08 13	81.84 100	80.06 117	83.91 89	89.92 99	90.62 105	86.21 133	83.48 70	73.88 131	65.75 137	66.38 123	65.30 122	84.63 62	85.90 139	87.25 118	93.45 149	91.13 132

Effi-MVS+-dtu		82.05 96	81.76 88	82.38 104	87.72 112	90.56 106	86.90 122	78.05 141	73.85 132	66.85 123	71.29 95	71.90 101	82.00 72	86.64 129	85.48 165	92.76 161	92.58 93

test-LLR		79.47 142	79.84 122	79.03 159	87.47 116	82.40 203	81.24 179	78.05 141	73.72 133	62.69 160	73.76 79	74.42 88	73.49 163	84.61 171	82.99 184	91.25 177	87.01 172

TESTMET0.1,1		77.78 164	79.84 122	75.38 188	80.86 199	82.40 203	81.24 179	62.72 219	73.72 133	62.69 160	73.76 79	74.42 88	73.49 163	84.61 171	82.99 184	91.25 177	87.01 172

tfpn		81.79 101	80.06 117	83.82 91	90.61 75	92.91 82	87.62 95	82.34 89	73.66 135	67.46 118	64.99 137	55.50 200	79.77 112	91.12 71	89.62 79	95.14 60	92.59 91

test-mter		77.79 163	80.02 119	75.18 189	81.18 198	82.85 198	80.52 187	62.03 220	73.62 136	62.16 165	73.55 81	73.83 92	73.81 161	84.67 170	83.34 181	91.37 175	88.31 157

IterMVS		78.79 154	79.71 125	77.71 167	85.26 137	85.91 180	84.54 155	69.84 195	73.38 137	61.25 178	70.53 99	70.35 104	74.43 157	85.21 165	83.80 179	90.95 181	88.77 154
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
UA-Net		86.07 64	87.78 56	84.06 86	92.85 58	95.11 47	87.73 93	84.38 56	73.22 138	73.18 87	79.99 51	89.22 30	71.47 176	93.22 33	93.03 32	94.76 81	90.69 142

FMVSNet5		75.50 192	76.07 178	74.83 191	76.16 212	81.19 206	81.34 177	70.21 192	73.20 139	61.59 174	58.97 185	68.33 114	68.50 186	85.87 140	85.85 158	91.18 180	79.11 209

test0.0.03 1		76.03 185	78.51 133	73.12 200	87.47 116	85.13 190	76.32 201	78.05 141	73.19 140	50.98 208	70.64 97	69.28 109	55.53 206	85.33 161	84.38 176	90.39 185	81.63 201

tfpn_n400		80.63 116	80.79 110	80.43 148	90.02 96	91.08 99	85.34 147	81.79 97	72.93 141	59.27 188	73.54 82	64.40 126	71.61 174	89.05 98	88.21 104	94.56 96	86.32 178

tfpnconf		80.63 116	80.79 110	80.43 148	90.02 96	91.08 99	85.34 147	81.79 97	72.93 141	59.27 188	73.54 82	64.40 126	71.61 174	89.05 98	88.21 104	94.56 96	86.32 178

Fast-Effi-MVS+-dtu		79.95 127	80.69 112	79.08 158	86.36 125	89.14 143	85.85 136	72.28 183	72.85 143	59.32 187	70.43 100	68.42 113	77.57 138	86.14 136	86.44 134	93.11 156	91.39 129

v18		79.71 136	77.98 152	81.73 117	84.02 155	86.67 167	87.37 100	76.35 155	72.61 144	68.86 109	61.35 151	62.65 140	79.94 104	85.49 153	86.21 140	93.85 136	90.92 135

v16		79.65 137	77.91 154	81.69 119	84.04 153	86.65 170	87.20 107	76.32 157	72.41 145	68.71 110	61.13 159	62.52 143	79.93 105	85.55 149	86.22 138	93.92 127	90.91 136

v17		79.59 139	77.88 155	81.60 122	84.03 154	86.66 168	87.13 116	76.31 158	72.09 146	68.29 113	61.15 158	62.57 142	79.90 106	85.55 149	86.20 143	93.93 125	90.93 134

tpm		76.30 183	76.05 180	76.59 179	86.97 121	83.01 197	83.83 160	67.06 204	71.83 147	63.87 154	69.56 106	62.88 138	73.41 165	79.79 200	78.59 199	84.41 215	86.68 175

v6		80.11 123	78.47 135	82.01 108	83.97 156	90.49 107	87.19 110	79.67 120	71.59 148	67.51 117	61.26 152	62.46 146	79.81 111	85.49 153	86.18 148	93.89 131	91.86 113

v1neww		80.09 124	78.45 137	82.00 109	83.97 156	90.49 107	87.18 111	79.67 120	71.49 149	67.44 119	61.24 154	62.41 147	79.83 108	85.49 153	86.19 145	93.88 133	91.86 113

v7new		80.09 124	78.45 137	82.00 109	83.97 156	90.49 107	87.18 111	79.67 120	71.49 149	67.44 119	61.24 154	62.41 147	79.83 108	85.49 153	86.19 145	93.88 133	91.86 113

v8		79.90 128	78.39 140	81.66 120	83.97 156	89.81 127	87.16 114	77.40 147	71.49 149	67.71 116	61.24 154	62.49 144	79.83 108	85.48 157	86.17 149	93.89 131	92.02 112

V42		79.59 139	78.43 139	80.94 139	82.79 186	89.71 131	86.66 126	76.73 153	71.38 152	67.42 121	61.01 160	62.30 149	78.39 133	85.56 148	86.48 132	93.65 147	92.60 90

PM-MVS		74.17 198	73.10 198	75.41 187	76.07 213	82.53 201	77.56 200	71.69 185	71.04 153	61.92 168	61.23 157	47.30 217	74.82 154	81.78 192	79.80 194	90.42 184	88.05 161

MIMVSNet		74.69 195	75.60 188	73.62 197	76.02 214	85.31 187	81.21 181	67.43 201	71.02 154	59.07 192	54.48 198	64.07 129	66.14 197	86.52 132	86.64 127	91.83 170	81.17 203

FMVSNet1		81.64 105	80.61 113	82.84 101	82.36 190	89.20 141	88.67 83	79.58 124	70.79 155	72.63 90	58.95 186	72.26 100	79.34 127	90.73 77	90.72 54	94.47 102	91.62 125

v2v482		79.84 129	78.07 145	81.90 112	83.75 169	90.21 119	87.17 113	79.85 119	70.65 156	65.93 136	61.93 149	60.07 172	80.82 79	85.25 162	86.71 125	93.88 133	91.70 124

TinyColmap		76.73 171	73.95 197	79.96 153	85.16 140	85.64 184	82.34 170	78.19 139	70.63 157	62.06 167	60.69 168	49.61 214	80.81 81	85.12 166	83.69 180	91.22 179	82.27 198

v7		79.79 132	78.28 142	81.54 124	83.73 170	90.34 115	87.27 103	78.27 138	70.50 158	65.59 138	60.59 169	60.47 168	80.46 88	86.90 121	86.63 128	93.92 127	92.56 95

v10		79.62 138	78.19 143	81.28 129	83.73 170	89.69 132	87.27 103	76.86 151	70.50 158	65.46 139	60.58 171	60.47 168	80.44 89	86.91 120	86.63 128	93.93 125	92.55 97

GA-MVS		79.52 141	79.71 125	79.30 157	85.68 131	90.36 114	84.55 154	78.44 136	70.47 160	57.87 195	68.52 113	61.38 163	76.21 145	89.40 95	87.89 109	93.04 158	89.96 148

ACMH	78.52 14	81.86 99	80.45 114	83.51 95	90.51 81	91.22 98	85.62 141	84.23 58	70.29 161	62.21 164	69.04 111	64.05 130	84.48 63	87.57 113	88.45 103	94.01 121	92.54 98
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
v1141		79.75 134	78.04 149	81.75 114	83.89 163	90.37 112	87.20 107	79.89 117	70.23 162	66.18 130	60.92 162	61.48 159	79.54 120	85.36 159	86.17 149	93.81 139	91.76 119

v1		79.76 133	78.06 147	81.74 116	83.89 163	90.38 111	87.20 107	79.88 118	70.23 162	66.17 133	60.92 162	61.56 155	79.50 124	85.37 158	86.17 149	93.81 139	91.77 117

divwei89l23v2f112		79.75 134	78.04 149	81.75 114	83.90 160	90.37 112	87.21 106	79.90 116	70.20 164	66.18 130	60.92 162	61.48 159	79.52 123	85.36 159	86.17 149	93.81 139	91.77 117

v15		79.13 145	77.37 160	81.19 130	83.90 160	86.56 172	87.01 118	76.15 162	70.20 164	66.48 125	60.71 167	61.55 156	79.60 118	85.59 147	86.19 145	93.98 123	90.80 141

V14		79.11 146	77.35 162	81.16 132	83.90 160	86.54 173	86.94 119	76.10 164	70.14 166	66.41 127	60.59 169	61.54 157	79.59 119	85.64 144	86.20 143	94.04 119	90.82 139

V9		79.08 147	77.32 164	81.14 134	83.89 163	86.52 174	86.85 123	76.06 165	70.02 167	66.42 126	60.44 172	61.52 158	79.54 120	85.68 143	86.21 140	94.08 116	90.83 138

v12		79.03 149	77.28 165	81.06 136	83.88 167	86.49 175	86.62 127	76.02 166	69.99 168	66.18 130	60.34 175	61.44 161	79.54 120	85.70 142	86.21 140	94.11 115	90.82 139

v13		78.99 151	77.25 167	81.02 137	83.87 168	86.47 177	86.60 128	75.96 168	69.87 169	66.07 134	60.25 176	61.41 162	79.49 125	85.72 141	86.22 138	94.14 114	90.84 137

v11		79.02 150	77.36 161	80.95 138	83.89 163	86.48 176	86.53 129	75.77 170	69.69 170	65.21 144	60.36 174	60.24 171	80.32 97	87.20 116	86.54 131	93.96 124	91.02 133

WR-MVS		76.63 173	78.02 151	75.02 190	84.14 152	89.76 130	78.34 197	80.64 105	69.56 171	52.32 203	61.26 152	61.24 164	60.66 203	84.45 173	87.07 120	93.99 122	92.77 85

v148		78.59 157	76.84 173	80.62 143	83.61 173	89.16 142	83.65 162	79.24 129	69.38 172	69.34 106	59.88 179	60.41 170	75.19 149	83.81 177	84.63 174	92.70 162	90.63 144

v1144		79.38 143	77.83 156	81.18 131	83.62 172	90.23 117	87.15 115	78.35 137	69.13 173	64.02 153	60.20 177	59.41 179	80.14 101	86.78 125	86.57 130	93.81 139	92.53 99

CDS-MVSNet		81.63 106	82.09 87	81.09 135	87.21 119	90.28 116	87.46 99	80.33 109	69.06 174	70.66 92	71.30 94	73.87 91	67.99 188	89.58 92	89.87 72	92.87 160	90.69 142
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
CVMVSNet		76.70 172	78.46 136	74.64 194	83.34 175	84.48 192	81.83 175	74.58 173	68.88 175	51.23 207	69.77 101	70.05 105	67.49 191	84.27 174	83.81 178	89.38 192	87.96 162

MDTV_nov1_ep13_2view		73.21 200	72.91 199	73.56 198	80.01 200	84.28 194	78.62 195	66.43 207	68.64 176	59.12 191	60.39 173	59.69 177	69.81 182	78.82 205	77.43 207	87.36 201	81.11 204

v1192		78.94 152	77.33 163	80.82 140	83.25 176	89.90 126	86.91 121	77.72 144	68.63 177	62.61 162	59.17 183	57.53 190	80.62 87	86.89 122	86.47 133	93.79 143	92.75 87

v1921920		78.57 158	76.99 171	80.41 150	82.93 182	89.63 134	86.38 132	77.14 149	68.31 178	61.80 170	58.89 187	56.79 193	80.19 100	86.50 133	86.05 156	94.02 120	92.76 86

v144192		78.81 153	77.22 168	80.67 142	82.95 181	89.79 129	86.40 131	77.42 146	68.26 179	63.13 158	59.50 181	58.13 188	80.08 103	85.93 138	86.08 154	94.06 118	92.83 83

conf0.05thres1000		81.00 112	79.12 129	83.20 97	90.14 93	92.15 89	87.05 117	82.09 92	68.11 180	66.19 128	59.67 180	61.10 165	79.05 129	90.47 84	89.11 92	94.68 87	93.22 72

CP-MVSNet		76.36 182	76.41 175	76.32 182	82.73 187	88.64 148	79.39 190	79.62 123	67.21 181	53.70 199	60.72 166	55.22 202	67.91 190	83.52 179	86.34 136	94.55 98	93.19 73

v1240		78.15 160	76.53 174	80.04 152	82.85 185	89.48 137	85.61 142	76.77 152	67.05 182	61.18 180	58.37 189	56.16 198	79.89 107	86.11 137	86.08 154	93.92 127	92.47 104

LP		68.35 206	67.23 207	69.67 205	77.49 209	79.38 211	72.84 209	61.37 221	66.94 183	55.08 197	47.00 212	50.35 212	65.16 199	75.61 212	76.03 209	86.08 210	75.28 215

WR-MVS_H		75.84 189	76.93 172	74.57 195	82.86 184	89.50 136	78.34 197	79.36 128	66.90 184	52.51 202	60.20 177	59.71 175	59.73 204	83.61 178	85.77 159	94.65 89	92.84 82

pmmvs-eth3d		74.32 197	71.96 202	77.08 175	77.33 210	82.71 199	78.41 196	76.02 166	66.65 185	65.98 135	54.23 201	49.02 216	73.14 167	82.37 190	82.69 186	91.61 172	86.05 182

PEN-MVS		76.02 186	76.07 178	75.95 185	83.17 178	87.97 157	79.65 188	80.07 114	66.57 186	51.45 205	60.94 161	55.47 201	66.81 194	82.72 185	86.80 123	94.59 93	92.03 111

N_pmnet		66.85 208	66.63 208	67.11 210	78.73 205	74.66 217	70.53 211	71.07 187	66.46 187	46.54 213	51.68 208	51.91 211	55.48 207	74.68 215	72.38 219	80.29 223	74.65 216

DTE-MVSNet		75.14 193	75.44 190	74.80 192	83.18 177	87.19 163	78.25 199	80.11 113	66.05 188	48.31 211	60.88 165	54.67 203	64.54 200	82.57 187	86.17 149	94.43 105	90.53 146

PS-CasMVS		75.90 188	75.86 185	75.96 184	82.59 188	88.46 153	79.23 193	79.56 125	66.00 189	52.77 201	59.48 182	54.35 205	67.14 193	83.37 182	86.23 137	94.47 102	93.10 75

v7n		77.22 168	76.23 177	78.38 165	81.89 193	89.10 145	82.24 173	76.36 154	65.96 190	61.21 179	56.56 193	55.79 199	75.07 152	86.55 130	86.68 126	93.52 148	92.95 79

anonymousdsp		77.94 162	79.00 130	76.71 178	79.03 204	87.83 158	79.58 189	72.87 182	65.80 191	58.86 194	65.82 125	62.48 145	75.99 146	86.77 126	88.66 99	93.92 127	95.68 44

tmp_tt				32.73 231	43.96 235	21.15 238	26.71 235	8.99 234	65.67 192	51.39 206	56.01 197	42.64 220	11.76 234	56.60 228	50.81 230	53.55 233

SixPastTwentyTwo		76.02 186	75.72 186	76.36 181	83.38 174	87.54 159	75.50 203	76.22 160	65.50 193	57.05 196	70.64 97	53.97 206	74.54 155	80.96 194	82.12 189	91.44 173	89.35 151

v52		76.55 175	75.89 183	77.31 172	79.94 203	88.49 152	81.07 182	73.62 179	65.49 194	61.66 173	56.29 196	58.90 185	74.30 159	83.47 181	85.62 163	93.28 151	92.99 76

V4		76.55 175	75.89 183	77.32 171	79.95 202	88.50 151	81.07 182	73.62 179	65.47 195	61.71 171	56.31 195	58.87 187	74.28 160	83.48 180	85.62 163	93.28 151	92.98 77

pmmvs5		76.93 170	76.33 176	77.62 168	81.97 192	88.40 154	81.32 178	74.35 176	65.42 196	61.42 176	63.07 146	57.95 189	73.23 166	85.60 146	85.35 167	93.41 150	88.55 155

TAMVS		76.42 179	77.16 169	75.56 186	83.05 179	85.55 185	80.58 186	71.43 186	65.40 197	61.04 181	67.27 119	69.22 110	67.99 188	84.88 169	84.78 172	89.28 193	83.01 197

v748		76.17 184	75.10 193	77.43 170	81.60 194	88.01 156	79.02 194	76.28 159	64.47 198	64.14 151	56.55 194	56.26 197	70.40 180	82.50 188	85.77 159	93.11 156	92.15 107

Anonymous20231206		70.80 202	70.59 205	71.04 202	81.60 194	82.49 202	74.64 205	75.87 169	64.17 199	49.27 209	44.85 217	53.59 208	54.68 209	83.07 183	82.34 188	90.17 186	83.65 194

testgi		71.92 201	74.20 196	69.27 206	84.58 147	83.06 195	73.40 206	74.39 175	64.04 200	46.17 214	68.90 112	57.15 191	48.89 217	84.07 176	83.08 183	88.18 197	79.09 210

EU-MVSNet		69.98 204	72.30 201	67.28 209	75.67 215	79.39 210	73.12 207	69.94 194	63.59 201	42.80 218	62.93 147	56.71 195	55.07 208	79.13 204	78.55 200	87.06 205	85.82 185

pm-mvs1		78.51 159	77.75 158	79.40 156	84.83 146	89.30 138	83.55 163	79.38 127	62.64 202	63.68 155	58.73 188	64.68 124	70.78 178	89.79 90	87.84 111	94.17 113	91.28 131

CMPMVS	56.49 17	73.84 199	71.73 203	76.31 183	85.20 138	85.67 183	75.80 202	73.23 181	62.26 203	65.40 140	53.40 204	59.70 176	71.77 172	80.25 197	79.56 196	86.45 207	81.28 202
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
MVS-HIRNet		68.83 205	66.39 209	71.68 201	77.58 208	75.52 216	66.45 216	65.05 213	62.16 204	62.84 159	44.76 218	56.60 196	71.96 171	78.04 206	75.06 214	86.18 209	72.56 218

FPMVS		63.63 214	60.08 220	67.78 208	80.01 200	71.50 221	72.88 208	69.41 197	61.82 205	53.11 200	45.12 216	42.11 221	50.86 214	66.69 224	63.84 226	80.41 222	69.46 223

tfpnnormal		77.46 167	74.86 194	80.49 146	86.34 126	88.92 147	84.33 157	81.26 101	61.39 206	61.70 172	51.99 207	53.66 207	74.84 153	88.63 104	87.38 117	94.50 100	92.08 108

testpf		63.91 212	65.23 211	62.38 216	81.32 197	69.95 223	62.71 222	54.16 228	61.29 207	48.73 210	57.31 191	52.50 209	50.97 213	67.50 223	68.86 224	76.36 226	79.21 208

new_pmnet		59.28 218	61.47 218	56.73 223	61.66 228	68.29 224	59.57 223	54.91 226	60.83 208	34.38 229	44.66 219	43.65 219	49.90 215	71.66 221	71.56 222	79.94 224	69.67 222

ambc			61.92 216		70.98 224	73.54 218	63.64 220		60.06 209	52.23 204	38.44 221	19.17 236	57.12 205	82.33 191	75.03 215	83.21 219	84.89 189

TransMVSNet (Re)		76.57 174	75.16 192	78.22 166	85.60 132	87.24 162	82.46 167	81.23 102	59.80 210	59.05 193	57.07 192	59.14 183	66.60 196	88.09 109	86.82 122	94.37 108	87.95 163

EG-PatchMatch MVS		76.40 181	75.47 189	77.48 169	85.86 129	90.22 118	82.45 168	73.96 178	59.64 211	59.60 186	52.75 205	62.20 150	68.44 187	88.23 108	87.50 113	94.55 98	87.78 164

MDA-MVSNet-bldmvs		66.22 209	64.49 212	68.24 207	61.67 227	82.11 205	70.07 212	76.16 161	59.14 212	47.94 212	54.35 200	35.82 229	67.33 192	64.94 227	75.68 211	86.30 208	79.36 207

test20.03		68.31 207	70.05 206	66.28 211	82.41 189	80.84 207	67.35 215	76.11 163	58.44 213	40.80 221	53.77 202	54.54 204	42.28 223	83.07 183	81.96 192	88.73 196	77.76 212

new-patchmatchnet		63.80 213	63.31 215	64.37 213	76.49 211	75.99 215	63.73 219	70.99 188	57.27 214	43.08 217	45.86 215	43.80 218	45.13 222	73.20 218	70.68 223	86.80 206	76.34 214

test2356		63.96 211	64.10 214	63.78 214	74.71 216	71.55 220	65.83 217	67.38 202	57.11 215	40.41 222	53.58 203	41.13 223	49.35 216	77.00 209	77.57 206	85.01 214	70.79 219

testus		63.31 215	64.48 213	61.94 218	73.99 218	71.99 219	63.56 221	63.25 217	57.01 216	39.41 226	54.38 199	38.73 227	46.24 221	77.01 208	77.93 203	85.20 213	74.29 217

DeepMVS_CX						48.31 234	48.03 230	26.08 233	56.42 217	25.77 233	47.51 211	31.31 232	51.30 212	48.49 230		53.61 232	61.52 227

MIMVSNet1		65.00 210	66.24 210	63.55 215	58.41 231	80.01 209	69.00 213	74.03 177	55.81 218	41.88 219	36.81 226	49.48 215	47.89 218	81.32 193	82.40 187	90.08 188	77.88 211

Gipuma		49.17 225	47.05 226	51.65 224	59.67 230	48.39 233	41.98 232	63.47 216	55.64 219	33.33 230	14.90 232	13.78 237	41.34 224	69.31 222	72.30 220	70.11 230	55.00 230
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
pmmvs6		74.83 194	72.89 200	77.09 174	82.11 191	87.50 160	80.88 185	76.97 150	52.79 220	61.91 169	46.66 213	60.49 167	69.28 183	86.74 128	85.46 166	91.39 174	90.56 145

gg-mvs-nofinetune		75.64 191	77.26 166	73.76 196	87.92 109	92.20 87	87.32 101	64.67 215	51.92 221	35.35 228	46.44 214	77.05 82	71.97 170	92.64 42	91.02 50	95.34 50	89.53 150

LTVRE_ROB	74.41 16	75.78 190	74.72 195	77.02 176	85.88 128	89.22 140	82.44 169	77.17 148	50.57 222	45.45 215	65.44 130	52.29 210	81.25 75	85.50 152	87.42 116	89.94 189	92.62 89
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
gm-plane-assit		70.29 203	70.65 204	69.88 204	85.03 141	78.50 213	58.41 224	65.47 211	50.39 223	40.88 220	49.60 209	50.11 213	75.14 151	91.43 54	89.78 75	94.32 109	84.73 192

pmmvs3		61.89 217	61.74 217	62.06 217	64.30 225	70.83 222	64.22 218	52.14 230	48.78 224	44.47 216	41.67 220	41.70 222	63.03 201	76.06 211	76.02 210	84.18 217	77.14 213

1111		57.32 220	57.20 221	57.46 220	71.89 222	67.50 227	52.34 225	58.78 223	46.57 225	39.69 223	37.38 224	38.78 225	46.37 219	74.15 216	74.36 218	75.70 227	61.66 226

.test1245		41.43 228	38.48 229	44.88 226	71.89 222	67.50 227	52.34 225	58.78 223	46.57 225	39.69 223	37.38 224	38.78 225	46.37 219	74.15 216	1.18 233	0.20 237	3.76 235

PMVS	50.48 18	55.81 223	51.93 224	60.33 219	72.90 219	49.34 232	48.78 229	69.51 196	43.49 227	54.25 198	36.26 227	41.04 224	39.71 227	65.07 226	60.70 227	76.85 225	67.58 224
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
testmv		56.62 221	56.41 222	56.86 221	71.92 220	67.58 225	52.17 227	65.69 208	40.60 228	28.53 231	37.90 222	31.52 230	40.10 225	72.64 219	74.73 216	82.78 220	69.91 220

test1235678		56.61 222	56.40 223	56.86 221	71.92 220	67.58 225	52.17 227	65.69 208	40.58 229	28.52 232	37.89 223	31.49 231	40.10 225	72.64 219	74.72 217	82.78 220	69.90 221

Anonymous20231211		62.95 216	60.42 219	65.89 212	74.22 217	78.37 214	67.66 214	74.47 174	40.37 230	39.59 225	27.51 229	38.26 228	52.13 211	75.39 214	77.89 204	87.28 203	85.16 187

test12356		50.02 224	51.22 225	48.61 225	63.00 226	60.15 230	47.60 231	56.49 225	38.02 231	24.74 234	36.14 228	25.93 233	24.79 230	66.19 225	71.68 221	75.07 228	60.44 228

PMMVS2		41.68 227	44.74 227	38.10 228	46.97 234	52.32 231	40.63 233	48.08 231	35.51 232	7.36 239	26.86 230	24.64 234	16.72 233	55.24 229	59.03 228	68.85 231	59.59 229

no-one		44.14 226	43.91 228	44.40 227	59.91 229	61.10 229	34.07 234	60.09 222	27.71 233	14.44 236	19.11 231	19.28 235	23.90 232	47.36 231	66.69 225	73.98 229	66.11 225

EMVS		30.49 231	25.44 232	36.39 230	51.47 232	29.89 237	20.17 237	54.00 229	26.49 234	12.02 238	13.94 235	8.84 238	34.37 228	25.04 234	34.37 232	46.29 235	39.53 233

E-PMN		31.40 229	26.80 231	36.78 229	51.39 233	29.96 236	20.20 236	54.17 227	25.93 235	12.75 237	14.73 233	8.58 239	34.10 229	27.36 233	37.83 231	48.07 234	43.18 232

MVE	30.17 19	30.88 230	33.52 230	27.80 233	23.78 236	39.16 235	18.69 238	46.90 232	21.88 236	15.39 235	14.37 234	7.31 240	24.41 231	41.63 232	56.22 229	37.64 236	54.07 231
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
testmvs		1.03 232	1.63 233	0.34 234	0.09 238	0.35 239	0.61 240	0.16 235	1.49 237	0.10 241	3.15 236	0.15 241	0.86 236	1.32 235	1.18 233	0.20 237	3.76 235

test123		0.87 233	1.40 234	0.25 235	0.03 239	0.25 240	0.35 241	0.08 237	1.21 238	0.05 242	2.84 237	0.03 242	0.89 235	0.43 236	1.16 235	0.13 239	3.87 234

sosnet-low-res		0.00 234	0.00 235	0.00 236	0.00 240	0.00 241	0.00 242	0.00 238	0.00 239	0.00 243	0.00 238	0.00 243	0.00 238	0.00 237	0.00 236	0.00 240	0.00 237

sosnet		0.00 234	0.00 235	0.00 236	0.00 240	0.00 241	0.00 242	0.00 238	0.00 239	0.00 243	0.00 238	0.00 243	0.00 238	0.00 237	0.00 236	0.00 240	0.00 237

MTAPA										92.97 2		91.03 16

MTMP										93.14 1		90.21 23

Patchmatch-RL test							8.55 239

XVS					93.11 53	96.70 20	91.91 48			83.95 42		88.82 33				95.79 28

X-MVStestdata					93.11 53	96.70 20	91.91 48			83.95 42		88.82 33				95.79 28

mPP-MVS					97.06 12							88.08 38

Patchmtry						85.54 186	82.30 171	68.23 198		65.37 141

High-resolution multi-view benchmark