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.

APDe-MVS		77.58 4	82.93 4	71.35 3	77.86 2	80.55 3	83.38 1	57.61 8	85.57 3	61.11 18	86.10 5	82.98 5	64.76 2	78.29 13	76.78 21	83.40 6	90.20 4

MSP-MVS		77.82 3	83.46 3	71.24 5	75.26 13	80.22 4	82.95 2	57.85 6	85.90 2	64.79 4	88.54 3	83.43 4	66.24 1	78.21 16	78.56 6	80.34 41	89.39 6

DPE-MVS		78.11 2	83.84 2	71.42 2	77.82 3	81.32 2	82.92 3	57.81 7	84.04 6	63.19 10	88.63 2	86.00 2	64.52 3	78.71 9	77.63 14	82.26 20	90.57 3

SMA-MVS		77.32 5	82.51 5	71.26 4	75.43 11	80.19 5	82.22 4	58.26 3	84.83 5	64.36 5	78.19 12	83.46 3	63.61 6	81.00 1	80.28 1	83.66 4	89.62 5

SteuartSystems-ACMMP		75.23 10	79.60 12	70.13 10	76.81 4	78.92 9	81.74 5	57.99 4	75.30 27	59.83 23	75.69 15	78.45 21	60.48 26	80.58 2	79.77 2	83.94 3	88.52 9
Skip Steuart: Steuart Systems R&D Blog.
CNVR-MVS		75.62 9	79.91 11	70.61 7	75.76 7	78.82 11	81.66 6	57.12 10	79.77 14	63.04 11	70.69 21	81.15 12	62.99 9	80.23 3	79.54 3	83.11 7	89.16 7

DVP-MVS		78.77 1	84.89 1	71.62 1	78.04 1	82.05 1	81.64 7	57.96 5	87.53 1	66.64 1	88.77 1	86.31 1	63.16 8	79.99 5	78.56 6	82.31 19	91.03 1

DPM-MVS		72.80 24	75.90 27	69.19 18	75.51 10	77.68 17	81.62 8	54.83 24	75.96 23	62.06 16	63.96 40	76.58 27	58.55 35	76.66 30	76.77 22	82.60 16	83.68 40

ACMMP_NAP		76.15 6	81.17 6	70.30 8	74.09 17	79.47 7	81.59 9	57.09 11	81.38 8	63.89 8	79.02 10	80.48 15	62.24 15	80.05 4	79.12 4	82.94 10	88.64 8

APD-MVS		75.80 8	80.90 8	69.86 13	75.42 12	78.48 13	81.43 10	57.44 9	80.45 12	59.32 24	85.28 6	80.82 14	63.96 5	76.89 26	76.08 26	81.58 35	88.30 11
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
TSAR-MVS + MP.		75.22 11	80.06 10	69.56 14	74.61 15	72.74 47	80.59 11	55.70 21	80.80 10	62.65 13	86.25 4	82.92 6	62.07 17	76.89 26	75.66 29	81.77 31	85.19 31
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
NCCC		74.27 16	77.83 22	70.13 10	75.70 8	77.41 20	80.51 12	57.09 11	78.25 18	62.28 15	65.54 34	78.26 22	62.18 16	79.13 6	78.51 8	83.01 9	87.68 15

HFP-MVS		74.87 12	78.86 17	70.21 9	73.99 18	77.91 15	80.36 13	56.63 13	78.41 17	64.27 6	74.54 17	77.75 25	62.96 10	78.70 10	77.82 11	83.02 8	86.91 19

HPM-MVS++		76.01 7	80.47 9	70.81 6	76.60 5	74.96 33	80.18 14	58.36 2	81.96 7	63.50 9	78.80 11	82.53 8	64.40 4	78.74 8	78.84 5	81.81 29	87.46 16

train_agg		73.89 19	78.25 19	68.80 21	75.25 14	72.27 49	79.75 15	56.05 18	74.87 30	58.97 25	81.83 8	79.76 18	61.05 23	77.39 24	76.01 27	81.71 32	85.61 28

zzz-MVS		74.25 17	77.97 21	69.91 12	73.43 21	74.06 41	79.69 16	56.44 15	80.74 11	64.98 3	68.72 27	79.98 17	62.92 11	78.24 15	77.77 13	81.99 27	86.30 21

OPM-MVS		69.33 35	71.05 43	67.32 25	72.34 26	75.70 30	79.57 17	56.34 17	55.21 67	53.81 51	59.51 59	68.96 51	59.67 30	77.61 22	76.44 24	82.19 24	83.88 38

ACMMPR		73.79 21	78.41 18	68.40 22	72.35 25	77.79 16	79.32 18	56.38 16	77.67 21	58.30 29	74.16 18	76.66 26	61.40 20	78.32 12	77.80 12	82.68 14	86.51 20

PGM-MVS		72.89 23	77.13 24	67.94 23	72.47 24	77.25 21	79.27 19	54.63 27	73.71 32	57.95 31	72.38 19	75.33 32	60.75 24	78.25 14	77.36 17	82.57 17	85.62 27

MP-MVS		74.31 15	78.87 15	68.99 19	73.49 20	78.56 12	79.25 20	56.51 14	75.33 25	60.69 20	75.30 16	79.12 20	61.81 18	77.78 20	77.93 10	82.18 25	88.06 13
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
CSCG		74.68 13	79.22 13	69.40 15	75.69 9	80.01 6	79.12 21	52.83 39	79.34 15	63.99 7	70.49 22	82.02 9	60.35 28	77.48 23	77.22 18	84.38 1	87.97 14

SD-MVS		74.43 14	78.87 15	69.26 17	74.39 16	73.70 43	79.06 22	55.24 23	81.04 9	62.71 12	80.18 9	82.61 7	61.70 19	75.43 38	73.92 41	82.44 18	85.22 30
Zhenlong Yuan, Jiakai Cao, Zhaoxin Li, Hao Jiang and Zhaoqi Wang: SD-MVS: Segmentation-driven Deformation Multi-View Stereo with Spherical Refinement and EM optimization. AAAI2024
MCST-MVS		73.67 22	77.39 23	69.33 16	76.26 6	78.19 14	78.77 23	54.54 28	75.33 25	59.99 22	67.96 29	79.23 19	62.43 14	78.00 17	75.71 28	84.02 2	87.30 17

TSAR-MVS + ACMM		72.56 26	79.07 14	64.96 38	73.24 22	73.16 46	78.50 24	48.80 62	79.34 15	55.32 38	85.04 7	81.49 11	58.57 34	75.06 41	73.75 42	75.35 98	85.61 28

ACMMP		71.57 28	75.84 28	66.59 28	70.30 37	76.85 26	78.46 25	53.95 32	73.52 33	55.56 36	70.13 23	71.36 45	58.55 35	77.00 25	76.23 25	82.71 13	85.81 26
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
DeepC-MVS	66.32 2	73.85 20	78.10 20	68.90 20	67.92 47	79.31 8	78.16 26	59.28 1	78.24 19	61.13 17	67.36 33	76.10 30	63.40 7	79.11 7	78.41 9	83.52 5	88.16 12
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
CDPH-MVS		71.47 29	75.82 29	66.41 29	72.97 23	77.15 22	78.14 27	54.71 25	69.88 45	53.07 53	70.98 20	74.83 34	56.95 48	76.22 31	76.57 23	82.62 15	85.09 32

CP-MVS		72.63 25	76.95 25	67.59 24	70.67 33	75.53 31	77.95 28	56.01 19	75.65 24	58.82 26	69.16 26	76.48 28	60.46 27	77.66 21	77.20 19	81.65 33	86.97 18

HQP-MVS		70.88 31	75.02 31	66.05 32	71.69 28	74.47 38	77.51 29	53.17 36	72.89 34	54.88 42	70.03 24	70.48 47	57.26 44	76.02 33	75.01 33	81.78 30	86.21 22

X-MVS		71.18 30	75.66 30	65.96 33	71.71 27	76.96 23	77.26 30	55.88 20	72.75 35	54.48 46	64.39 38	74.47 35	54.19 62	77.84 19	77.37 16	82.21 23	85.85 25

LGP-MVS_train		68.87 37	72.03 39	65.18 37	69.33 41	74.03 42	76.67 31	53.88 33	68.46 46	52.05 56	63.21 42	63.89 63	56.31 52	75.99 34	74.43 37	82.83 12	84.18 34

DeepC-MVS_fast	65.08 3	72.00 27	76.11 26	67.21 26	68.93 43	77.46 18	76.54 32	54.35 29	74.92 29	58.64 28	65.18 35	74.04 40	62.62 12	77.92 18	77.02 20	82.16 26	86.21 22
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
ACMM	60.30 7	67.58 45	68.82 54	66.13 31	70.59 34	72.01 51	76.54 32	54.26 30	65.64 51	54.78 44	50.35 99	61.72 73	58.74 33	75.79 36	75.03 31	81.88 28	81.17 50
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
CPTT-MVS		68.76 39	73.01 35	63.81 45	65.42 58	73.66 44	76.39 34	52.08 41	72.61 36	50.33 59	60.73 54	72.65 43	59.43 32	73.32 49	72.12 47	79.19 54	85.99 24

ACMP	61.42 5	68.72 40	71.37 41	65.64 35	69.06 42	74.45 39	75.88 35	53.30 35	68.10 47	55.74 35	61.53 53	62.29 69	56.97 47	74.70 42	74.23 39	82.88 11	84.31 33
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
DeepPCF-MVS	66.49 1	74.25 17	80.97 7	66.41 29	67.75 49	78.87 10	75.61 36	54.16 31	84.86 4	58.22 30	77.94 13	81.01 13	62.52 13	78.34 11	77.38 15	80.16 45	88.40 10

AdaColmap		67.89 43	68.85 53	66.77 27	73.73 19	74.30 40	75.28 37	53.58 34	70.24 43	57.59 32	51.19 96	59.19 85	60.74 25	75.33 40	73.72 43	79.69 50	77.96 64

MSLP-MVS++		68.17 41	70.72 46	65.19 36	69.41 40	70.64 53	74.99 38	45.76 72	70.20 44	60.17 21	56.42 70	73.01 41	61.14 21	72.80 51	70.54 54	79.70 48	81.42 49

MVS_0304		69.49 34	73.96 33	64.28 43	67.92 47	76.13 29	74.90 39	47.60 64	63.29 55	54.09 50	67.44 32	76.35 29	59.53 31	75.81 35	75.03 31	81.62 34	83.70 39

PCF-MVS	59.98 8	67.32 46	71.04 44	62.97 47	64.77 60	74.49 37	74.78 40	49.54 56	67.44 48	54.39 49	58.35 63	72.81 42	55.79 58	71.54 56	69.24 63	78.57 56	83.41 41
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
MAR-MVS		68.04 42	70.74 45	64.90 39	71.68 29	76.33 28	74.63 41	50.48 53	63.81 53	55.52 37	54.88 78	69.90 49	57.39 42	75.42 39	74.79 35	79.71 47	80.03 53
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
XVS					70.49 35	76.96 23	74.36 42			54.48 46		74.47 35				82.24 21

X-MVStestdata					70.49 35	76.96 23	74.36 42			54.48 46		74.47 35				82.24 21

PHI-MVS		69.27 36	74.84 32	62.76 48	66.83 51	74.83 34	73.88 44	49.32 58	70.61 42	50.93 57	69.62 25	74.84 33	57.25 45	75.53 37	74.32 38	78.35 61	84.17 35

CANet		68.77 38	73.01 35	63.83 44	68.30 44	75.19 32	73.73 45	47.90 63	63.86 52	54.84 43	67.51 31	74.36 38	57.62 40	74.22 44	73.57 45	80.56 40	82.36 44

3Dnovator+	62.63 4	69.51 33	72.62 37	65.88 34	68.21 46	76.47 27	73.50 46	52.74 40	70.85 41	58.65 27	55.97 72	69.95 48	61.11 22	76.80 28	75.09 30	81.09 39	83.23 43

abl_6				64.36 42	70.08 38	77.45 19	72.88 47	50.15 54	71.31 40	54.77 45	62.79 45	77.99 24	56.80 49			81.50 36	83.91 37

TSAR-MVS + GP.		69.71 32	73.92 34	64.80 40	68.27 45	70.56 54	71.90 48	50.75 49	71.38 39	57.46 33	68.68 28	75.42 31	60.10 29	73.47 48	73.99 40	80.32 42	83.97 36

CLD-MVS		67.02 47	71.57 40	61.71 49	71.01 32	74.81 35	71.62 49	38.91 151	71.86 38	60.70 19	64.97 36	67.88 58	51.88 85	76.77 29	74.98 34	76.11 88	69.75 117
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
3Dnovator	60.86 6	66.99 48	70.32 48	63.11 46	66.63 52	74.52 36	71.56 50	45.76 72	67.37 49	55.00 41	54.31 83	68.19 55	58.49 37	73.97 45	73.63 44	81.22 38	80.23 52

MVS_111021_HR		67.62 44	70.39 47	64.39 41	69.77 39	70.45 55	71.44 51	51.72 45	60.77 61	55.06 40	62.14 50	66.40 60	58.13 38	76.13 32	74.79 35	80.19 44	82.04 47

DELS-MVS		65.87 49	70.30 49	60.71 51	64.05 68	72.68 48	70.90 52	45.43 76	57.49 64	49.05 65	64.43 37	68.66 52	55.11 60	74.31 43	73.02 46	79.70 48	81.51 48
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
OMC-MVS		65.16 52	71.35 42	57.94 67	52.95 140	68.82 59	69.00 53	38.28 158	79.89 13	55.20 39	62.76 46	68.31 54	56.14 55	71.30 58	68.70 69	76.06 90	79.67 54

Effi-MVS+		63.28 57	65.96 63	60.17 53	64.26 64	68.06 64	68.78 54	45.71 74	54.08 71	46.64 71	55.92 73	63.13 67	55.94 56	70.38 67	71.43 49	79.68 51	78.70 59

casdiffmvs		64.09 54	68.13 55	59.37 58	61.81 74	68.32 63	68.48 55	44.45 87	61.95 58	49.12 64	63.04 43	69.67 50	53.83 65	70.46 64	66.06 106	78.55 57	77.43 67

CS-MVS		63.45 56	65.72 65	60.80 50	64.20 65	67.86 66	68.46 56	43.50 109	53.86 72	49.90 61	56.44 69	60.45 80	57.27 43	73.56 47	70.13 59	81.45 37	77.73 66

ETV-MVS		63.10 58	66.18 62	59.51 56	62.37 72	67.42 75	67.80 57	42.86 121	54.77 68	45.90 74	60.15 56	57.01 93	57.83 39	73.78 46	71.30 51	80.27 43	78.82 58

MSDG		58.46 85	58.97 113	57.85 71	66.27 56	66.23 87	67.72 58	42.33 124	53.43 74	43.68 85	43.39 152	45.35 154	49.75 93	68.66 80	67.77 79	77.38 69	67.96 132

EPNet		65.14 53	69.54 51	60.00 54	66.61 53	67.67 70	67.53 59	55.32 22	62.67 57	46.22 73	67.74 30	65.93 61	48.07 104	72.17 53	72.12 47	76.28 84	78.47 61
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
ACMH+	53.71 12	59.26 75	60.28 93	58.06 64	64.17 67	68.46 61	67.51 60	50.93 48	52.46 83	35.83 126	40.83 166	45.12 158	52.32 81	69.88 70	69.00 67	77.59 67	76.21 82

LS3D		60.20 72	61.70 81	58.45 62	64.18 66	67.77 67	67.19 61	48.84 61	61.67 59	41.27 100	45.89 132	51.81 113	54.18 63	68.78 77	66.50 101	75.03 100	69.48 123

v2v482		58.69 81	60.12 100	57.03 75	57.16 114	66.05 88	67.17 62	43.52 107	46.33 128	45.19 79	49.46 103	51.02 116	52.51 79	67.30 107	66.03 107	76.61 78	74.62 92

v1144		58.88 78	60.16 97	57.39 73	58.03 95	67.26 76	67.14 63	44.46 86	45.17 136	44.33 83	47.81 114	49.92 123	53.20 76	67.77 99	66.62 98	77.15 73	76.58 76

diffmvs		61.64 65	66.55 58	55.90 82	56.63 116	63.71 106	67.13 64	41.27 134	59.49 63	46.70 70	63.93 41	68.01 57	50.46 90	67.30 107	65.51 114	73.24 122	77.87 65

v10		59.17 77	60.60 89	57.50 72	57.95 96	66.73 81	67.09 65	44.11 89	46.85 123	45.42 77	48.18 113	51.07 115	53.63 66	67.84 97	66.59 99	76.79 76	76.92 72

CostFormer		56.57 104	59.13 111	53.60 95	57.52 102	61.12 121	66.94 66	35.95 169	53.44 73	44.68 81	55.87 74	54.44 104	48.21 101	60.37 156	58.33 164	68.27 155	70.33 115

ACMH	52.42 13	58.24 90	59.56 107	56.70 79	66.34 55	69.59 56	66.71 67	49.12 59	46.08 131	28.90 153	42.67 161	41.20 176	52.60 78	71.39 57	70.28 56	76.51 80	75.72 85
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
v8		58.88 78	60.57 91	56.92 76	57.35 107	65.69 91	66.69 68	42.64 122	47.89 118	45.77 75	49.04 105	52.98 109	52.77 77	67.51 104	65.57 113	76.26 85	75.30 90

MVS_Test		62.40 63	66.23 61	57.94 67	59.77 88	64.77 100	66.50 69	41.76 129	57.26 65	49.33 62	62.68 47	67.47 59	53.50 70	68.57 82	66.25 103	76.77 77	76.58 76

PVSNet_Blended_VisFu		63.65 55	66.92 56	59.83 55	60.03 85	73.44 45	66.33 70	48.95 60	52.20 84	50.81 58	56.07 71	60.25 81	53.56 67	73.23 50	70.01 60	79.30 52	83.24 42

V42		56.97 100	60.14 98	53.28 98	48.16 169	62.78 111	66.30 71	37.93 160	47.44 120	42.68 91	48.19 112	52.59 111	51.90 84	67.46 105	65.94 109	72.72 127	76.55 78

v1192		58.51 82	59.66 103	57.17 74	57.82 97	67.72 68	66.21 72	44.83 82	44.15 144	43.49 86	46.68 119	47.94 127	53.55 68	67.39 106	66.51 100	77.13 74	77.20 70

DI_MVS_plusplus_trai		61.88 64	65.17 69	58.06 64	60.05 84	65.26 94	66.03 73	44.22 88	55.75 66	46.73 69	54.64 81	68.12 56	54.13 64	69.13 75	66.66 95	77.18 72	76.61 75

Effi-MVS+-dtu		60.34 71	62.32 80	58.03 66	64.31 62	67.44 74	65.99 74	42.26 125	49.55 95	42.00 96	48.92 107	59.79 83	56.27 53	68.07 93	67.03 87	77.35 70	75.45 88

CNLPA		62.78 61	66.31 60	58.65 61	58.47 93	68.41 62	65.98 75	41.22 135	78.02 20	56.04 34	46.65 120	59.50 84	57.50 41	69.67 71	65.27 118	72.70 129	76.67 74

MVS_111021_LR		63.05 59	66.43 59	59.10 59	61.33 78	63.77 105	65.87 76	43.58 105	60.20 62	53.70 52	62.09 51	62.38 68	55.84 57	70.24 68	68.08 74	74.30 105	78.28 63

ET-MVSNet_ETH3D		58.38 87	61.57 82	54.67 89	42.15 189	65.26 94	65.70 77	43.82 97	48.84 105	42.34 93	59.76 58	47.76 130	56.68 50	67.02 114	68.60 72	77.33 71	73.73 100

v144192		58.23 91	59.40 109	56.87 77	57.56 99	66.89 79	65.70 77	45.01 80	44.06 145	42.88 88	46.61 121	48.09 126	53.49 71	66.94 115	65.90 110	76.61 78	77.29 68

QAPM		65.27 51	69.49 52	60.35 52	65.43 57	72.20 50	65.69 79	47.23 65	63.46 54	49.14 63	53.56 84	71.04 46	57.01 46	72.60 52	71.41 50	77.62 65	82.14 46

Fast-Effi-MVS+-dtu		56.30 106	59.29 110	52.82 106	58.64 92	64.89 98	65.56 80	32.89 186	45.80 133	35.04 128	45.89 132	54.14 105	49.41 94	67.16 110	66.45 102	75.37 97	70.69 112

TSAR-MVS + COLMAP		62.65 62	69.90 50	54.19 91	46.31 177	66.73 81	65.49 81	41.36 133	76.57 22	46.31 72	76.80 14	56.68 94	53.27 75	69.50 72	66.65 96	72.40 134	76.36 81

EIA-MVS		61.53 68	63.79 76	58.89 60	63.82 70	67.61 71	65.35 82	42.15 128	49.98 92	45.66 76	57.47 67	56.62 95	56.59 51	70.91 62	69.15 64	79.78 46	74.80 91

TAPA-MVS	54.74 10	60.85 69	66.61 57	54.12 93	47.38 173	65.33 92	65.35 82	36.51 167	75.16 28	48.82 66	54.70 80	63.51 65	53.31 74	68.36 84	64.97 124	73.37 117	74.27 94
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
Fast-Effi-MVS+		60.36 70	63.35 77	56.87 77	58.70 90	65.86 89	65.08 84	37.11 164	53.00 79	45.36 78	52.12 91	56.07 100	56.27 53	71.28 59	69.42 62	78.71 55	75.69 86

MS-PatchMatch		58.19 92	60.20 96	55.85 83	65.17 59	64.16 103	64.82 85	41.48 132	50.95 87	42.17 95	45.38 137	56.42 96	48.08 103	68.30 85	66.70 94	73.39 116	69.46 125

HyFIR lowres test		56.87 102	58.60 117	54.84 87	56.62 117	69.27 58	64.77 86	42.21 126	45.66 134	37.50 121	33.08 183	57.47 92	53.33 73	65.46 134	67.94 75	74.60 102	71.35 107

v1921920		57.89 94	59.02 112	56.58 80	57.55 100	66.66 85	64.72 87	44.70 84	43.55 148	42.73 90	46.17 129	46.93 140	53.51 69	66.78 116	65.75 112	76.29 83	77.28 69

canonicalmvs		65.62 50	72.06 38	58.11 63	63.94 69	71.05 52	64.49 88	43.18 117	74.08 31	47.35 67	64.17 39	71.97 44	51.17 88	71.87 54	70.74 52	78.51 59	80.56 51

PLC	52.09 14	59.21 76	62.47 79	55.41 86	53.24 138	64.84 99	64.47 89	40.41 144	65.92 50	44.53 82	46.19 128	55.69 101	55.33 59	68.24 88	65.30 117	74.50 103	71.09 108
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
OpenMVS	57.13 9	62.81 60	65.75 64	59.39 57	66.47 54	69.52 57	64.26 90	43.07 119	61.34 60	50.19 60	47.29 117	64.41 62	54.60 61	70.18 69	68.62 71	77.73 63	78.89 57

CANet_DTU		58.88 78	64.68 72	52.12 110	55.77 120	66.75 80	63.92 91	37.04 165	53.32 75	37.45 122	59.81 57	61.81 72	44.43 121	68.25 86	67.47 85	74.12 107	75.33 89

v1240		57.55 96	58.63 116	56.29 81	57.30 110	66.48 86	63.77 92	44.56 85	42.77 158	42.48 92	45.64 135	46.28 147	53.46 72	66.32 122	65.80 111	76.16 87	77.13 71

tpm cat1		53.30 128	53.41 150	53.17 102	58.16 94	59.15 139	63.73 93	38.27 159	50.73 89	46.98 68	45.57 136	44.00 170	49.20 95	55.90 184	54.02 183	62.65 175	64.50 158

PVSNet_BlendedMVS		61.63 66	64.82 70	57.91 69	57.21 112	67.55 72	63.47 94	46.08 70	54.72 69	52.46 54	58.59 61	60.73 76	51.82 86	70.46 64	65.20 120	76.44 81	76.50 79

PVSNet_Blended		61.63 66	64.82 70	57.91 69	57.21 112	67.55 72	63.47 94	46.08 70	54.72 69	52.46 54	58.59 61	60.73 76	51.82 86	70.46 64	65.20 120	76.44 81	76.50 79

CHOSEN 1792x2688		55.85 109	58.01 121	53.33 97	57.26 111	62.82 110	63.29 96	41.55 131	46.65 125	38.34 115	34.55 181	53.50 106	52.43 80	67.10 112	67.56 84	67.13 159	73.92 99

IB-MVS	54.11 11	58.36 88	60.70 88	55.62 84	58.67 91	68.02 65	61.56 97	43.15 118	46.09 130	44.06 84	44.24 144	50.99 118	48.71 98	66.70 117	70.33 55	77.60 66	78.50 60
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
MVSTER		57.19 97	61.11 85	52.62 107	50.82 160	58.79 142	61.55 98	37.86 161	48.81 107	41.31 99	57.43 68	52.10 112	48.60 99	68.19 90	66.75 93	75.56 94	75.68 87

EG-PatchMatch MVS		56.98 99	58.24 120	55.50 85	64.66 61	68.62 60	61.48 99	43.63 104	38.44 182	41.44 97	38.05 172	46.18 149	43.95 122	71.71 55	70.61 53	77.87 62	74.08 97

DCV-MVSNet		59.49 73	64.00 75	54.23 90	61.81 74	64.33 102	61.42 100	43.77 98	52.85 80	38.94 114	55.62 75	62.15 71	43.24 129	69.39 73	67.66 82	76.22 86	75.97 83

v148		55.58 112	57.61 127	53.20 100	54.59 130	61.86 113	61.18 101	38.70 156	44.30 143	42.25 94	47.53 115	50.24 122	48.73 97	65.15 136	62.61 146	73.79 110	71.61 106

GA-MVS		55.67 110	58.33 118	52.58 108	55.23 125	63.09 107	61.08 102	40.15 146	42.95 153	37.02 124	52.61 88	47.68 131	47.51 106	65.92 128	65.35 115	74.49 104	70.68 113

v7n		55.67 110	57.46 128	53.59 96	56.06 118	65.29 93	61.06 103	43.26 116	40.17 173	37.99 118	40.79 167	45.27 157	47.09 108	67.67 101	66.21 104	76.08 89	76.82 73

Anonymous202405211			60.60 89		63.44 71	66.71 84	61.00 104	47.23 65	50.62 90		36.85 175	60.63 79	43.03 130	69.17 74	67.72 81	75.41 95	72.54 102

Anonymous20231211		57.71 95	60.79 87	54.13 92	61.68 76	65.81 90	60.81 105	43.70 102	51.97 85	39.67 109	34.82 180	63.59 64	43.31 127	68.55 83	66.63 97	75.59 93	74.13 96

DWT-MVSNet_training		53.80 125	54.31 145	53.21 99	57.65 98	59.04 140	60.65 106	40.11 147	46.35 127	42.77 89	49.07 104	41.07 177	51.06 89	58.62 166	58.96 162	67.00 162	67.06 137

COLMAP_ROB	46.52 15	51.99 138	54.86 141	48.63 134	49.13 167	61.73 115	60.53 107	36.57 166	53.14 76	32.95 133	37.10 173	38.68 188	40.49 139	65.72 130	63.08 139	72.11 138	64.60 157
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
pmmvs4		54.66 121	56.07 132	53.00 103	54.63 127	57.08 157	60.43 108	44.10 90	51.69 86	40.55 104	46.55 124	44.79 163	45.95 114	62.54 145	63.66 134	72.36 135	66.20 144

Vis-MVSNet		58.48 84	65.70 66	50.06 119	53.40 137	67.20 77	60.24 109	43.32 114	48.83 106	30.23 146	62.38 49	61.61 74	40.35 140	71.03 61	69.77 61	72.82 125	79.11 56
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
baseline2		55.89 107	57.82 123	53.64 94	57.36 106	61.09 122	59.75 110	40.45 142	47.38 121	41.26 101	51.23 95	46.90 141	48.11 102	65.63 132	64.38 129	74.90 101	68.16 131

IterMVS-LS		58.30 89	61.39 83	54.71 88	59.92 87	58.40 146	59.42 111	43.64 103	48.71 109	40.25 107	57.53 66	58.55 87	52.15 83	65.42 135	65.34 116	72.85 123	75.77 84
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
thisisatest0530		56.68 103	59.68 102	53.19 101	52.97 139	60.96 124	59.41 112	40.51 140	48.26 115	41.06 102	52.67 87	46.30 146	49.78 91	67.66 102	67.83 77	75.39 96	74.07 98

TDRefinement		49.31 152	52.44 158	45.67 155	30.44 202	59.42 135	59.24 113	39.78 149	48.76 108	31.20 141	35.73 177	29.90 201	42.81 131	64.24 140	62.59 147	70.55 147	66.43 140

tttt0517		56.53 105	59.59 104	52.95 104	52.66 142	60.99 123	59.21 114	40.51 140	47.89 118	40.40 105	52.50 90	46.04 150	49.78 91	67.75 100	67.83 77	75.15 99	74.17 95

IterMVS		53.45 127	57.12 129	49.17 125	49.23 166	60.93 125	59.05 115	34.63 175	44.53 139	33.22 131	51.09 98	51.01 117	48.38 100	62.43 147	60.79 154	70.54 148	69.05 128
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
baseline		55.19 117	60.88 86	48.55 135	49.87 164	58.10 151	58.70 116	34.75 173	52.82 81	39.48 113	60.18 55	60.86 75	45.41 116	61.05 152	60.74 155	63.10 173	72.41 103

EPP-MVSNet		59.39 74	65.45 67	52.32 109	60.96 80	67.70 69	58.42 117	44.75 83	49.71 94	27.23 161	59.03 60	62.20 70	43.34 126	70.71 63	69.13 65	79.25 53	79.63 55

MDTV_nov1_ep13		50.32 148	52.43 159	47.86 145	49.87 164	54.70 161	58.10 118	34.29 177	45.59 135	37.71 119	47.44 116	47.42 135	41.86 134	58.07 170	55.21 176	65.34 167	58.56 180

TranMVSNet+NR-MVSNet		55.87 108	60.14 98	50.88 113	59.46 89	63.82 104	57.93 119	52.98 37	48.94 104	20.52 177	52.87 86	47.33 136	36.81 158	69.12 76	69.03 66	77.56 68	69.89 116

NR-MVSNet		55.35 114	59.46 108	50.56 115	61.33 78	62.97 108	57.91 120	51.80 43	48.62 112	20.59 176	51.99 92	44.73 164	34.10 168	68.58 81	68.64 70	77.66 64	70.67 114

UniMVSNet_NR-MVSNet		56.94 101	61.14 84	52.05 111	60.02 86	65.21 97	57.44 121	52.93 38	49.37 98	24.31 171	54.62 82	50.54 119	39.04 144	68.69 78	68.84 68	78.53 58	70.72 110

DU-MVS		55.41 113	59.59 104	50.54 116	54.60 128	62.97 108	57.44 121	51.80 43	48.62 112	24.31 171	51.99 92	47.00 139	39.04 144	68.11 91	67.75 80	76.03 91	70.72 110

IS_MVSNet		57.95 93	64.26 74	50.60 114	61.62 77	65.25 96	57.18 123	45.42 77	50.79 88	26.49 164	57.81 65	60.05 82	34.51 165	71.24 60	70.20 58	78.36 60	74.44 93

GBi-Net		55.20 115	60.25 94	49.31 122	52.42 143	61.44 116	57.03 124	44.04 92	49.18 101	30.47 142	48.28 109	58.19 88	38.22 147	68.05 94	66.96 88	73.69 112	69.65 118

test1		55.20 115	60.25 94	49.31 122	52.42 143	61.44 116	57.03 124	44.04 92	49.18 101	30.47 142	48.28 109	58.19 88	38.22 147	68.05 94	66.96 88	73.69 112	69.65 118

FMVSNet2		55.04 119	59.95 101	49.31 122	52.42 143	61.44 116	57.03 124	44.08 91	49.55 95	30.40 145	46.89 118	58.84 86	38.22 147	67.07 113	66.21 104	73.69 112	69.65 118

UniMVSNet_ETH3D		52.62 130	55.98 133	48.70 133	51.04 157	60.71 126	56.87 127	46.74 67	42.52 160	26.96 162	42.50 162	45.95 151	37.87 151	66.22 124	65.15 123	72.74 126	68.78 130

FMVSNet1		54.08 123	58.68 115	48.71 132	50.90 159	61.35 119	56.73 128	43.94 96	45.91 132	29.32 152	42.72 160	56.26 99	37.70 152	68.05 94	66.96 88	73.69 112	69.50 122

FMVSNet3		54.78 120	59.58 106	49.17 125	52.37 146	61.31 120	56.72 129	44.04 92	49.18 101	30.47 142	48.28 109	58.19 88	38.09 150	65.48 133	65.20 120	73.31 119	69.45 126

UGNet		57.03 98	65.25 68	47.44 147	46.54 176	66.73 81	56.30 130	43.28 115	50.06 91	32.99 132	62.57 48	63.26 66	33.31 170	68.25 86	67.58 83	72.20 137	78.29 62
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
Baseline_NR-MVSNet		53.50 126	57.89 122	48.37 138	54.60 128	59.25 138	56.10 131	51.84 42	49.32 99	17.92 184	45.38 137	47.68 131	36.93 157	68.11 91	65.95 108	72.84 124	69.57 121

tpm		48.82 158	51.27 166	45.96 152	54.10 133	47.35 185	56.05 132	30.23 190	46.70 124	43.21 87	52.54 89	47.55 134	37.28 155	54.11 189	50.50 192	54.90 192	60.12 176

pmmvs-eth3d		51.33 141	52.25 160	50.26 118	50.82 160	54.65 162	56.03 133	43.45 113	43.51 149	37.20 123	39.20 170	39.04 187	42.28 132	61.85 150	62.78 143	71.78 140	64.72 156

thisisatest0515		53.85 124	56.84 131	50.37 117	50.25 163	58.17 150	55.99 134	39.90 148	41.88 163	38.16 117	45.91 131	45.30 155	44.58 120	66.15 126	66.89 91	73.36 118	73.57 101

FC-MVSNet-train		58.40 86	63.15 78	52.85 105	64.29 63	61.84 114	55.98 135	46.47 68	53.06 77	34.96 129	61.95 52	56.37 98	39.49 142	68.67 79	68.36 73	75.92 92	71.81 105

UniMVSNet (Re)		55.15 118	60.39 92	49.03 128	55.31 122	64.59 101	55.77 136	50.63 50	48.66 111	20.95 175	51.47 94	50.40 120	34.41 167	67.81 98	67.89 76	77.11 75	71.88 104

baseline1		54.48 122	58.69 114	49.57 120	60.63 83	58.29 149	55.70 137	44.95 81	49.20 100	29.62 149	54.77 79	54.75 103	35.29 162	67.15 111	64.08 130	71.21 144	62.58 167

thres200		52.39 133	55.37 138	48.90 129	57.39 105	60.18 129	55.60 138	43.73 100	42.93 154	27.41 159	43.35 153	45.09 159	36.61 159	66.36 120	63.92 133	72.66 130	65.78 149

CDS-MVSNet		52.42 132	57.06 130	47.02 149	53.92 135	58.30 148	55.50 139	46.47 68	42.52 160	29.38 151	49.50 102	52.85 110	28.49 180	66.70 117	66.89 91	68.34 154	62.63 166
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
thres400		52.38 134	55.51 135	48.74 131	57.49 103	60.10 131	55.45 140	43.54 106	42.90 155	26.72 163	43.34 154	45.03 162	36.61 159	66.20 125	64.53 127	72.66 130	66.43 140

tfpn200view9		52.53 131	55.51 135	49.06 127	57.31 108	60.24 128	55.42 141	43.77 98	42.85 156	27.81 157	43.00 158	45.06 160	37.32 154	66.38 119	64.54 126	72.71 128	66.54 139

thres100view900		52.04 137	54.81 142	48.80 130	57.31 108	59.33 136	55.30 142	42.92 120	42.85 156	27.81 157	43.00 158	45.06 160	36.99 156	64.74 138	63.51 135	72.47 133	65.21 153

our_test_3					51.15 155	57.31 156	55.12 143

IterMVS-SCA-FT		52.18 135	57.75 125	45.68 154	51.01 158	62.06 112	55.10 144	34.75 173	44.85 137	32.86 134	51.13 97	51.22 114	48.74 96	62.47 146	61.51 150	51.61 199	71.02 109

MDTV_nov1_ep13_2view		47.62 167	49.72 176	45.18 157	48.05 170	53.70 165	54.90 145	33.80 181	39.90 175	29.79 148	38.85 171	41.89 174	39.17 143	58.99 160	55.55 173	65.34 167	59.17 178

thres600view7		51.91 140	55.14 139	48.14 140	57.43 104	60.18 129	54.60 146	43.73 100	42.61 159	25.20 167	43.10 157	44.47 167	35.19 163	66.36 120	63.28 138	72.66 130	66.01 147

tfpnnormal		50.16 149	52.19 161	47.78 146	56.86 115	58.37 147	54.15 147	44.01 95	38.35 184	25.94 165	36.10 176	37.89 190	34.50 166	65.93 127	63.42 136	71.26 143	65.28 152

TransMVSNet (Re)		51.92 139	55.38 137	47.88 144	60.95 81	59.90 132	53.95 148	45.14 79	39.47 176	24.85 168	43.87 147	46.51 145	29.15 177	67.55 103	65.23 119	73.26 121	65.16 154

dps		50.42 146	51.20 167	49.51 121	55.88 119	56.07 159	53.73 149	38.89 152	43.66 146	40.36 106	45.66 134	37.63 192	45.23 117	59.05 159	56.18 168	62.94 174	60.16 175

anonymousdsp		52.84 129	57.78 124	47.06 148	40.24 192	58.95 141	53.70 150	33.54 183	36.51 189	32.69 135	43.88 146	45.40 153	47.97 105	67.17 109	70.28 56	74.22 106	82.29 45

UA-Net		58.50 83	64.68 72	51.30 112	66.97 50	67.13 78	53.68 151	45.65 75	49.51 97	31.58 140	62.91 44	68.47 53	35.85 161	68.20 89	67.28 86	74.03 108	69.24 127

tpmrst		48.08 163	49.88 175	45.98 151	52.71 141	48.11 183	53.62 152	33.70 182	48.70 110	39.74 108	48.96 106	46.23 148	40.29 141	50.14 197	49.28 194	55.80 189	57.71 182

gg-mvs-nofinetune		49.07 157	52.56 157	45.00 158	61.99 73	59.78 133	53.55 153	41.63 130	31.62 198	12.08 192	29.56 192	53.28 108	29.57 176	66.27 123	64.49 128	71.19 145	62.92 163

PatchmatchNet		49.92 151	51.29 165	48.32 139	51.83 150	51.86 172	53.38 154	37.63 163	47.90 117	40.83 103	48.54 108	45.30 155	45.19 118	56.86 174	53.99 185	61.08 180	54.57 187
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
USDC		51.11 142	53.71 147	48.08 142	44.76 181	55.99 160	53.01 155	40.90 136	52.49 82	36.14 125	44.67 142	33.66 197	43.27 128	63.23 141	61.10 152	70.39 149	64.82 155

EPNet_dtu		52.05 136	58.26 119	44.81 159	54.10 133	50.09 178	52.01 156	40.82 138	53.03 78	27.41 159	54.90 77	57.96 91	26.72 182	62.97 142	62.70 145	67.78 157	66.19 145
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
TinyColmap		47.08 169	47.56 183	46.52 150	42.35 188	53.44 166	51.77 157	40.70 139	43.44 150	31.92 138	29.78 191	23.72 206	45.04 119	61.99 149	59.54 160	67.35 158	61.03 171

pm-mvs1		51.02 143	55.55 134	45.73 153	54.16 132	58.52 144	50.92 158	42.56 123	40.32 172	25.67 166	43.66 149	50.34 121	30.06 175	65.85 129	63.97 132	70.99 146	66.21 143

MIMVSNet		43.79 181	48.53 179	38.27 183	41.46 190	48.97 181	50.81 159	32.88 187	44.55 138	22.07 173	32.05 184	47.15 137	24.76 185	58.73 163	56.09 170	57.63 188	52.14 188

SCA		50.99 144	53.22 154	48.40 137	51.07 156	56.78 158	50.25 160	39.05 150	48.31 114	41.38 98	49.54 101	46.70 144	46.00 113	58.31 167	56.28 167	62.65 175	56.60 184

PatchMatch-RL		50.11 150	51.56 164	48.43 136	46.23 178	51.94 171	50.21 161	38.62 157	46.62 126	37.51 120	42.43 163	39.38 185	52.24 82	60.98 153	59.56 159	65.76 164	60.01 177

test-LLR		49.28 153	50.29 171	48.10 141	55.26 123	47.16 186	49.52 162	43.48 111	39.22 177	31.98 136	43.65 150	47.93 128	41.29 137	56.80 175	55.36 174	67.08 160	61.94 168

TESTMET0.1,1		46.09 175	50.29 171	41.18 175	36.91 195	47.16 186	49.52 162	20.32 204	39.22 177	31.98 136	43.65 150	47.93 128	41.29 137	56.80 175	55.36 174	67.08 160	61.94 168

pmmvs6		48.35 161	51.64 163	44.51 161	51.92 149	57.94 153	49.44 164	42.17 127	34.45 191	24.62 170	28.87 194	46.90 141	29.07 179	64.60 139	63.08 139	69.83 150	65.68 150

PMMVS		49.20 156	54.28 146	43.28 167	34.13 197	45.70 193	48.98 165	26.09 199	46.31 129	34.92 130	55.22 76	53.47 107	47.48 107	59.43 158	59.04 161	68.05 156	60.77 172

GG-mvs-BLEND		36.62 194	53.39 151	17.06 203	0.01 214	58.61 143	48.63 166	0.01 211	47.13 122	0.02 214	43.98 145	60.64 78	0.03 210	54.92 188	51.47 191	53.64 195	56.99 183

CR-MVSNet		50.47 145	52.61 156	47.98 143	49.03 168	52.94 167	48.27 167	38.86 153	44.41 140	39.59 110	44.34 143	44.65 166	46.63 110	58.97 161	60.31 156	65.48 165	62.66 164

Patchmtry						47.61 184	48.27 167	38.86 153		39.59 110

pmmvs5		47.07 170	51.02 169	42.46 169	45.18 180	51.47 173	48.23 169	33.09 185	38.17 185	28.62 155	46.60 122	43.48 171	30.74 173	58.28 168	58.63 163	68.92 152	60.48 173

SixPastTwentyTwo		47.55 168	50.25 173	44.41 162	47.30 174	54.31 164	47.81 170	40.36 145	33.76 192	19.93 179	43.75 148	32.77 199	42.07 133	59.82 157	60.94 153	68.98 151	66.37 142

test-mter		45.30 176	50.37 170	39.38 180	33.65 199	46.99 188	47.59 171	18.59 205	38.75 180	28.00 156	43.28 155	46.82 143	41.50 136	57.28 173	55.78 171	66.93 163	63.70 161

EPMVS		44.66 178	47.86 182	40.92 176	47.97 171	44.70 195	47.58 172	33.27 184	48.11 116	29.58 150	49.65 100	44.38 168	34.65 164	51.71 192	47.90 196	52.49 197	48.57 198

CMPMVS	37.70 17	49.24 154	52.71 155	45.19 156	45.97 179	51.23 174	47.44 173	29.31 191	43.04 152	44.69 80	34.45 182	48.35 125	43.64 123	62.59 144	59.82 158	60.08 181	69.48 123
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
FMVSNet5		40.96 186	45.81 187	35.29 191	34.30 196	44.55 196	47.28 174	28.84 193	40.76 169	21.62 174	29.85 190	42.44 172	24.77 184	57.53 172	55.00 177	54.93 191	50.56 193

PEN-MVS		49.21 155	54.32 144	43.24 168	54.33 131	59.26 137	47.04 175	51.37 47	41.67 164	9.97 197	46.22 127	41.80 175	22.97 189	60.52 154	64.03 131	73.73 111	66.75 138

CP-MVSNet		48.37 160	53.53 149	42.34 170	51.35 153	58.01 152	46.56 176	50.54 51	41.62 165	10.61 194	46.53 125	40.68 181	23.18 187	58.71 164	61.83 148	71.81 139	67.36 136

PS-CasMVS		48.18 162	53.25 153	42.27 171	51.26 154	57.94 153	46.51 177	50.52 52	41.30 166	10.56 195	45.35 139	40.34 183	23.04 188	58.66 165	61.79 149	71.74 141	67.38 135

CVMVSNet		46.38 174	52.01 162	39.81 179	42.40 187	50.26 176	46.15 178	37.68 162	40.03 174	15.09 187	46.56 123	47.56 133	33.72 169	56.50 179	55.65 172	63.80 171	67.53 133

PM-MVS		44.55 179	48.13 181	40.37 178	32.85 201	46.82 190	46.11 179	29.28 192	40.48 171	29.99 147	39.98 169	34.39 196	41.80 135	56.08 182	53.88 187	62.19 178	65.31 151

RPMNet		46.41 172	48.72 178	43.72 163	47.77 172	52.94 167	46.02 180	33.92 179	44.41 140	31.82 139	36.89 174	37.42 193	37.41 153	53.88 190	54.02 183	65.37 166	61.47 170

Vis-MVSNet (Re-imp)		50.37 147	57.73 126	41.80 173	57.53 101	54.35 163	45.70 181	45.24 78	49.80 93	13.43 190	58.23 64	56.42 96	20.11 192	62.96 143	63.36 137	68.76 153	58.96 179

FPMVS		38.36 193	40.41 197	35.97 188	38.92 194	39.85 200	45.50 182	25.79 200	41.13 167	18.70 181	30.10 189	24.56 204	31.86 172	49.42 199	46.80 199	55.04 190	51.03 191

RPSCF		46.41 172	54.42 143	37.06 187	25.70 209	45.14 194	45.39 183	20.81 203	62.79 56	35.10 127	44.92 141	55.60 102	43.56 124	56.12 181	52.45 189	51.80 198	63.91 160

TAMVS		44.02 180	49.18 177	37.99 185	47.03 175	45.97 192	45.04 184	28.47 194	39.11 179	20.23 178	43.22 156	48.52 124	28.49 180	58.15 169	57.95 166	58.71 183	51.36 190

CHOSEN 280x420		40.80 187	45.05 190	35.84 190	32.95 200	29.57 205	44.98 185	23.71 202	37.54 187	18.42 182	31.36 187	47.07 138	46.41 112	56.71 177	54.65 181	48.55 201	58.47 181

MDA-MVSNet-bldmvs		41.36 185	43.15 194	39.27 181	28.74 204	52.68 169	44.95 186	40.84 137	32.89 194	18.13 183	31.61 186	22.09 207	38.97 146	50.45 196	56.11 169	64.01 170	56.23 185

WR-MVS_H		47.65 166	53.67 148	40.63 177	51.45 151	59.74 134	44.71 187	49.37 57	40.69 170	7.61 204	46.04 130	44.34 169	17.32 194	57.79 171	61.18 151	73.30 120	65.86 148

DTE-MVSNet		48.03 165	53.28 152	41.91 172	54.64 126	57.50 155	44.63 188	51.66 46	41.02 168	7.97 203	46.26 126	40.90 178	20.24 191	60.45 155	62.89 142	72.33 136	63.97 159

WR-MVS		48.78 159	55.06 140	41.45 174	55.50 121	60.40 127	43.77 189	49.99 55	41.92 162	8.10 202	45.24 140	45.56 152	17.47 193	61.57 151	64.60 125	73.85 109	66.14 146

Anonymous20231206		42.28 183	45.89 186	38.07 184	51.96 148	48.98 180	43.66 190	38.81 155	38.74 181	14.32 189	26.74 196	40.90 178	20.94 190	56.64 178	54.67 180	58.71 183	54.59 186

LTVRE_ROB	44.17 16	47.06 171	50.15 174	43.44 165	51.39 152	58.42 145	42.90 191	43.51 108	22.27 205	14.85 188	41.94 165	34.57 195	45.43 115	62.28 148	62.77 144	62.56 177	68.83 129
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
test0.0.03 1		43.15 182	46.95 184	38.72 182	55.26 123	50.56 175	42.48 192	43.48 111	38.16 186	15.11 186	35.07 179	44.69 165	16.47 195	55.95 183	54.34 182	59.54 182	49.87 196

ADS-MVSNet		40.67 188	43.38 193	37.50 186	44.36 183	39.79 201	42.09 193	32.67 188	44.34 142	28.87 154	40.76 168	40.37 182	30.22 174	48.34 201	45.87 200	46.81 202	44.21 202

ambc			45.54 189		50.66 162	52.63 170	40.99 194		38.36 183	24.67 169	22.62 201	13.94 209	29.14 178	65.71 131	58.06 165	58.60 185	67.43 134

PatchT		48.08 163	51.03 168	44.64 160	42.96 186	50.12 177	40.36 195	35.09 171	43.17 151	39.59 110	42.00 164	39.96 184	46.63 110	58.97 161	60.31 156	63.21 172	62.66 164

EU-MVSNet		40.63 189	45.65 188	34.78 192	39.11 193	46.94 189	40.02 196	34.03 178	33.50 193	10.37 196	35.57 178	37.80 191	23.65 186	51.90 191	50.21 193	61.49 179	63.62 162

test20.03		40.38 190	44.20 191	35.92 189	53.73 136	49.05 179	38.54 197	43.49 110	32.55 195	9.54 198	27.88 195	39.12 186	12.24 200	56.28 180	54.69 179	57.96 187	49.83 197

MIMVSNet1		35.51 195	41.41 195	28.63 197	27.53 206	43.36 197	38.09 198	33.82 180	32.01 196	6.77 205	21.63 202	35.43 194	11.97 202	55.05 187	53.99 185	53.59 196	48.36 199

gm-plane-assit		44.74 177	45.95 185	43.33 166	60.88 82	46.79 191	36.97 199	32.24 189	24.15 204	11.79 193	29.26 193	32.97 198	46.64 109	65.09 137	62.95 141	71.45 142	60.42 174

N_pmnet		32.67 199	36.85 200	27.79 199	40.55 191	32.13 204	35.80 200	26.79 197	37.24 188	9.10 199	32.02 185	30.94 200	16.30 196	47.22 202	41.21 201	38.21 205	37.21 203

MVS-HIRNet		42.24 184	41.15 196	43.51 164	44.06 185	40.74 198	35.77 201	35.35 170	35.38 190	38.34 115	25.63 198	38.55 189	43.48 125	50.77 194	47.03 198	64.07 169	49.98 194

testgi		38.71 192	43.64 192	32.95 193	52.30 147	48.63 182	35.59 202	35.05 172	31.58 199	9.03 201	30.29 188	40.75 180	11.19 205	55.30 185	53.47 188	54.53 194	45.48 200

pmmvs3		35.10 196	38.47 198	31.17 195	26.37 208	40.47 199	34.51 203	18.09 206	24.75 203	16.88 185	23.05 200	26.69 203	32.69 171	50.73 195	51.60 190	58.46 186	51.98 189

PMVS	27.84 18	33.81 197	35.28 201	32.09 194	34.13 197	24.81 207	32.51 204	26.48 198	26.41 202	19.37 180	23.76 199	24.02 205	25.18 183	50.78 193	47.24 197	54.89 193	49.95 195
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
FC-MVSNet-test		39.65 191	48.35 180	29.49 196	44.43 182	39.28 202	30.23 205	40.44 143	43.59 147	3.12 210	53.00 85	42.03 173	10.02 207	55.09 186	54.77 178	48.66 200	50.71 192

new-patchmatchnet		33.24 198	37.20 199	28.62 198	44.32 184	38.26 203	29.68 206	36.05 168	31.97 197	6.33 206	26.59 197	27.33 202	11.12 206	50.08 198	41.05 202	44.23 203	45.15 201

Gipuma		25.87 200	26.91 203	24.66 200	28.98 203	20.17 208	20.46 207	34.62 176	29.55 200	9.10 199	4.91 210	5.31 213	15.76 197	49.37 200	49.10 195	39.03 204	29.95 205
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
new_pmnet		23.19 201	28.17 202	17.37 201	17.03 210	24.92 206	19.66 208	16.16 208	27.05 201	4.42 207	20.77 203	19.20 208	12.19 201	37.71 203	36.38 203	34.77 206	31.17 204

PMMVS2		15.84 202	19.68 204	11.35 205	15.74 211	16.95 209	13.31 209	17.64 207	16.08 207	0.36 213	13.12 204	11.47 210	1.69 209	28.82 204	27.24 205	19.38 209	24.09 207

EMVS		14.49 204	12.45 207	16.87 204	27.02 207	12.56 212	8.13 210	27.19 196	15.05 208	3.14 209	6.69 208	2.67 215	15.08 199	14.60 208	18.05 207	20.67 208	17.56 210

E-PMN		15.09 203	13.19 206	17.30 202	27.80 205	12.62 211	7.81 211	27.54 195	14.62 209	3.19 208	6.89 207	2.52 216	15.09 198	15.93 206	20.22 206	22.38 207	19.53 208

MVE	12.28 19	13.53 205	15.72 205	10.96 206	7.39 212	15.71 210	6.05 212	23.73 201	10.29 211	3.01 211	5.77 209	3.41 214	11.91 203	20.11 205	29.79 204	13.67 210	24.98 206
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
DeepMVS_CX						6.95 213	5.98 213	2.25 209	11.73 210	2.07 212	11.85 205	5.43 212	11.75 204	11.40 209		8.10 212	18.38 209

tmp_tt				5.40 207	3.97 213	2.35 214	3.26 214	0.44 210	17.56 206	12.09 191	11.48 206	7.14 211	1.98 208	15.68 207	15.49 208	10.69 211

Patchmatch-RL test							1.04 215

testmvs		0.01 206	0.02 208	0.00 208	0.00 215	0.00 215	0.01 216	0.00 212	0.01 212	0.00 215	0.03 212	0.00 217	0.01 211	0.01 210	0.01 209	0.00 213	0.06 212

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

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

test123		0.01 206	0.02 208	0.00 208	0.00 215	0.00 215	0.00 217	0.00 212	0.01 212	0.00 215	0.04 211	0.00 217	0.01 211	0.00 211	0.01 209	0.00 213	0.07 211

SR-MVS					71.46 31			54.67 26				81.54 10

test_part1																	90.62 2

MTAPA										65.14 2		80.20 16

MTMP										62.63 14		78.04 23

mPP-MVS					71.67 30							74.36 38

NP-MVS									72.00 37

High-resolution multi-view benchmark