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.

SD-MVS		98.52 4	98.77 3	98.23 11	98.15 44	99.26 18	98.79 23	97.59 10	98.52 1	96.25 11	97.99 11	99.75 2	99.01 3	98.27 22	97.97 23	99.59 4	99.63 1

CSCG		97.44 28	97.18 35	97.75 23	99.47 5	99.52 3	98.55 28	95.41 34	97.69 18	95.72 15	94.29 46	95.53 53	98.10 26	96.20 99	97.38 43	99.24 76	99.62 2

TSAR-MVS + MP.		98.49 5	98.78 2	98.15 15	98.14 45	99.17 25	99.34 3	97.18 23	98.44 3	95.72 15	97.84 12	99.28 8	98.87 6	99.05 1	98.05 20	99.66 1	99.60 3

TSAR-MVS + ACMM		97.71 24	98.60 8	96.66 35	98.64 35	99.05 29	98.85 22	97.23 22	98.45 2	89.40 73	97.51 19	99.27 9	96.88 56	98.53 11	97.81 30	98.96 112	99.59 4

APDe-MVS		98.87 1	98.96 1	98.77 1	99.58 1	99.53 2	99.44 1	97.81 1	98.22 7	97.33 2	98.70 2	99.33 6	98.86 7	98.96 3	98.40 10	99.63 3	99.57 5

SteuartSystems-ACMMP		98.38 10	98.71 5	97.99 19	99.34 16	99.46 5	99.34 3	97.33 19	97.31 29	94.25 25	98.06 9	99.17 12	98.13 25	98.98 2	98.46 8	99.55 9	99.54 6
Skip Steuart: Steuart Systems R&D Blog.
canonicalmvs		95.25 55	95.45 58	95.00 59	95.27 91	98.72 65	96.89 55	89.82 99	96.51 46	90.84 53	93.72 47	86.01 97	97.66 34	95.78 111	97.94 25	99.54 11	99.50 7

EPP-MVSNet		95.27 54	96.18 50	94.20 76	94.88 104	98.64 69	94.97 105	90.70 89	95.34 76	89.67 69	91.66 69	93.84 58	95.42 77	97.32 49	97.00 50	99.58 6	99.47 8

HSP-MVS		98.59 2	98.65 6	98.52 3	99.44 9	99.57 1	99.34 3	97.65 6	97.36 28	96.62 8	98.49 5	99.65 4	98.67 16	98.60 10	97.44 40	99.40 49	99.46 9

ESAPD		98.59 2	98.77 3	98.39 6	99.46 7	99.50 4	99.11 13	97.80 2	97.20 32	96.06 13	98.56 3	99.83 1	98.43 22	98.84 6	98.03 22	99.45 30	99.45 10

PVSNet_Blended_VisFu		94.77 63	95.54 57	93.87 82	96.48 64	98.97 40	94.33 118	91.84 79	94.93 86	90.37 60	85.04 126	94.99 54	90.87 145	98.12 30	97.30 46	99.30 67	99.45 10

MVS_0304		96.31 43	96.91 41	95.62 48	97.21 57	99.20 23	98.55 28	93.10 54	97.04 38	89.73 67	90.30 83	96.35 45	95.71 70	98.14 28	97.93 27	99.38 53	99.40 12

MSLP-MVS++		98.04 19	97.93 28	98.18 12	99.10 22	99.09 28	98.34 32	96.99 26	97.54 24	96.60 9	94.82 42	98.45 29	98.89 5	97.46 47	98.77 4	99.17 88	99.37 13

DeepC-MVS	94.87 4	96.76 41	96.50 45	97.05 31	98.21 43	99.28 16	98.67 24	97.38 15	97.31 29	90.36 61	89.19 91	93.58 61	98.19 24	98.31 20	98.50 6	99.51 17	99.36 14
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
CANet		96.84 38	97.20 33	96.42 36	97.92 47	99.24 22	98.60 26	93.51 46	97.11 35	93.07 33	91.16 73	97.24 38	96.21 66	98.24 25	98.05 20	99.22 82	99.35 15

3Dnovator+	93.91 7	97.23 31	97.22 32	97.24 28	98.89 30	98.85 58	98.26 33	93.25 51	97.99 11	95.56 18	90.01 87	98.03 34	98.05 27	97.91 36	98.43 9	99.44 40	99.35 15

TSAR-MVS + GP.		97.45 27	98.36 14	96.39 37	95.56 76	98.93 48	97.74 43	93.31 48	97.61 22	94.24 26	98.44 7	99.19 10	98.03 28	97.60 43	97.41 42	99.44 40	99.33 17

UGNet		94.92 56	96.63 43	92.93 96	96.03 70	98.63 71	94.53 115	91.52 84	96.23 53	90.03 63	92.87 56	96.10 50	86.28 192	96.68 73	96.60 59	99.16 91	99.32 18
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
MCST-MVS		98.20 14	98.36 14	98.01 18	99.40 11	99.05 29	99.00 18	97.62 8	97.59 23	93.70 29	97.42 22	99.30 7	98.77 12	98.39 19	97.48 38	99.59 4	99.31 19

FC-MVSNet-train		93.85 78	93.91 83	93.78 84	94.94 103	96.79 114	94.29 119	91.13 86	93.84 101	88.26 84	90.40 82	85.23 104	94.65 85	96.54 83	95.31 103	99.38 53	99.28 20

X-MVS		97.84 20	98.19 23	97.42 26	99.40 11	99.35 9	99.06 15	97.25 20	97.38 27	90.85 50	96.06 31	98.72 23	98.53 20	98.41 18	98.15 16	99.46 26	99.28 20

EPNet		96.27 44	96.97 38	95.46 51	98.47 38	98.28 80	97.41 48	93.67 44	95.86 66	92.86 38	97.51 19	93.79 59	91.76 129	97.03 57	97.03 49	98.61 159	99.28 20
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
CP-MVS		98.32 13	98.34 17	98.29 8	99.34 16	99.30 14	99.15 11	97.35 16	97.49 25	95.58 17	97.72 14	98.62 27	98.82 10	98.29 21	97.67 33	99.51 17	99.28 20

DELS-MVS		96.06 45	96.04 51	96.07 44	97.77 49	99.25 20	98.10 36	93.26 49	94.42 91	92.79 39	88.52 98	93.48 62	95.06 81	98.51 12	98.83 1	99.45 30	99.28 20
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
HFP-MVS		98.48 6	98.62 7	98.32 7	99.39 14	99.33 13	99.27 8	97.42 13	98.27 5	95.25 19	98.34 8	98.83 20	99.08 1	98.26 23	98.08 19	99.48 21	99.26 25

HPM-MVS++		98.34 12	98.47 11	98.18 12	99.46 7	99.15 26	99.10 14	97.69 5	97.67 19	94.93 22	97.62 15	99.70 3	98.60 17	98.45 15	97.46 39	99.31 65	99.26 25

3Dnovator	93.79 8	97.08 33	97.20 33	96.95 33	99.09 23	99.03 34	98.20 34	93.33 47	97.99 11	93.82 28	90.61 81	96.80 42	97.82 30	97.90 37	98.78 3	99.47 24	99.26 25

MP-MVS		98.09 18	98.30 20	97.84 22	99.34 16	99.19 24	99.23 10	97.40 14	97.09 36	93.03 36	97.58 17	98.85 19	98.57 19	98.44 17	97.69 32	99.48 21	99.23 28

IS_MVSNet		95.28 53	96.43 47	93.94 79	95.30 89	99.01 38	95.90 83	91.12 87	94.13 97	87.50 93	91.23 72	94.45 57	94.17 92	98.45 15	98.50 6	99.65 2	99.23 28

ACMMPR		98.40 9	98.49 9	98.28 9	99.41 10	99.40 6	99.36 2	97.35 16	98.30 4	95.02 21	97.79 13	98.39 30	99.04 2	98.26 23	98.10 17	99.50 19	99.22 30

APD-MVS		98.36 11	98.32 18	98.41 5	99.47 5	99.26 18	99.12 12	97.77 4	96.73 42	96.12 12	97.27 23	98.88 18	98.46 21	98.47 14	98.39 11	99.52 13	99.22 30
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
UniMVSNet (Re)		90.03 135	89.61 143	90.51 121	89.97 160	96.12 130	92.32 150	89.26 106	90.99 144	80.95 125	78.25 158	75.08 157	91.14 136	93.78 143	93.87 138	99.41 47	99.21 32

DeepPCF-MVS	95.28 2	97.00 35	98.35 16	95.42 52	97.30 55	98.94 42	94.82 110	96.03 32	98.24 6	92.11 44	95.80 33	98.64 26	95.51 75	98.95 4	98.66 5	96.78 194	99.20 33

conf0.05thres1000		92.47 105	91.39 128	93.73 85	95.21 94	98.52 74	95.66 91	91.56 82	90.87 146	84.27 104	82.79 141	76.12 149	96.29 64	96.59 78	95.68 93	99.39 51	99.19 34

ACMMP_Plus		98.20 14	98.49 9	97.85 21	99.50 3	99.40 6	99.26 9	97.64 7	97.47 26	92.62 42	97.59 16	99.09 15	98.71 14	98.82 8	97.86 29	99.40 49	99.19 34

train_agg		97.65 25	98.06 25	97.18 29	98.94 27	98.91 54	98.98 21	97.07 25	96.71 43	90.66 55	97.43 21	99.08 17	98.20 23	97.96 35	97.14 48	99.22 82	99.19 34

QAPM		96.78 40	97.14 36	96.36 38	99.05 24	99.14 27	98.02 37	93.26 49	97.27 31	90.84 53	91.16 73	97.31 37	97.64 35	97.70 41	98.20 14	99.33 60	99.18 37

MPTG		98.43 8	98.31 19	98.57 2	99.48 4	99.40 6	99.32 6	97.62 8	97.70 16	96.67 6	96.59 27	99.09 15	98.86 7	98.65 9	97.56 36	99.45 30	99.17 38

anonymousdsp		88.90 150	91.00 132	86.44 193	88.74 200	95.97 134	90.40 190	82.86 176	88.77 173	67.33 207	81.18 148	81.44 125	90.22 170	96.23 97	94.27 129	99.12 97	99.16 39

ACMMP		97.37 29	97.48 31	97.25 27	98.88 31	99.28 16	98.47 30	96.86 28	97.04 38	92.15 43	97.57 18	96.05 51	97.67 33	97.27 50	95.99 78	99.46 26	99.14 40
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 1792x2688		92.66 103	92.49 107	92.85 97	97.13 58	98.89 57	95.90 83	88.50 114	95.32 77	83.31 110	71.99 201	88.96 84	94.10 94	96.69 72	96.49 60	98.15 176	99.10 41

DeepC-MVS_fast	96.13 1	98.13 16	98.27 21	97.97 20	99.16 21	99.03 34	99.05 16	97.24 21	98.22 7	94.17 27	95.82 32	98.07 32	98.69 15	98.83 7	98.80 2	99.52 13	99.10 41
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
PGM-MVS		97.81 21	98.11 24	97.46 25	99.55 2	99.34 12	99.32 6	94.51 39	96.21 54	93.07 33	98.05 10	97.95 35	98.82 10	98.22 26	97.89 28	99.48 21	99.09 43

CNVR-MVS		98.47 7	98.46 12	98.48 4	99.40 11	99.05 29	99.02 17	97.54 11	97.73 14	96.65 7	97.20 24	99.13 13	98.85 9	98.91 5	98.10 17	99.41 47	99.08 44

PVSNet_BlendedMVS		95.41 51	95.28 59	95.57 49	97.42 53	99.02 36	95.89 85	93.10 54	96.16 55	93.12 31	91.99 65	85.27 102	94.66 83	98.09 32	97.34 44	99.24 76	99.08 44

PVSNet_Blended		95.41 51	95.28 59	95.57 49	97.42 53	99.02 36	95.89 85	93.10 54	96.16 55	93.12 31	91.99 65	85.27 102	94.66 83	98.09 32	97.34 44	99.24 76	99.08 44

IB-MVS	89.56 15	91.71 112	92.50 106	90.79 118	95.94 72	98.44 77	87.05 204	91.38 85	93.15 109	92.98 37	84.78 127	85.14 105	78.27 212	92.47 165	94.44 127	99.10 99	99.08 44
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
ACMP	92.88 9	94.43 67	94.38 75	94.50 73	96.01 71	97.69 93	95.85 88	92.09 73	95.74 70	89.12 76	95.14 39	82.62 121	94.77 82	95.73 112	94.67 115	99.14 94	99.06 48
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
CLD-MVS		94.79 61	94.36 76	95.30 54	95.21 94	97.46 96	97.23 50	92.24 70	96.43 47	91.77 46	92.69 57	84.31 108	96.06 67	95.52 116	95.03 108	99.31 65	99.06 48
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
UniMVSNet_NR-MVSNet		90.35 130	89.96 139	90.80 117	89.66 163	95.83 142	92.48 142	90.53 92	90.96 145	79.57 129	79.33 155	77.14 143	93.21 110	92.91 159	94.50 126	99.37 56	99.05 50

DU-MVS		89.67 139	88.84 149	90.63 120	89.26 187	95.61 147	92.48 142	89.91 97	91.22 141	79.57 129	77.72 159	71.18 194	93.21 110	92.53 163	94.57 120	99.35 57	99.05 50

CPTT-MVS		97.78 22	97.54 29	98.05 17	98.91 29	99.05 29	99.00 18	96.96 27	97.14 34	95.92 14	95.50 35	98.78 22	98.99 4	97.20 52	96.07 74	98.54 163	99.04 52

tfpnnormal		88.50 153	87.01 189	90.23 124	91.36 143	95.78 144	92.74 137	90.09 95	83.65 212	76.33 145	71.46 206	69.58 203	91.84 127	95.54 115	94.02 134	99.06 106	99.03 53

LGP-MVS_train		94.12 72	94.62 69	93.53 88	96.44 65	97.54 94	97.40 49	91.84 79	94.66 87	81.09 124	95.70 34	83.36 117	95.10 80	96.36 93	95.71 92	99.32 62	99.03 53

PHI-MVS		97.78 22	98.44 13	97.02 32	98.73 32	99.25 20	98.11 35	95.54 33	96.66 45	92.79 39	98.52 4	99.38 5	97.50 37	97.84 38	98.39 11	99.45 30	99.03 53

MVS_111021_HR		97.04 34	98.20 22	95.69 47	98.44 40	99.29 15	96.59 71	93.20 52	97.70 16	89.94 65	98.46 6	96.89 40	96.71 59	98.11 31	97.95 24	99.27 71	99.01 56

HQP-MVS		94.43 67	94.57 70	94.27 75	96.41 66	97.23 101	96.89 55	93.98 41	95.94 63	83.68 108	95.01 41	84.46 107	95.58 73	95.47 117	94.85 114	99.07 103	99.00 57

NR-MVSNet		89.34 142	88.66 150	90.13 129	90.40 153	95.61 147	93.04 135	89.91 97	91.22 141	78.96 132	77.72 159	68.90 207	89.16 176	94.24 139	93.95 135	99.32 62	98.99 58

MAR-MVS		95.50 47	95.60 55	95.39 53	98.67 34	98.18 86	95.89 85	89.81 100	94.55 90	91.97 45	92.99 53	90.21 76	97.30 39	96.79 66	97.49 37	98.72 150	98.99 58
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
Effi-MVS+		92.93 99	93.86 85	91.86 104	94.07 117	98.09 90	95.59 92	85.98 144	94.27 94	79.54 131	91.12 76	81.81 123	96.71 59	96.67 74	96.06 75	99.27 71	98.98 60

CP-MVSNet		87.89 171	87.27 180	88.62 149	89.30 183	95.06 172	90.60 188	85.78 146	87.43 193	75.98 147	74.60 175	68.14 209	90.76 146	93.07 157	93.60 143	99.30 67	98.98 60

NCCC		98.10 17	98.05 26	98.17 14	99.38 15	99.05 29	99.00 18	97.53 12	98.04 10	95.12 20	94.80 43	99.18 11	98.58 18	98.49 13	97.78 31	99.39 51	98.98 60

ACMH	90.77 13	91.51 117	91.63 123	91.38 110	95.62 75	96.87 108	91.76 175	89.66 102	91.58 138	78.67 133	86.73 110	78.12 135	93.77 98	94.59 129	94.54 123	98.78 141	98.98 60
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
HyFIR lowres test		92.03 107	91.55 125	92.58 101	97.13 58	98.72 65	94.65 113	86.54 136	93.58 105	82.56 113	67.75 216	90.47 75	95.67 71	95.87 107	95.54 99	98.91 116	98.93 64

CDPH-MVS		96.84 38	97.49 30	96.09 42	98.92 28	98.85 58	98.61 25	95.09 35	96.00 61	87.29 96	95.45 37	97.42 36	97.16 42	97.83 39	97.94 25	99.44 40	98.92 65

TranMVSNet+NR-MVSNet		89.23 145	88.48 153	90.11 130	89.07 193	95.25 168	92.91 136	90.43 93	90.31 153	77.10 139	76.62 162	71.57 192	91.83 128	92.12 176	94.59 119	99.32 62	98.92 65

PS-CasMVS		87.33 186	86.68 194	88.10 159	89.22 192	94.93 177	90.35 191	85.70 147	86.44 199	74.01 169	73.43 194	66.59 215	90.04 171	92.92 158	93.52 144	99.28 69	98.91 67

Vis-MVSNet		92.77 101	95.00 66	90.16 126	94.10 116	98.79 60	94.76 112	88.26 115	92.37 130	79.95 127	88.19 100	91.58 68	84.38 202	97.59 44	97.58 35	99.52 13	98.91 67
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
Baseline_NR-MVSNet		89.27 144	88.01 159	90.73 119	89.26 187	93.71 196	92.71 139	89.78 101	90.73 148	81.28 123	73.53 193	72.85 173	92.30 120	92.53 163	93.84 140	99.07 103	98.88 69

OpenMVS	92.33 11	95.50 47	95.22 61	95.82 46	98.98 25	98.97 40	97.67 45	93.04 57	94.64 88	89.18 75	84.44 130	94.79 55	96.79 57	97.23 51	97.61 34	99.24 76	98.88 69

LTVRE_ROB	87.32 16	87.55 177	88.25 155	86.73 188	90.66 150	95.80 143	93.05 134	84.77 160	83.35 213	60.32 219	83.12 139	67.39 210	93.32 107	94.36 136	94.86 113	98.28 174	98.87 71
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
MVS_111021_LR		97.16 32	98.01 27	96.16 41	98.47 38	98.98 39	96.94 54	93.89 42	97.64 21	91.44 47	98.89 1	96.41 44	97.20 41	98.02 34	97.29 47	99.04 108	98.85 72

WR-MVS_H		87.93 167	87.85 167	88.03 165	89.62 165	95.58 151	90.47 189	85.55 149	87.20 195	76.83 141	74.42 179	72.67 184	86.37 191	93.22 154	93.04 151	99.33 60	98.83 73

v748		85.88 198	85.66 201	86.14 196	88.03 203	94.63 187	87.02 205	84.59 164	84.30 209	74.56 163	70.94 208	67.27 211	83.94 206	90.96 195	92.74 173	98.71 151	98.81 74

WR-MVS		87.93 167	88.09 157	87.75 171	89.26 187	95.28 165	90.81 186	86.69 135	88.90 170	75.29 154	74.31 181	73.72 163	85.19 198	92.26 166	93.32 148	99.27 71	98.81 74

DI_MVS_plusplus_trai		94.01 75	93.63 91	94.44 74	94.54 110	98.26 83	97.51 47	90.63 90	95.88 65	89.34 74	80.54 151	89.36 80	95.48 76	96.33 94	96.27 67	99.17 88	98.78 76

v7n		86.43 194	86.52 197	86.33 194	87.91 205	94.93 177	90.15 192	83.05 174	86.57 197	70.21 198	71.48 205	66.78 213	87.72 183	94.19 141	92.96 154	98.92 115	98.76 77

tfpn		92.91 100	91.44 127	94.63 72	95.42 77	98.92 52	96.41 77	92.10 72	93.19 108	87.34 95	86.85 108	69.20 205	97.01 53	96.88 58	96.28 66	99.47 24	98.75 78

view800		93.45 94	92.37 114	94.71 70	95.42 77	98.92 52	96.51 74	92.19 71	93.14 110	87.62 91	86.72 111	76.54 148	97.08 52	96.86 59	95.74 91	99.45 30	98.70 79

diffmvs		94.83 58	95.64 54	93.89 81	94.73 106	97.96 92	96.49 75	89.13 109	96.82 41	89.47 71	91.66 69	93.63 60	95.15 79	94.76 127	95.93 79	98.36 173	98.69 80

Effi-MVS+-dtu		91.78 111	93.59 93	89.68 134	92.44 136	97.11 103	94.40 117	84.94 159	92.43 126	75.48 150	91.09 77	83.75 112	93.55 103	96.61 75	95.47 100	97.24 190	98.67 81

thres600view7		93.49 93	92.37 114	94.79 69	95.42 77	98.93 48	96.58 72	92.31 63	93.04 112	87.88 89	86.62 114	76.94 145	97.09 51	96.82 61	95.63 95	99.45 30	98.63 82

view600		93.50 92	92.39 113	94.80 68	95.41 80	98.93 48	96.60 70	92.30 68	93.09 111	87.96 88	86.67 113	76.97 144	97.12 45	96.83 60	95.64 94	99.43 45	98.62 83

abl_6				96.82 34	98.60 36	98.74 62	97.74 43	93.73 43	96.25 52	94.37 24	94.55 45	98.60 28	97.25 40			99.27 71	98.61 84

IterMVS-LS		92.56 104	93.18 99	91.84 105	93.90 119	94.97 175	94.99 104	86.20 140	94.18 96	82.68 112	85.81 123	87.36 91	94.43 87	95.31 119	96.02 77	98.87 118	98.60 85
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
tfpn111		94.05 73	93.34 98	94.88 63	95.33 83	98.94 42	96.82 58	92.31 63	92.63 117	88.26 84	92.61 58	78.01 137	97.12 45	96.82 61	95.85 86	99.45 30	98.56 86

conf200view11		93.64 82	92.57 102	94.88 63	95.33 83	98.94 42	96.82 58	92.31 63	92.63 117	88.26 84	87.21 104	78.01 137	97.12 45	96.82 61	95.85 86	99.45 30	98.56 86

tfpn200view9		93.64 82	92.57 102	94.89 62	95.33 83	98.94 42	96.82 58	92.31 63	92.63 117	88.29 81	87.21 104	78.01 137	97.12 45	96.82 61	95.85 86	99.45 30	98.56 86

thres400		93.56 88	92.43 110	94.87 66	95.40 81	98.91 54	96.70 67	92.38 62	92.93 114	88.19 87	86.69 112	77.35 142	97.13 43	96.75 70	95.85 86	99.42 46	98.56 86

MVS_Test		94.82 59	95.66 53	93.84 83	94.79 105	98.35 79	96.49 75	89.10 110	96.12 57	87.09 97	92.58 60	90.61 74	96.48 62	96.51 87	96.89 53	99.11 98	98.54 90

Vis-MVSNet (Re-imp)		94.46 66	96.24 49	92.40 102	95.23 93	98.64 69	95.56 95	90.99 88	94.42 91	85.02 102	90.88 79	94.65 56	88.01 182	98.17 27	98.37 13	99.57 8	98.53 91

v52		86.57 192	86.63 195	86.50 191	87.47 209	94.89 180	89.90 193	83.39 171	86.36 200	71.17 193	71.53 204	71.65 189	88.34 180	91.14 193	92.32 183	98.74 149	98.52 92

V4		86.56 193	86.61 196	86.50 191	87.49 208	94.90 179	89.87 194	83.39 171	86.25 201	71.20 192	71.57 203	71.58 191	88.30 181	91.14 193	92.31 184	98.75 148	98.52 92

Fast-Effi-MVS+		91.87 109	92.08 117	91.62 109	92.91 132	97.21 102	94.93 106	84.60 163	93.61 103	81.49 122	83.50 136	78.95 132	96.62 61	96.55 82	96.22 71	99.16 91	98.51 94

MVSTER		94.89 57	95.07 64	94.68 71	94.71 107	96.68 117	97.00 52	90.57 91	95.18 83	93.05 35	95.21 38	86.41 94	93.72 99	97.59 44	95.88 84	99.00 109	98.50 95

thres200		93.62 85	92.54 104	94.88 63	95.36 82	98.93 48	96.75 66	92.31 63	92.84 115	88.28 83	86.99 107	77.81 141	97.13 43	96.82 61	95.92 80	99.45 30	98.49 96

v1921920		87.31 187	87.13 186	87.52 181	88.87 197	94.72 183	91.96 173	84.59 164	88.28 184	69.86 201	72.50 199	70.03 202	91.10 137	93.33 152	92.61 179	98.71 151	98.44 97

v144192		87.40 183	87.20 182	87.64 176	88.89 195	94.88 181	91.65 176	84.70 162	87.80 188	71.17 193	73.20 196	70.91 195	90.75 147	92.69 161	92.49 180	98.71 151	98.43 98

v1192		87.51 178	87.31 179	87.74 172	89.04 194	94.87 182	92.07 166	85.03 157	88.49 176	70.32 196	72.65 198	70.35 199	91.21 135	93.59 144	92.80 161	98.78 141	98.42 99

v7		88.18 161	88.01 159	88.39 151	89.45 172	95.14 171	92.36 146	85.37 152	89.29 164	72.94 183	73.98 188	72.77 176	91.38 133	93.59 144	92.87 157	98.82 121	98.42 99

v10		88.00 165	87.96 164	88.05 163	89.44 173	94.68 184	92.36 146	83.35 173	89.37 163	72.96 181	73.98 188	72.79 175	91.35 134	93.59 144	92.88 156	98.81 125	98.42 99

thres100view900		93.55 91	92.47 109	94.81 67	95.33 83	98.74 62	96.78 65	92.30 68	92.63 117	88.29 81	87.21 104	78.01 137	96.78 58	96.38 91	95.92 80	99.38 53	98.40 102

AdaColmap		97.53 26	96.93 39	98.24 10	99.21 19	98.77 61	98.47 30	97.34 18	96.68 44	96.52 10	95.11 40	96.12 49	98.72 13	97.19 54	96.24 70	99.17 88	98.39 103

PCF-MVS	93.95 6	95.65 46	95.14 62	96.25 39	97.73 51	98.73 64	97.59 46	97.13 24	92.50 125	89.09 77	89.85 88	96.65 43	96.90 55	94.97 126	94.89 112	99.08 101	98.38 104
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
Fast-Effi-MVS+-dtu		91.19 120	93.64 90	88.33 153	92.19 139	96.46 123	93.99 122	81.52 192	92.59 123	71.82 187	92.17 64	85.54 100	91.68 130	95.73 112	94.64 117	98.80 128	98.34 105

v1144		87.92 170	87.79 168	88.07 160	89.27 186	95.15 170	92.17 164	85.62 148	88.52 175	71.52 188	73.80 191	72.40 187	91.06 139	93.54 149	92.80 161	98.81 125	98.33 106

V42		88.31 158	87.95 165	88.73 148	89.44 173	95.34 164	92.23 161	87.21 130	88.83 171	74.49 165	74.89 174	73.43 169	90.41 164	92.08 183	92.77 167	98.60 161	98.33 106

CDS-MVSNet		92.77 101	93.60 92	91.80 106	92.63 134	96.80 111	95.24 101	89.14 108	90.30 154	84.58 103	86.76 109	90.65 73	90.42 161	95.89 106	96.49 60	98.79 134	98.32 108
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
PEN-MVS		87.22 188	86.50 198	88.07 160	88.88 196	94.44 190	90.99 185	86.21 138	86.53 198	73.66 175	74.97 173	66.56 216	89.42 175	91.20 192	93.48 145	99.24 76	98.31 109

conf0.01		93.33 95	91.89 120	95.00 59	95.32 87	98.94 42	96.82 58	92.41 61	92.63 117	88.91 79	88.02 102	72.75 178	97.12 45	96.78 67	95.85 86	99.44 40	98.27 110

v1240		86.89 189	86.75 193	87.06 186	88.75 199	94.65 186	91.30 182	84.05 167	87.49 192	68.94 205	71.96 202	68.86 208	90.65 155	93.33 152	92.72 175	98.67 155	98.24 111

v8		88.21 160	87.94 166	88.51 150	89.62 165	95.01 174	92.31 151	84.99 158	88.94 169	74.70 162	75.03 169	73.51 167	90.67 154	92.11 179	92.74 173	98.80 128	98.24 111

v1neww		88.41 155	88.00 162	88.89 141	89.61 167	95.44 157	92.31 151	87.65 125	89.09 165	74.30 167	75.02 170	73.42 170	90.68 152	92.12 176	92.77 167	98.79 134	98.18 113

v7new		88.41 155	88.00 162	88.89 141	89.61 167	95.44 157	92.31 151	87.65 125	89.09 165	74.30 167	75.02 170	73.42 170	90.68 152	92.12 176	92.77 167	98.79 134	98.18 113

v6		88.43 154	88.01 159	88.92 140	89.60 169	95.43 159	92.36 146	87.66 124	89.07 167	74.50 164	75.06 168	73.47 168	90.59 157	92.11 179	92.76 171	98.79 134	98.18 113

conf0.002		93.20 98	91.63 123	95.02 57	95.31 88	98.94 42	96.82 58	92.43 60	92.63 117	88.99 78	88.16 101	70.49 197	97.12 45	96.77 68	96.30 62	99.44 40	98.16 116

CNLPA		96.90 37	96.28 48	97.64 24	98.56 37	98.63 71	96.85 57	96.60 30	97.73 14	97.08 4	89.78 89	96.28 48	97.80 32	96.73 71	96.63 58	98.94 113	98.14 117

v1141		88.17 162	87.69 171	88.74 146	89.44 173	95.41 160	92.25 159	87.98 119	88.38 179	73.54 178	74.43 178	72.71 182	90.45 159	92.08 183	92.72 175	98.79 134	98.09 118

divwei89l23v2f112		88.17 162	87.69 171	88.74 146	89.44 173	95.41 160	92.26 157	87.97 121	88.29 183	73.57 177	74.45 177	72.75 178	90.42 161	92.08 183	92.72 175	98.81 125	98.09 118

v1		88.17 162	87.66 173	88.77 145	89.44 173	95.40 162	92.29 154	87.98 119	88.21 186	73.75 172	74.41 180	72.75 178	90.36 167	92.07 186	92.71 178	98.80 128	98.09 118

PLC	94.95 3	97.37 29	96.77 42	98.07 16	98.97 26	98.21 85	97.94 40	96.85 29	97.66 20	97.58 1	93.33 51	96.84 41	98.01 29	97.13 56	96.20 73	99.09 100	98.01 121
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
v2v482		88.25 159	87.71 170	88.88 143	89.23 191	95.28 165	92.10 165	87.89 123	88.69 174	73.31 180	75.32 165	71.64 190	91.89 126	92.10 182	92.92 155	98.86 120	97.99 122

ACMH+	90.88 12	91.41 118	91.13 130	91.74 107	95.11 98	96.95 105	93.13 133	89.48 105	92.42 127	79.93 128	85.13 125	78.02 136	93.82 97	93.49 150	93.88 137	98.94 113	97.99 122

v148		87.51 178	86.79 191	88.36 152	89.39 179	95.21 169	89.84 195	88.20 117	87.61 191	77.56 136	73.38 195	70.32 200	86.80 189	90.70 196	92.31 184	98.37 172	97.98 124

v11		87.58 176	87.50 178	87.67 175	89.34 181	91.91 209	92.22 163	81.63 189	89.01 168	72.95 182	74.11 186	72.51 186	91.08 138	94.01 142	93.00 153	98.77 145	97.93 125

v12		87.38 184	87.13 186	87.68 174	89.30 183	91.92 208	92.01 172	81.94 187	88.35 180	73.69 174	74.10 187	72.57 185	90.33 169	92.23 168	92.82 159	98.80 128	97.91 126

v16		87.87 172	87.60 176	88.19 156	89.70 162	92.01 204	92.37 145	82.54 181	89.67 159	75.00 160	75.02 170	73.65 164	90.73 150	92.14 175	92.80 161	98.77 145	97.90 127

v17		87.83 173	87.56 177	88.13 158	89.65 164	92.02 203	92.34 149	82.55 180	89.38 162	74.76 161	75.14 167	73.59 165	90.70 151	92.15 174	92.78 165	98.78 141	97.89 128

v13		87.34 185	87.11 188	87.62 177	89.30 183	91.91 209	92.04 168	81.86 188	88.35 180	73.36 179	73.88 190	72.69 183	90.34 168	92.23 168	92.82 159	98.80 128	97.88 129

v18		87.93 167	87.61 175	88.31 154	89.74 161	92.04 202	92.59 141	82.71 178	89.70 157	75.32 153	75.23 166	73.55 166	90.74 148	92.11 179	92.77 167	98.78 141	97.87 130

V9		87.41 182	87.15 185	87.72 173	89.33 182	91.93 207	92.23 161	82.02 186	88.35 180	73.59 176	74.13 185	72.77 176	90.37 166	92.21 170	92.80 161	98.79 134	97.86 131

V14		87.47 180	87.19 183	87.80 170	89.37 180	91.95 206	92.25 159	82.12 185	88.39 178	73.83 171	74.31 181	72.84 174	90.44 160	92.20 171	92.78 165	98.80 128	97.84 132

v15		87.46 181	87.16 184	87.81 169	89.41 178	91.96 205	92.26 157	82.28 184	88.42 177	73.72 173	74.29 183	72.73 181	90.41 164	92.17 173	92.76 171	98.79 134	97.83 133

tfpn1000		94.14 71	94.54 71	93.67 86	95.27 91	98.50 76	95.36 99	91.84 79	96.31 50	87.38 94	92.98 54	84.04 109	92.60 116	96.49 88	95.62 96	99.55 9	97.82 134

TAPA-MVS	94.18 5	96.38 42	96.49 46	96.25 39	98.26 42	98.66 67	98.00 38	94.96 37	97.17 33	89.48 70	92.91 55	96.35 45	97.53 36	96.59 78	95.90 82	99.28 69	97.82 134
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
tfpn_ndepth		94.36 70	94.64 68	94.04 78	95.16 96	98.51 75	95.58 93	92.09 73	95.78 69	88.52 80	92.38 62	85.74 99	93.34 106	96.39 89	95.90 82	99.54 11	97.79 136

CANet_DTU		93.92 77	96.57 44	90.83 116	95.63 74	98.39 78	96.99 53	87.38 128	96.26 51	71.97 186	96.31 29	93.02 63	94.53 86	97.38 48	96.83 55	98.49 166	97.79 136

GBi-Net		93.81 79	94.18 79	93.38 90	91.34 144	95.86 139	96.22 79	88.68 111	95.23 80	90.40 57	86.39 117	91.16 69	94.40 89	96.52 84	96.30 62	99.21 85	97.79 136

test1		93.81 79	94.18 79	93.38 90	91.34 144	95.86 139	96.22 79	88.68 111	95.23 80	90.40 57	86.39 117	91.16 69	94.40 89	96.52 84	96.30 62	99.21 85	97.79 136

FMVSNet3		93.79 81	94.17 81	93.35 92	91.21 147	95.99 132	96.62 68	88.68 111	95.23 80	90.40 57	86.39 117	91.16 69	94.11 93	95.96 104	96.67 57	99.07 103	97.79 136

FMVSNet2		93.30 96	93.36 97	93.22 94	91.34 144	95.86 139	96.22 79	88.24 116	95.15 84	89.92 66	81.64 145	89.36 80	94.40 89	96.77 68	96.98 51	99.21 85	97.79 136

pm-mvs1		89.19 146	89.02 148	89.38 137	90.40 153	95.74 145	92.05 167	88.10 118	86.13 203	77.70 135	73.72 192	79.44 131	88.97 177	95.81 110	94.51 125	99.08 101	97.78 142

FMVSNet1		91.54 116	90.93 133	92.26 103	90.35 155	95.27 167	95.22 102	87.16 131	91.37 140	87.62 91	75.45 164	83.84 111	94.43 87	96.52 84	96.30 62	98.82 121	97.74 143

LS3D		95.46 49	95.14 62	95.84 45	97.91 48	98.90 56	98.58 27	97.79 3	97.07 37	83.65 109	88.71 94	88.64 86	97.82 30	97.49 46	97.42 41	99.26 75	97.72 144

OMC-MVS		97.00 35	96.92 40	97.09 30	98.69 33	98.66 67	97.85 41	95.02 36	98.09 9	94.47 23	93.15 52	96.90 39	97.38 38	97.16 55	96.82 56	99.13 95	97.65 145

DTE-MVSNet		86.67 191	86.09 199	87.35 183	88.45 202	94.08 194	90.65 187	86.05 143	86.13 203	72.19 185	74.58 176	66.77 214	87.61 185	90.31 198	93.12 150	99.13 95	97.62 146

CHOSEN 280x420		95.46 49	97.01 37	93.66 87	97.28 56	97.98 91	96.40 78	85.39 151	96.10 59	91.07 49	96.53 28	96.34 47	95.61 72	97.65 42	96.95 52	96.21 198	97.49 147

IterMVS		90.20 131	92.43 110	87.61 178	92.82 133	94.31 193	94.11 120	81.54 191	92.97 113	69.90 200	84.71 128	88.16 90	89.96 172	95.25 120	94.17 130	97.31 189	97.46 148
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
FC-MVSNet-test		91.63 113	93.82 88	89.08 139	92.02 140	96.40 126	93.26 131	87.26 129	93.72 102	77.26 138	88.61 97	89.86 78	85.50 195	95.72 114	95.02 109	99.16 91	97.44 149

gg-mvs-nofinetune		86.17 196	88.57 152	83.36 205	93.44 125	98.15 88	96.58 72	72.05 224	74.12 224	49.23 231	64.81 219	90.85 72	89.90 173	97.83 39	96.84 54	98.97 111	97.41 150

test-mter		90.95 122	93.54 96	87.93 168	90.28 156	96.80 111	91.44 177	82.68 179	92.15 135	74.37 166	89.57 90	88.23 89	90.88 144	96.37 92	94.31 128	97.93 183	97.37 151

ACMM	92.75 10	94.41 69	93.84 86	95.09 56	96.41 66	96.80 111	94.88 109	93.54 45	96.41 48	90.16 62	92.31 63	83.11 118	96.32 63	96.22 98	94.65 116	99.22 82	97.35 152
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
GA-MVS		89.28 143	90.75 136	87.57 179	91.77 141	96.48 122	92.29 154	87.58 127	90.61 151	65.77 210	84.48 129	76.84 146	89.46 174	95.84 108	93.68 142	98.52 164	97.34 153

pmmvs6		85.98 197	84.89 207	87.25 184	88.83 198	94.35 192	89.36 197	85.30 155	78.51 221	75.44 151	62.71 222	75.41 154	87.65 184	93.58 147	92.40 182	96.89 192	97.29 154

thresconf0.02		93.57 87	93.84 86	93.25 93	95.03 102	98.16 87	95.80 90	92.46 59	96.12 57	83.88 106	92.61 58	80.39 127	92.83 114	96.11 103	96.21 72	99.49 20	97.28 155

OPM-MVS		93.61 86	92.43 110	95.00 59	96.94 60	97.34 99	97.78 42	94.23 40	89.64 160	85.53 100	88.70 95	82.81 119	96.28 65	96.28 96	95.00 111	99.24 76	97.22 156

SixPastTwentyTwo		88.37 157	89.47 144	87.08 185	90.01 159	95.93 138	87.41 201	85.32 153	90.26 155	70.26 197	86.34 120	71.95 188	90.93 141	92.89 160	91.72 191	98.55 162	97.22 156

pmmvs5		87.83 173	88.09 157	87.51 182	89.59 170	95.48 152	89.75 196	84.73 161	86.07 205	71.44 189	80.57 150	70.09 201	90.74 148	94.47 132	92.87 157	98.82 121	97.10 158

CVMVSNet		89.77 138	91.66 122	87.56 180	93.21 130	95.45 154	91.94 174	89.22 107	89.62 161	69.34 204	83.99 133	85.90 98	84.81 200	94.30 137	95.28 104	96.85 193	97.09 159

PMMVS		94.61 65	95.56 56	93.50 89	94.30 113	96.74 115	94.91 108	89.56 104	95.58 72	87.72 90	96.15 30	92.86 64	96.06 67	95.47 117	95.02 109	98.43 171	97.09 159

CR-MVSNet		90.16 133	91.96 119	88.06 162	93.32 127	95.95 136	93.36 129	75.99 213	92.40 128	75.19 155	83.18 138	85.37 101	92.05 122	95.21 121	94.56 121	98.47 168	97.08 161

PatchT		89.13 147	91.71 121	86.11 197	92.92 131	95.59 149	83.64 211	75.09 217	91.87 137	75.19 155	82.63 142	85.06 106	92.05 122	95.21 121	94.56 121	97.76 185	97.08 161

tfpnview11		93.63 84	94.42 73	92.71 98	95.08 99	98.26 83	95.58 93	92.06 75	96.32 49	81.88 115	93.44 48	83.43 114	92.14 121	96.58 80	95.88 84	99.52 13	97.07 163

tfpn_n400		93.56 88	94.36 76	92.63 99	95.07 100	98.28 80	95.50 97	91.98 77	95.48 73	81.88 115	93.44 48	83.43 114	92.01 124	96.60 76	96.27 67	99.34 58	97.04 164

tfpnconf		93.56 88	94.36 76	92.63 99	95.07 100	98.28 80	95.50 97	91.98 77	95.48 73	81.88 115	93.44 48	83.43 114	92.01 124	96.60 76	96.27 67	99.34 58	97.04 164

UA-Net		93.96 76	95.95 52	91.64 108	96.06 69	98.59 73	95.29 100	90.00 96	91.06 143	82.87 111	90.64 80	98.06 33	86.06 193	98.14 28	98.20 14	99.58 6	96.96 166

tpm		87.95 166	89.44 145	86.21 195	92.53 135	94.62 188	91.40 178	76.36 211	91.46 139	69.80 202	87.43 103	75.14 155	91.55 131	89.85 204	90.60 197	95.61 204	96.96 166

RPMNet		90.19 132	92.03 118	88.05 163	93.46 124	95.95 136	93.41 128	74.59 219	92.40 128	75.91 148	84.22 131	86.41 94	92.49 117	94.42 134	93.85 139	98.44 169	96.96 166

test-LLR		91.62 114	93.56 94	89.35 138	93.31 128	96.57 120	92.02 170	87.06 132	92.34 131	75.05 158	90.20 84	88.64 86	90.93 141	96.19 100	94.07 132	97.75 186	96.90 169

TESTMET0.1,1		91.07 121	93.56 94	88.17 157	90.43 152	96.57 120	92.02 170	82.83 177	92.34 131	75.05 158	90.20 84	88.64 86	90.93 141	96.19 100	94.07 132	97.75 186	96.90 169

pmmvs4		90.55 127	89.91 140	91.30 111	90.26 157	94.95 176	92.73 138	87.94 122	93.44 107	85.35 101	82.28 144	76.09 151	93.02 113	93.56 148	92.26 187	98.51 165	96.77 171

TSAR-MVS + COLMAP		94.79 61	94.51 72	95.11 55	96.50 63	97.54 94	97.99 39	94.54 38	97.81 13	85.88 99	96.73 26	81.28 126	96.99 54	96.29 95	95.21 106	98.76 147	96.73 172

PatchMatch-RL		94.69 64	94.41 74	95.02 57	97.63 52	98.15 88	94.50 116	91.99 76	95.32 77	91.31 48	95.47 36	83.44 113	96.02 69	96.56 81	95.23 105	98.69 154	96.67 173

PM-MVS		84.72 203	84.47 208	85.03 200	84.67 214	91.57 212	86.27 207	82.31 183	87.65 190	70.62 195	76.54 163	56.41 227	88.75 179	92.59 162	89.85 201	97.54 188	96.66 174

test0.0.03 1		91.97 108	93.91 83	89.72 131	93.31 128	96.40 126	91.34 180	87.06 132	93.86 99	81.67 120	91.15 75	89.16 82	86.02 194	95.08 123	95.09 107	98.91 116	96.64 175

testgi		89.42 140	91.50 126	87.00 187	92.40 137	95.59 149	89.15 198	85.27 156	92.78 116	72.42 184	91.75 68	76.00 152	84.09 204	94.38 135	93.82 141	98.65 157	96.15 176

CostFormer		90.69 124	90.48 138	90.93 114	94.18 114	96.08 131	94.03 121	78.20 202	93.47 106	89.96 64	90.97 78	80.30 128	93.72 99	87.66 212	88.75 204	95.51 206	96.12 177

EU-MVSNet		85.62 199	87.65 174	83.24 206	88.54 201	92.77 200	87.12 203	85.32 153	86.71 196	64.54 212	78.52 157	75.11 156	78.35 211	92.25 167	92.28 186	95.58 205	95.93 178

tpmp4_e23		89.82 136	89.31 147	90.42 122	94.01 118	95.45 154	94.63 114	78.37 199	93.59 104	87.09 97	86.62 114	76.59 147	93.06 112	88.50 206	88.52 205	95.36 207	95.88 179

DWT-MVSNet_training		91.30 119	89.73 141	93.13 95	94.64 109	96.87 108	94.93 106	86.17 141	94.22 95	93.18 30	89.11 92	73.28 172	93.59 102	88.00 209	90.73 196	96.26 197	95.87 180

TransMVSNet (Re)		87.73 175	86.79 191	88.83 144	90.76 149	94.40 191	91.33 181	89.62 103	84.73 208	75.41 152	72.73 197	71.41 193	86.80 189	94.53 131	93.93 136	99.06 106	95.83 181

EPNet_dtu		92.45 106	95.02 65	89.46 135	98.02 46	95.47 153	94.79 111	92.62 58	94.97 85	70.11 199	94.76 44	92.61 66	84.07 205	95.94 105	95.56 98	97.15 191	95.82 182
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
MSDG		94.82 59	93.73 89	96.09 42	98.34 41	97.43 98	97.06 51	96.05 31	95.84 67	90.56 56	86.30 121	89.10 83	95.55 74	96.13 102	95.61 97	99.00 109	95.73 183

pmmvs-eth3d		84.33 205	82.94 212	85.96 199	84.16 217	90.94 213	86.55 206	83.79 168	84.25 210	75.85 149	70.64 210	56.43 226	87.44 187	92.20 171	90.41 199	97.97 182	95.68 184

GG-mvs-BLEND		66.17 225	94.91 67	32.63 233	1.32 238	96.64 118	91.40 178	0.85 237	94.39 93	2.20 241	90.15 86	95.70 52	2.27 237	96.39 89	95.44 101	97.78 184	95.68 184

gm-plane-assit		83.26 208	85.29 203	80.89 209	89.52 171	89.89 216	70.26 225	78.24 201	77.11 222	58.01 224	74.16 184	66.90 212	90.63 156	97.20 52	96.05 76	98.66 156	95.68 184

TAMVS		90.54 128	90.87 135	90.16 126	91.48 142	96.61 119	93.26 131	86.08 142	87.71 189	81.66 121	83.11 140	84.04 109	90.42 161	94.54 130	94.60 118	98.04 181	95.48 187

EG-PatchMatch MVS		86.68 190	87.24 181	86.02 198	90.58 151	96.26 128	91.08 184	81.59 190	84.96 207	69.80 202	71.35 207	75.08 157	84.23 203	94.24 139	93.35 147	98.82 121	95.46 188

COLMAP_ROB	90.49 14	93.27 97	92.71 101	93.93 80	97.75 50	97.44 97	96.07 82	93.17 53	95.40 75	83.86 107	83.76 135	88.72 85	93.87 95	94.25 138	94.11 131	98.87 118	95.28 189
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
Anonymous20231211		75.89 217	74.18 222	77.88 218	81.42 220	87.72 219	79.33 221	81.05 193	66.49 231	60.00 222	45.74 230	51.46 230	71.22 221	85.70 217	86.91 217	94.25 220	95.25 190

ambc			73.83 223		76.23 229	85.13 224	82.27 215		84.16 211	65.58 211	52.82 228	23.31 239	73.55 219	91.41 191	85.26 223	92.97 224	94.70 191

MS-PatchMatch		91.82 110	92.51 105	91.02 112	95.83 73	96.88 106	95.05 103	84.55 166	93.85 100	82.01 114	82.51 143	91.71 67	90.52 158	95.07 124	93.03 152	98.13 177	94.52 192

TDRefinement		89.07 148	88.15 156	90.14 128	95.16 96	96.88 106	95.55 96	90.20 94	89.68 158	76.42 144	76.67 161	74.30 160	84.85 199	93.11 155	91.91 189	98.64 158	94.47 193

MDTV_nov1_ep13		91.57 115	93.18 99	89.70 132	93.39 126	96.97 104	93.53 126	80.91 194	95.70 71	81.86 118	92.40 61	89.93 77	93.25 109	91.97 188	90.80 195	95.25 211	94.46 194

dps		90.11 134	89.37 146	90.98 113	93.89 120	96.21 129	93.49 127	77.61 204	91.95 136	92.74 41	88.85 93	78.77 134	92.37 119	87.71 211	87.71 210	95.80 201	94.38 195

Anonymous20231206		83.84 206	85.19 204	82.26 207	87.38 210	92.87 198	85.49 208	83.65 169	86.07 205	63.44 215	68.42 213	69.01 206	75.45 215	93.34 151	92.44 181	98.12 179	94.20 196

USDC		90.69 124	90.52 137	90.88 115	94.17 115	96.43 124	95.82 89	86.76 134	93.92 98	76.27 146	86.49 116	74.30 160	93.67 101	95.04 125	93.36 146	98.61 159	94.13 197

tpm cat1		88.90 150	87.78 169	90.22 125	93.88 121	95.39 163	93.79 124	78.11 203	92.55 124	89.43 72	81.31 147	79.84 130	91.40 132	84.95 218	86.34 220	94.68 218	94.09 198

RPSCF		94.05 73	94.00 82	94.12 77	96.20 68	96.41 125	96.61 69	91.54 83	95.83 68	89.73 67	96.94 25	92.80 65	95.35 78	91.63 190	90.44 198	95.27 210	93.94 199

tpmrst		88.86 152	89.62 142	87.97 167	94.33 112	95.98 133	92.62 140	76.36 211	94.62 89	76.94 140	85.98 122	82.80 120	92.80 115	86.90 213	87.15 214	94.77 215	93.93 200

MDTV_nov1_ep13_2view		86.30 195	88.27 154	84.01 202	87.71 207	94.67 185	88.08 200	76.78 207	90.59 152	68.66 206	80.46 152	80.12 129	87.58 186	89.95 203	88.20 207	95.25 211	93.90 201

ADS-MVSNet		89.80 137	91.33 129	88.00 166	94.43 111	96.71 116	92.29 154	74.95 218	96.07 60	77.39 137	88.67 96	86.09 96	93.26 108	88.44 207	89.57 202	95.68 203	93.81 202

PatchmatchNet		90.56 126	92.49 107	88.31 154	93.83 122	96.86 110	92.42 144	76.50 210	95.96 62	78.31 134	91.96 67	89.66 79	93.48 104	90.04 201	89.20 203	95.32 208	93.73 203
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
MIMVSNet		88.99 149	91.07 131	86.57 190	86.78 212	95.62 146	91.20 183	75.40 216	90.65 150	76.57 142	84.05 132	82.44 122	91.01 140	95.84 108	95.38 102	98.48 167	93.50 204

EPMVS		90.88 123	92.12 116	89.44 136	94.71 107	97.24 100	93.55 125	76.81 206	95.89 64	81.77 119	91.49 71	86.47 93	93.87 95	90.21 199	90.07 200	95.92 200	93.49 205

testpf		83.57 207	85.70 200	81.08 208	90.99 148	88.96 218	82.71 214	65.32 232	90.22 156	73.86 170	81.58 146	76.10 150	81.19 209	84.14 222	85.41 222	92.43 225	93.45 206

TinyColmap		89.42 140	88.58 151	90.40 123	93.80 123	95.45 154	93.96 123	86.54 136	92.24 133	76.49 143	80.83 149	70.44 198	93.37 105	94.45 133	93.30 149	98.26 175	93.37 207

MDA-MVSNet-bldmvs		80.11 213	80.24 215	79.94 211	77.01 228	93.21 197	78.86 222	85.94 145	82.71 216	60.86 216	79.71 154	51.77 229	83.71 207	75.60 227	86.37 219	93.28 223	92.35 208

N_pmnet		84.80 201	85.10 205	84.45 201	89.25 190	92.86 199	84.04 210	86.21 138	88.78 172	66.73 209	72.41 200	74.87 159	85.21 197	88.32 208	86.45 218	95.30 209	92.04 209

test20.03		82.92 209	85.52 202	79.90 212	87.75 206	91.84 211	82.80 213	82.99 175	82.65 217	60.32 219	78.90 156	70.50 196	67.10 223	92.05 187	90.89 194	98.44 169	91.80 210

pmmvs3		79.16 215	80.12 216	78.05 216	79.36 223	86.59 223	78.13 223	73.87 220	76.42 223	57.51 225	70.59 211	57.02 224	84.66 201	90.10 200	88.32 206	94.75 216	91.77 211

testus		81.33 211	84.13 209	78.06 215	84.54 215	87.72 219	79.66 218	80.42 195	87.36 194	54.13 230	83.83 134	56.63 225	73.21 220	90.51 197	91.74 190	96.40 195	91.11 212

LP		84.43 204	85.10 205	83.66 203	92.31 138	93.89 195	87.13 202	72.88 221	90.81 147	67.08 208	70.65 209	75.76 153	86.87 188	86.43 216	87.15 214	95.70 202	90.98 213

FMVSNet5		90.36 129	90.93 133	89.70 132	87.99 204	92.25 201	92.03 169	83.51 170	92.20 134	84.13 105	85.59 124	86.48 92	92.43 118	94.61 128	94.52 124	98.13 177	90.85 214

new-patchmatchnet		78.49 216	78.19 217	78.84 214	84.13 218	90.06 215	77.11 224	80.39 196	79.57 220	59.64 223	66.01 217	55.65 228	75.62 214	84.55 221	80.70 224	96.14 199	90.77 215

test2356		81.26 212	84.10 210	77.95 217	84.35 216	87.38 221	79.56 219	79.53 198	86.17 202	54.14 229	83.24 137	60.71 219	73.77 216	90.01 202	91.18 193	96.33 196	90.01 216

MIMVSNet1		80.03 214	80.93 214	78.97 213	72.46 231	90.73 214	80.81 217	82.44 182	80.39 218	63.64 214	57.57 224	64.93 217	76.37 213	91.66 189	91.55 192	98.07 180	89.70 217

MVS-HIRNet		85.36 200	86.89 190	83.57 204	90.13 158	94.51 189	83.57 212	72.61 222	88.27 185	71.22 191	68.97 212	81.81 123	88.91 178	93.08 156	91.94 188	94.97 214	89.64 218

CMPMVS	65.18 17	84.76 202	83.10 211	86.69 189	95.29 90	95.05 173	88.37 199	85.51 150	80.27 219	71.31 190	68.37 214	73.85 162	85.25 196	87.72 210	87.75 209	94.38 219	88.70 219
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
new_pmnet		81.53 210	82.68 213	80.20 210	83.47 219	89.47 217	82.21 216	78.36 200	87.86 187	60.14 221	67.90 215	69.43 204	82.03 208	89.22 205	87.47 211	94.99 213	87.39 220

testmv		72.66 220	74.40 219	70.62 221	80.64 221	81.51 228	64.99 230	76.60 208	68.76 227	44.81 232	63.78 220	48.00 231	62.52 225	84.74 219	87.17 212	94.19 221	86.86 221

test1235678		72.65 221	74.40 219	70.62 221	80.64 221	81.50 229	64.99 230	76.59 209	68.74 228	44.81 232	63.78 220	47.99 232	62.51 226	84.73 220	87.17 212	94.19 221	86.85 222

DeepMVS_CX						86.86 222	79.50 220	70.43 226	90.73 148	63.66 213	80.36 153	60.83 218	79.68 210	76.23 226		89.46 228	86.53 223

test12356		69.55 222	71.53 224	67.24 225	77.70 227	78.48 230	65.92 227	75.55 215	68.39 229	44.26 234	61.80 223	40.70 234	47.92 233	81.45 225	87.01 216	92.09 226	82.89 224

PMMVS2		64.36 226	65.94 227	62.52 227	67.37 233	77.44 231	64.39 232	69.32 229	61.47 232	34.59 236	46.09 229	41.03 233	48.02 232	74.56 229	78.23 225	91.43 227	82.76 225

FPMVS		75.84 218	74.59 218	77.29 219	86.92 211	83.89 225	85.01 209	80.05 197	82.91 215	60.61 218	65.25 218	60.41 220	63.86 224	75.60 227	73.60 229	87.29 230	80.47 226

Gipuma		68.35 223	66.71 225	70.27 223	74.16 230	68.78 234	63.93 233	71.77 225	83.34 214	54.57 228	34.37 231	31.88 235	68.69 222	83.30 223	85.53 221	88.48 229	79.78 227
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
1111		73.35 219	74.40 219	72.12 220	78.22 224	82.24 226	65.06 228	65.61 230	70.28 225	55.42 226	56.30 225	57.35 222	73.66 217	86.73 214	88.16 208	94.75 216	79.76 228

PMVS	63.12 18	67.27 224	66.39 226	68.30 224	77.98 226	60.24 235	59.53 234	76.82 205	66.65 230	60.74 217	54.39 227	59.82 221	51.24 229	73.92 230	70.52 230	83.48 232	79.17 229
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
no-one		55.96 228	55.63 229	56.35 229	68.48 232	73.29 233	43.03 235	72.52 223	44.01 235	34.80 235	32.83 232	29.11 236	35.21 234	56.63 232	75.72 227	84.04 231	77.79 230

MVE	50.86 19	49.54 231	51.43 230	47.33 232	44.14 236	59.20 236	36.45 238	60.59 233	41.47 236	31.14 237	29.58 233	17.06 240	48.52 231	62.22 231	74.63 228	63.12 236	75.87 231
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
E-PMN		50.67 229	47.85 231	53.96 230	64.13 235	50.98 238	38.06 236	69.51 227	51.40 234	24.60 238	29.46 235	24.39 238	56.07 228	48.17 233	59.70 231	71.40 234	70.84 232

EMVS		49.98 230	46.76 232	53.74 231	64.96 234	51.29 237	37.81 237	69.35 228	51.83 233	22.69 239	29.57 234	25.06 237	57.28 227	44.81 234	56.11 232	70.32 235	68.64 233

.test1245		56.65 227	56.09 228	57.30 228	78.22 224	82.24 226	65.06 228	65.61 230	70.28 225	55.42 226	56.30 225	57.35 222	73.66 217	86.73 214	15.01 233	5.84 237	24.75 234

testmvs		12.09 232	16.94 233	6.42 234	3.15 237	6.08 239	9.51 240	3.84 235	21.46 237	5.31 240	27.49 236	6.76 241	10.89 235	17.06 235	15.01 233	5.84 237	24.75 234

test123		9.58 233	13.53 234	4.97 235	1.31 239	5.47 240	8.32 241	2.95 236	18.14 238	2.03 242	20.82 237	2.34 242	10.60 236	10.00 236	14.16 235	4.60 239	23.77 236

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										96.83 5		99.12 14

MTMP										97.18 3		98.83 20

Patchmatch-RL test							34.61 239

tmp_tt				66.88 226	86.07 213	73.86 232	68.22 226	33.38 234	96.88 40	80.67 126	88.23 99	78.82 133	49.78 230	82.68 224	77.47 226	83.19 233

XVS					96.60 61	99.35 9	96.82 58			90.85 50		98.72 23				99.46 26

X-MVStestdata					96.60 61	99.35 9	96.82 58			90.85 50		98.72 23				99.46 26

mPP-MVS					99.21 19							98.29 31

NP-MVS									95.32 77

Patchmtry						95.96 135	93.36 129	75.99 213		75.19 155

High-resolution multi-view benchmark