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.

DVP-MVS		88.67 1	91.62 1	85.22 2	90.47 14	92.36 1	90.69 6	76.15 2	93.08 1	82.75 3	92.19 4	90.71 2	80.45 4	89.27 4	87.91 7	90.82 8	95.84 1

DPE-MVS		88.63 2	91.29 2	85.53 1	90.87 6	92.20 2	91.98 2	76.00 4	90.55 6	82.09 5	93.85 1	90.75 1	81.25 1	88.62 6	87.59 12	90.96 6	95.48 2

MSP-MVS		88.09 3	90.84 3	84.88 5	90.00 20	91.80 4	91.63 3	75.80 5	91.99 2	81.23 6	92.54 2	89.18 4	80.89 2	87.99 13	87.91 7	89.70 40	94.51 6

APDe-MVS		88.00 4	90.50 4	85.08 3	90.95 5	91.58 5	92.03 1	75.53 9	91.15 3	80.10 12	92.27 3	88.34 8	80.80 3	88.00 12	86.99 17	91.09 4	95.16 5

SMA-MVS		87.56 5	90.17 5	84.52 6	91.71 2	90.57 6	90.77 5	75.19 10	90.67 5	80.50 11	86.59 14	88.86 5	78.09 13	89.92 1	89.41 1	90.84 7	95.19 4

HPM-MVS++		87.09 6	88.92 10	84.95 4	92.61 1	87.91 37	90.23 12	76.06 3	88.85 9	81.20 7	87.33 10	87.93 9	79.47 7	88.59 7	88.23 5	90.15 31	93.60 18

SD-MVS		86.96 7	89.45 6	84.05 12	90.13 17	89.23 19	89.77 15	74.59 11	89.17 7	80.70 8	89.93 8	89.67 3	78.47 9	87.57 17	86.79 21	90.67 14	93.76 14
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
TSAR-MVS + MP.		86.88 8	89.23 7	84.14 10	89.78 23	88.67 28	90.59 7	73.46 24	88.99 8	80.52 10	91.26 5	88.65 6	79.91 6	86.96 27	86.22 29	90.59 15	93.83 12
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
APD-MVS		86.84 9	88.91 11	84.41 7	90.66 9	90.10 9	90.78 4	75.64 6	87.38 14	78.72 16	90.68 7	86.82 13	80.15 5	87.13 22	86.45 27	90.51 17	93.83 12
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
ACMMP_NAP		86.52 10	89.01 8	83.62 14	90.28 16	90.09 10	90.32 10	74.05 17	88.32 11	79.74 13	87.04 12	85.59 20	76.97 26	89.35 2	88.44 4	90.35 26	94.27 10

CNVR-MVS		86.36 11	88.19 14	84.23 9	91.33 4	89.84 11	90.34 8	75.56 7	87.36 15	78.97 15	81.19 25	86.76 14	78.74 8	89.30 3	88.58 2	90.45 23	94.33 9

HFP-MVS		86.15 12	87.95 15	84.06 11	90.80 7	89.20 20	89.62 17	74.26 13	87.52 12	80.63 9	86.82 13	84.19 26	78.22 11	87.58 16	87.19 15	90.81 9	93.13 22

SteuartSystems-ACMMP		85.99 13	88.31 13	83.27 18	90.73 8	89.84 11	90.27 11	74.31 12	84.56 27	75.88 27	87.32 11	85.04 21	77.31 21	89.01 5	88.46 3	91.14 3	93.96 11
Skip Steuart: Steuart Systems R&D Blog.
zzz-MVS		85.71 14	86.88 20	84.34 8	90.54 13	87.11 41	89.77 15	74.17 15	88.54 10	83.08 2	78.60 29	86.10 16	78.11 12	87.80 15	87.46 13	90.35 26	92.56 24

ACMMPR		85.52 15	87.53 17	83.17 19	90.13 17	89.27 17	89.30 18	73.97 18	86.89 17	77.14 22	86.09 15	83.18 29	77.74 17	87.42 18	87.20 14	90.77 10	92.63 23

MP-MVS		85.50 16	87.40 18	83.28 17	90.65 10	89.51 16	89.16 21	74.11 16	83.70 31	78.06 19	85.54 17	84.89 24	77.31 21	87.40 19	87.14 16	90.41 24	93.65 17
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
NCCC		85.34 17	86.59 22	83.88 13	91.48 3	88.88 22	89.79 14	75.54 8	86.67 18	77.94 20	76.55 32	84.99 22	78.07 14	88.04 10	87.68 10	90.46 22	93.31 19

DeepPCF-MVS	79.04 1	85.30 18	88.93 9	81.06 29	88.77 33	90.48 7	85.46 43	73.08 26	90.97 4	73.77 34	84.81 19	85.95 17	77.43 20	88.22 9	87.73 9	87.85 74	94.34 8

CSCG		85.28 19	87.68 16	82.49 22	89.95 21	91.99 3	88.82 22	71.20 35	86.41 19	79.63 14	79.26 26	88.36 7	73.94 37	86.64 29	86.67 24	91.40 2	94.41 7

MCST-MVS		85.13 20	86.62 21	83.39 15	90.55 12	89.82 13	89.29 19	73.89 20	84.38 28	76.03 26	79.01 28	85.90 18	78.47 9	87.81 14	86.11 31	92.11 1	93.29 20

TSAR-MVS + ACMM		85.10 21	88.81 12	80.77 32	89.55 26	88.53 30	88.59 25	72.55 28	87.39 13	71.90 40	90.95 6	87.55 10	74.57 31	87.08 24	86.54 25	87.47 81	93.67 15

train_agg		84.86 22	87.21 19	82.11 24	90.59 11	85.47 52	89.81 13	73.55 23	83.95 29	73.30 35	89.84 9	87.23 12	75.61 29	86.47 31	85.46 36	89.78 36	92.06 30

DeepC-MVS	78.47 2	84.81 23	86.03 26	83.37 16	89.29 29	90.38 8	88.61 24	76.50 1	86.25 20	77.22 21	75.12 36	80.28 42	77.59 19	88.39 8	88.17 6	91.02 5	93.66 16
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
CP-MVS		84.74 24	86.43 24	82.77 21	89.48 27	88.13 36	88.64 23	73.93 19	84.92 22	76.77 23	81.94 23	83.50 27	77.29 23	86.92 28	86.49 26	90.49 18	93.14 21

PGM-MVS		84.42 25	86.29 25	82.23 23	90.04 19	88.82 24	89.23 20	71.74 33	82.82 34	74.61 30	84.41 20	82.09 32	77.03 25	87.13 22	86.73 23	90.73 12	92.06 30

DeepC-MVS_fast	78.24 3	84.27 26	85.50 28	82.85 20	90.46 15	89.24 18	87.83 30	74.24 14	84.88 23	76.23 25	75.26 35	81.05 40	77.62 18	88.02 11	87.62 11	90.69 13	92.41 26
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.		83.69 27	86.58 23	80.32 33	85.14 52	86.96 42	84.91 47	70.25 39	84.71 26	73.91 33	85.16 18	85.63 19	77.92 15	85.44 39	85.71 34	89.77 37	92.45 25

ACMMP		83.42 28	85.27 29	81.26 28	88.47 34	88.49 31	88.31 28	72.09 30	83.42 32	72.77 38	82.65 21	78.22 46	75.18 30	86.24 35	85.76 33	90.74 11	92.13 29
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
DPM-MVS		83.30 29	84.33 32	82.11 24	89.56 25	88.49 31	90.33 9	73.24 25	83.85 30	76.46 24	72.43 45	82.65 30	73.02 44	86.37 33	86.91 18	90.03 33	89.62 49

X-MVS		83.23 30	85.20 30	80.92 31	89.71 24	88.68 25	88.21 29	73.60 21	82.57 35	71.81 43	77.07 30	81.92 34	71.72 55	86.98 26	86.86 19	90.47 19	92.36 27

CDPH-MVS		82.64 31	85.03 31	79.86 36	89.41 28	88.31 33	88.32 27	71.84 32	80.11 42	67.47 58	82.09 22	81.44 38	71.85 53	85.89 37	86.15 30	90.24 29	91.25 36

3Dnovator+	75.73 4	82.40 32	82.76 37	81.97 26	88.02 35	89.67 14	86.60 34	71.48 34	81.28 40	78.18 18	64.78 80	77.96 48	77.13 24	87.32 20	86.83 20	90.41 24	91.48 34

PHI-MVS		82.36 33	85.89 27	78.24 46	86.40 45	89.52 15	85.52 41	69.52 46	82.38 37	65.67 65	81.35 24	82.36 31	73.07 42	87.31 21	86.76 22	89.24 47	91.56 33

MSLP-MVS++		82.09 34	82.66 38	81.42 27	87.03 41	87.22 40	85.82 39	70.04 40	80.30 41	78.66 17	68.67 64	81.04 41	77.81 16	85.19 43	84.88 41	89.19 50	91.31 35

CPTT-MVS		81.77 35	83.10 36	80.21 34	85.93 48	86.45 47	87.72 31	70.98 36	82.54 36	71.53 46	74.23 41	81.49 37	76.31 28	82.85 61	81.87 57	88.79 58	92.26 28

MVS_0304		81.73 36	83.86 33	79.26 39	86.22 47	89.18 21	86.41 35	67.15 60	75.28 52	70.75 50	74.59 38	83.49 28	74.42 33	87.05 25	86.34 28	90.58 16	91.08 38

CANet		81.62 37	83.41 34	79.53 38	87.06 40	88.59 29	85.47 42	67.96 56	76.59 50	74.05 31	74.69 37	81.98 33	72.98 45	86.14 36	85.47 35	89.68 41	90.42 44

HQP-MVS		81.19 38	83.27 35	78.76 43	87.40 38	85.45 53	86.95 32	70.47 38	81.31 39	66.91 61	79.24 27	76.63 50	71.67 56	84.43 47	83.78 47	89.19 50	92.05 32

OMC-MVS		80.26 39	82.59 39	77.54 49	83.04 60	85.54 51	83.25 54	65.05 75	87.32 16	72.42 39	72.04 48	78.97 44	73.30 40	83.86 50	81.60 61	88.15 65	88.83 53

MVS_111021_HR		80.13 40	81.46 42	78.58 44	85.77 49	85.17 56	83.45 53	69.28 47	74.08 58	70.31 51	74.31 40	75.26 55	73.13 41	86.46 32	85.15 39	89.53 43	89.81 47

LGP-MVS_train		79.83 41	81.22 44	78.22 47	86.28 46	85.36 55	86.76 33	69.59 44	77.34 47	65.14 67	75.68 34	70.79 72	71.37 59	84.60 45	84.01 44	90.18 30	90.74 40

ACMP	73.23 7	79.79 42	80.53 47	78.94 41	85.61 50	85.68 50	85.61 40	69.59 44	77.33 48	71.00 49	74.45 39	69.16 83	71.88 51	83.15 58	83.37 50	89.92 34	90.57 43
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
3Dnovator	73.76 5	79.75 43	80.52 48	78.84 42	84.94 57	87.35 38	84.43 49	65.54 71	78.29 46	73.97 32	63.00 88	75.62 54	74.07 36	85.00 44	85.34 37	90.11 32	89.04 51

AdaColmap		79.74 44	78.62 56	81.05 30	89.23 30	86.06 49	84.95 46	71.96 31	79.39 45	75.51 28	63.16 86	68.84 89	76.51 27	83.55 54	82.85 52	88.13 66	86.46 71

OPM-MVS		79.68 45	79.28 54	80.15 35	87.99 36	86.77 44	88.52 26	72.72 27	64.55 89	67.65 57	67.87 68	74.33 58	74.31 34	86.37 33	85.25 38	89.73 39	89.81 47

PCF-MVS	73.28 6	79.42 46	80.41 49	78.26 45	84.88 58	88.17 34	86.08 36	69.85 41	75.23 54	68.43 53	68.03 67	78.38 45	71.76 54	81.26 79	80.65 79	88.56 61	91.18 37
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
CLD-MVS		79.35 47	81.23 43	77.16 51	85.01 55	86.92 43	85.87 38	60.89 122	80.07 44	75.35 29	72.96 43	73.21 62	68.43 73	85.41 41	84.63 42	87.41 82	85.44 82
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
MAR-MVS		79.21 48	80.32 50	77.92 48	87.46 37	88.15 35	83.95 50	67.48 59	74.28 56	68.25 54	64.70 81	77.04 49	72.17 49	85.42 40	85.00 40	88.22 62	87.62 61
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
canonicalmvs		79.16 49	82.37 40	75.41 59	82.33 66	86.38 48	80.80 59	63.18 89	82.90 33	67.34 59	72.79 44	76.07 52	69.62 65	83.46 57	84.41 43	89.20 49	90.60 42

DELS-MVS		79.15 50	81.07 45	76.91 52	83.54 59	87.31 39	84.45 48	64.92 76	69.98 65	69.34 52	71.62 50	76.26 51	69.84 64	86.57 30	85.90 32	89.39 45	89.88 46
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
EPNet		79.08 51	80.62 46	77.28 50	88.90 32	83.17 72	83.65 51	72.41 29	74.41 55	67.15 60	76.78 31	74.37 57	64.43 92	83.70 53	83.69 48	87.15 85	88.19 56
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
ACMM	72.26 8	78.86 52	78.13 58	79.71 37	86.89 42	83.40 69	86.02 37	70.50 37	75.28 52	71.49 47	63.01 87	69.26 82	73.57 39	84.11 49	83.98 45	89.76 38	87.84 59
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
QAPM		78.47 53	80.22 51	76.43 54	85.03 54	86.75 45	80.62 62	66.00 68	73.77 59	65.35 66	65.54 76	78.02 47	72.69 46	83.71 52	83.36 51	88.87 56	90.41 45

TSAR-MVS + COLMAP		78.34 54	81.64 41	74.48 68	80.13 83	85.01 57	81.73 55	65.93 70	84.75 25	61.68 77	85.79 16	66.27 97	71.39 58	82.91 60	80.78 70	86.01 122	85.98 73

MVS_111021_LR		78.13 55	79.85 53	76.13 55	81.12 72	81.50 82	80.28 64	65.25 73	76.09 51	71.32 48	76.49 33	72.87 64	72.21 48	82.79 62	81.29 63	86.59 107	87.91 58

TAPA-MVS	71.42 9	77.69 56	80.05 52	74.94 62	80.68 76	84.52 59	81.36 56	63.14 90	84.77 24	64.82 69	68.72 62	75.91 53	71.86 52	81.62 68	79.55 95	87.80 76	85.24 85
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
ETV-MVS		77.41 57	78.94 55	75.62 58	81.86 68	83.04 73	80.59 63	63.41 86	70.65 64	63.89 72	72.11 46	68.87 88	74.10 35	85.61 38	83.89 46	89.88 35	88.38 55

CNLPA		77.20 58	77.54 63	76.80 53	82.63 62	84.31 60	79.77 68	64.64 77	85.17 21	73.18 36	56.37 122	69.81 79	74.53 32	81.12 82	78.69 105	86.04 121	87.29 65

CS-MVS		76.92 59	78.01 59	75.64 57	81.47 69	83.59 67	80.68 60	62.47 108	68.39 70	65.83 64	67.84 69	70.74 73	73.07 42	85.31 42	82.79 53	90.33 28	87.42 62

casdiffmvs		76.76 60	78.46 57	74.77 64	80.32 80	83.73 66	80.65 61	63.24 88	73.58 60	66.11 63	69.39 59	74.09 59	69.49 67	82.52 64	79.35 99	88.84 57	86.52 70

PVSNet_Blended_VisFu		76.57 61	77.90 60	75.02 61	80.56 77	86.58 46	79.24 73	66.18 65	64.81 86	68.18 55	65.61 74	71.45 67	67.05 75	84.16 48	81.80 58	88.90 54	90.92 39

PVSNet_BlendedMVS		76.21 62	77.52 64	74.69 65	79.46 86	83.79 64	77.50 90	64.34 81	69.88 66	71.88 41	68.54 65	70.42 75	67.05 75	83.48 55	79.63 91	87.89 72	86.87 67

PVSNet_Blended		76.21 62	77.52 64	74.69 65	79.46 86	83.79 64	77.50 90	64.34 81	69.88 66	71.88 41	68.54 65	70.42 75	67.05 75	83.48 55	79.63 91	87.89 72	86.87 67

OpenMVS	70.44 10	76.15 64	76.82 72	75.37 60	85.01 55	84.79 58	78.99 77	62.07 112	71.27 63	67.88 56	57.91 115	72.36 65	70.15 63	82.23 66	81.41 62	88.12 67	87.78 60

PLC	68.99 11	75.68 65	75.31 77	76.12 56	82.94 61	81.26 85	79.94 66	66.10 66	77.15 49	66.86 62	59.13 104	68.53 90	73.73 38	80.38 91	79.04 100	87.13 89	81.68 121
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
EIA-MVS		75.64 66	76.60 73	74.53 67	82.43 65	83.84 63	78.32 83	62.28 111	65.96 79	63.28 75	68.95 60	67.54 93	71.61 57	82.55 63	81.63 60	89.24 47	85.72 76

MVS_Test		75.37 67	77.13 70	73.31 72	79.07 89	81.32 84	79.98 65	60.12 133	69.72 68	64.11 71	70.53 54	73.22 61	68.90 69	80.14 97	79.48 97	87.67 78	85.50 80

Effi-MVS+		75.28 68	76.20 74	74.20 69	81.15 71	83.24 70	81.11 57	63.13 91	66.37 75	60.27 82	64.30 84	68.88 87	70.93 62	81.56 70	81.69 59	88.61 59	87.35 63

DI_MVS_plusplus_trai		75.13 69	76.12 75	73.96 70	78.18 94	81.55 80	80.97 58	62.54 105	68.59 69	65.13 68	61.43 90	74.81 56	69.32 68	81.01 84	79.59 93	87.64 79	85.89 74

diffmvs		74.86 70	77.37 67	71.93 75	75.62 117	80.35 96	79.42 72	60.15 132	72.81 62	64.63 70	71.51 51	73.11 63	66.53 85	79.02 110	77.98 113	85.25 136	86.83 69

UA-Net		74.47 71	77.80 61	70.59 85	85.33 51	85.40 54	73.54 133	65.98 69	60.65 121	56.00 103	72.11 46	79.15 43	54.63 158	83.13 59	82.25 55	88.04 68	81.92 119

LS3D		74.08 72	73.39 84	74.88 63	85.05 53	82.62 76	79.71 69	68.66 50	72.82 61	58.80 86	57.61 116	61.31 111	71.07 61	80.32 92	78.87 104	86.00 123	80.18 133

EPP-MVSNet		74.00 73	77.41 66	70.02 91	80.53 78	83.91 62	74.99 110	62.68 103	65.06 84	49.77 138	68.68 63	72.09 66	63.06 100	82.49 65	80.73 71	89.12 52	88.91 52

DCV-MVSNet		73.65 74	75.78 76	71.16 79	80.19 81	79.27 105	77.45 92	61.68 118	66.73 74	58.72 87	65.31 77	69.96 78	62.19 105	81.29 78	80.97 67	86.74 100	86.91 66

IS_MVSNet		73.33 75	77.34 68	68.65 106	81.29 70	83.47 68	74.45 114	63.58 85	65.75 81	48.49 142	67.11 73	70.61 74	54.63 158	84.51 46	83.58 49	89.48 44	86.34 72

CANet_DTU		73.29 76	76.96 71	69.00 102	77.04 106	82.06 78	79.49 71	56.30 159	67.85 72	53.29 119	71.12 53	70.37 77	61.81 114	81.59 69	80.96 68	86.09 116	84.73 93

Fast-Effi-MVS+		73.11 77	73.66 82	72.48 74	77.72 100	80.88 91	78.55 80	58.83 148	65.19 83	60.36 81	59.98 99	62.42 109	71.22 60	81.66 67	80.61 81	88.20 63	84.88 92

UGNet		72.78 78	77.67 62	67.07 128	71.65 155	83.24 70	75.20 104	63.62 84	64.93 85	56.72 99	71.82 49	73.30 60	49.02 171	81.02 83	80.70 77	86.22 113	88.67 54
Wanjuan Su, Qingshan Xu, Wenbing Tao: Uncertainty-guided Multi-view Stereo Network for Depth Estimation. IEEE Transactions on Circuits and Systems for Video Technology, 2022
Vis-MVSNet		72.77 79	77.20 69	67.59 118	74.19 131	84.01 61	76.61 100	61.69 117	60.62 122	50.61 134	70.25 56	71.31 70	55.57 154	83.85 51	82.28 54	86.90 94	88.08 57
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
FC-MVSNet-train		72.60 80	75.07 78	69.71 94	81.10 73	78.79 111	73.74 132	65.23 74	66.10 78	53.34 118	70.36 55	63.40 106	56.92 143	81.44 72	80.96 68	87.93 70	84.46 97

ET-MVSNet_ETH3D		72.46 81	74.19 80	70.44 86	62.50 189	81.17 86	79.90 67	62.46 109	64.52 90	57.52 95	71.49 52	59.15 120	72.08 50	78.61 115	81.11 65	88.16 64	83.29 107

MVSTER		72.06 82	74.24 79	69.51 97	70.39 166	75.97 141	76.91 96	57.36 155	64.64 88	61.39 79	68.86 61	63.76 104	63.46 97	81.44 72	79.70 90	87.56 80	85.31 84

Anonymous20231211		71.90 83	72.48 92	71.21 78	80.14 82	81.53 81	76.92 95	62.89 94	64.46 91	58.94 84	43.80 181	70.98 71	62.22 104	80.70 86	80.19 86	86.18 114	85.73 75

Effi-MVS+-dtu		71.82 84	71.86 97	71.78 76	78.77 90	80.47 94	78.55 80	61.67 119	60.68 120	55.49 104	58.48 108	65.48 99	68.85 70	76.92 132	75.55 144	87.35 83	85.46 81

IterMVS-LS		71.69 85	72.82 90	70.37 87	77.54 102	76.34 138	75.13 108	60.46 128	61.53 115	57.57 94	64.89 79	67.33 94	66.04 88	77.09 131	77.37 125	85.48 132	85.18 86
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
MSDG		71.52 86	69.87 109	73.44 71	82.21 67	79.35 104	79.52 70	64.59 78	66.15 77	61.87 76	53.21 146	56.09 135	65.85 89	78.94 111	78.50 107	86.60 106	76.85 156

thisisatest0530		71.48 87	73.01 87	69.70 95	73.83 136	78.62 113	74.53 113	59.12 142	64.13 92	58.63 88	64.60 82	58.63 122	64.27 93	80.28 94	80.17 87	87.82 75	84.64 95

tttt0517		71.41 88	72.95 88	69.60 96	73.70 138	78.70 112	74.42 117	59.12 142	63.89 96	58.35 91	64.56 83	58.39 124	64.27 93	80.29 93	80.17 87	87.74 77	84.69 94

ACMH+	66.54 13	71.36 89	70.09 107	72.85 73	82.59 63	81.13 87	78.56 79	68.04 54	61.55 114	52.52 125	51.50 161	54.14 145	68.56 72	78.85 112	79.50 96	86.82 97	83.94 101

IB-MVS	66.94 12	71.21 90	71.66 98	70.68 82	79.18 88	82.83 75	72.61 139	61.77 116	59.66 126	63.44 74	53.26 144	59.65 118	59.16 127	76.78 135	82.11 56	87.90 71	87.33 64
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
GBi-Net		70.78 91	73.37 85	67.76 111	72.95 143	78.00 118	75.15 105	62.72 98	64.13 92	51.44 127	58.37 109	69.02 84	57.59 135	81.33 75	80.72 72	86.70 101	82.02 113

test1		70.78 91	73.37 85	67.76 111	72.95 143	78.00 118	75.15 105	62.72 98	64.13 92	51.44 127	58.37 109	69.02 84	57.59 135	81.33 75	80.72 72	86.70 101	82.02 113

ACMH	65.37 14	70.71 93	70.00 108	71.54 77	82.51 64	82.47 77	77.78 87	68.13 53	56.19 148	46.06 157	54.30 132	51.20 172	68.68 71	80.66 87	80.72 72	86.07 117	84.45 98
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
UniMVSNet_NR-MVSNet		70.59 94	72.19 93	68.72 104	77.72 100	80.72 92	73.81 130	69.65 43	61.99 109	43.23 166	60.54 95	57.50 127	58.57 128	79.56 103	81.07 66	89.34 46	83.97 99

FMVSNet3		70.49 95	72.90 89	67.67 116	72.88 146	77.98 121	74.96 111	62.72 98	64.13 92	51.44 127	58.37 109	69.02 84	57.43 138	79.43 105	79.57 94	86.59 107	81.81 120

baseline		70.45 96	74.09 81	66.20 136	70.95 163	75.67 142	74.26 122	53.57 163	68.33 71	58.42 89	69.87 57	71.45 67	61.55 115	74.84 146	74.76 149	78.42 169	83.72 104

FMVSNet2		70.39 97	72.67 91	67.72 114	72.95 143	78.00 118	75.15 105	62.69 102	63.29 100	51.25 131	55.64 124	68.49 91	57.59 135	80.91 85	80.35 84	86.70 101	82.02 113

v8		70.23 98	69.86 110	70.67 83	74.69 126	79.82 100	78.79 78	59.18 141	58.80 130	58.20 92	55.00 129	57.33 128	66.31 87	77.51 125	76.71 134	86.82 97	83.88 102

v10		70.22 99	69.76 112	70.74 80	74.79 125	80.30 98	79.22 74	59.81 136	57.71 137	56.58 101	54.22 137	55.31 138	66.95 78	78.28 118	77.47 122	87.12 91	85.07 88

MS-PatchMatch		70.17 100	70.49 104	69.79 93	80.98 74	77.97 123	77.51 89	58.95 145	62.33 107	55.22 107	53.14 147	65.90 98	62.03 108	79.08 109	77.11 129	84.08 146	77.91 147

baseline1		70.10 101	72.17 94	67.69 115	79.74 84	76.80 133	73.91 126	64.38 80	62.74 105	48.30 144	64.94 78	64.08 103	54.17 160	81.46 71	78.92 102	85.66 129	76.22 158

v2v482		70.05 102	69.46 116	70.74 80	74.62 127	80.32 97	79.00 76	60.62 125	57.41 139	56.89 98	55.43 128	55.14 140	66.39 86	77.25 128	77.14 128	86.90 94	83.57 106

v1144		69.93 103	69.36 117	70.61 84	74.89 124	80.93 88	79.11 75	60.64 124	55.97 150	55.31 106	53.85 139	54.14 145	66.54 84	78.10 120	77.44 123	87.14 88	85.09 87

baseline2		69.69 104	70.27 106	69.01 101	75.72 116	77.13 131	73.82 129	58.94 146	61.35 116	57.09 97	61.68 89	57.17 130	61.99 109	78.10 120	76.58 136	86.48 110	79.85 135

DU-MVS		69.63 105	70.91 101	68.13 110	75.99 112	79.54 101	73.81 130	69.20 48	61.20 118	43.23 166	58.52 106	53.50 152	58.57 128	79.22 107	80.45 82	87.97 69	83.97 99

UniMVSNet (Re)		69.53 106	71.90 96	66.76 133	76.42 109	80.93 88	72.59 140	68.03 55	61.75 113	41.68 171	58.34 112	57.23 129	53.27 163	79.53 104	80.62 80	88.57 60	84.90 91

v1192		69.50 107	68.83 123	70.29 88	74.49 128	80.92 90	78.55 80	60.54 126	55.04 156	54.21 109	52.79 153	52.33 165	66.92 79	77.88 122	77.35 126	87.04 92	85.51 79

HyFIR lowres test		69.47 108	68.94 122	70.09 90	76.77 108	82.93 74	76.63 99	60.17 131	59.00 129	54.03 112	40.54 190	65.23 100	67.89 74	76.54 138	78.30 110	85.03 139	80.07 134

v144192		69.34 109	68.68 127	70.12 89	74.06 132	80.54 93	78.08 86	60.54 126	54.99 158	54.13 111	52.92 151	52.80 163	66.73 82	77.13 130	76.72 133	87.15 85	85.63 77

TranMVSNet+NR-MVSNet		69.25 110	70.81 102	67.43 119	77.23 105	79.46 103	73.48 135	69.66 42	60.43 123	39.56 174	58.82 105	53.48 154	55.74 152	79.59 101	81.21 64	88.89 55	82.70 109

CHOSEN 1792x2688		69.20 111	69.26 118	69.13 99	76.86 107	78.93 107	77.27 93	60.12 133	61.86 111	54.42 108	42.54 185	61.61 110	66.91 80	78.55 116	78.14 112	79.23 167	83.23 108

v1921920		69.03 112	68.32 131	69.86 92	74.03 133	80.37 95	77.55 88	60.25 130	54.62 160	53.59 117	52.36 157	51.50 171	66.75 81	77.17 129	76.69 135	86.96 93	85.56 78

CostFormer		68.92 113	69.58 114	68.15 109	75.98 114	76.17 140	78.22 85	51.86 175	65.80 80	61.56 78	63.57 85	62.83 107	61.85 112	70.40 178	68.67 176	79.42 165	79.62 138

FMVSNet1		68.84 114	70.47 105	66.94 130	71.35 160	77.68 126	74.71 112	62.35 110	56.93 141	49.94 137	50.01 167	64.59 101	57.07 140	81.33 75	80.72 72	86.25 112	82.00 116

NR-MVSNet		68.79 115	70.56 103	66.71 135	77.48 103	79.54 101	73.52 134	69.20 48	61.20 118	39.76 173	58.52 106	50.11 178	51.37 167	80.26 95	80.71 76	88.97 53	83.59 105

V42		68.76 116	69.63 113	67.74 113	64.93 186	78.01 117	78.30 84	56.48 158	58.65 131	56.30 102	54.26 135	57.03 131	64.85 91	77.47 126	77.01 130	85.60 130	84.96 90

v1240		68.64 117	67.89 136	69.51 97	73.89 135	80.26 99	76.73 98	59.97 135	53.43 167	53.08 120	51.82 160	50.84 174	66.62 83	76.79 134	76.77 132	86.78 99	85.34 83

Fast-Effi-MVS+-dtu		68.34 118	69.47 115	67.01 129	75.15 120	77.97 123	77.12 94	55.40 161	57.87 132	46.68 154	56.17 123	60.39 112	62.36 103	76.32 139	76.25 140	85.35 135	81.34 123

GA-MVS		68.14 119	69.17 120	66.93 131	73.77 137	78.50 115	74.45 114	58.28 150	55.11 155	48.44 143	60.08 97	53.99 148	61.50 116	78.43 117	77.57 120	85.13 137	80.54 129

tfpn200view9		68.11 120	68.72 126	67.40 120	77.83 98	78.93 107	74.28 120	62.81 95	56.64 143	46.82 152	52.65 154	53.47 155	56.59 144	80.41 88	78.43 108	86.11 115	80.52 130

EPNet_dtu		68.08 121	71.00 100	64.67 144	79.64 85	68.62 175	75.05 109	63.30 87	66.36 76	45.27 161	67.40 71	66.84 96	43.64 180	75.37 142	74.98 148	81.15 158	77.44 150
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
thres200		67.98 122	68.55 129	67.30 123	77.89 97	78.86 109	74.18 124	62.75 96	56.35 146	46.48 155	52.98 150	53.54 151	56.46 145	80.41 88	77.97 114	86.05 119	79.78 137

thres400		67.95 123	68.62 128	67.17 125	77.90 95	78.59 114	74.27 121	62.72 98	56.34 147	45.77 159	53.00 149	53.35 158	56.46 145	80.21 96	78.43 108	85.91 126	80.43 131

pmmvs4		67.89 124	67.39 141	68.48 107	71.60 157	73.57 157	74.45 114	60.98 121	64.65 87	57.97 93	54.95 130	51.73 170	61.88 111	73.78 152	75.11 146	83.99 148	77.91 147

v148		67.85 125	67.53 137	68.23 108	73.25 141	77.57 129	74.26 122	57.36 155	55.70 151	57.45 96	53.53 140	55.42 137	61.96 110	75.23 143	73.92 152	85.08 138	81.32 124

Vis-MVSNet (Re-imp)		67.83 126	73.52 83	61.19 160	78.37 93	76.72 135	66.80 165	62.96 92	65.50 82	34.17 184	67.19 72	69.68 80	39.20 188	79.39 106	79.44 98	85.68 128	76.73 157

PatchMatch-RL		67.78 127	66.65 146	69.10 100	73.01 142	72.69 160	68.49 155	61.85 115	62.93 103	60.20 83	56.83 121	50.42 176	69.52 66	75.62 141	74.46 151	81.51 156	73.62 175

thres600view7		67.68 128	68.43 130	66.80 132	77.90 95	78.86 109	73.84 128	62.75 96	56.07 149	44.70 164	52.85 152	52.81 162	55.58 153	80.41 88	77.77 116	86.05 119	80.28 132

COLMAP_ROB	62.73 15	67.66 129	66.76 145	68.70 105	80.49 79	77.98 121	75.29 103	62.95 93	63.62 98	49.96 136	47.32 176	50.72 175	58.57 128	76.87 133	75.50 145	84.94 141	75.33 167
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
CDS-MVSNet		67.65 130	69.83 111	65.09 140	75.39 119	76.55 136	74.42 117	63.75 83	53.55 166	49.37 140	59.41 102	62.45 108	44.44 178	79.71 100	79.82 89	83.17 152	77.36 151
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
RPSCF		67.64 131	71.25 99	63.43 153	61.86 191	70.73 167	67.26 160	50.86 180	74.20 57	58.91 85	67.49 70	69.33 81	64.10 95	71.41 165	68.45 180	77.61 171	77.17 152

thres100view900		67.60 132	68.02 133	67.12 127	77.83 98	77.75 125	73.90 127	62.52 106	56.64 143	46.82 152	52.65 154	53.47 155	55.92 149	78.77 113	77.62 119	85.72 127	79.23 140

Baseline_NR-MVSNet		67.53 133	68.77 125	66.09 137	75.99 112	74.75 152	72.43 141	68.41 51	61.33 117	38.33 177	51.31 162	54.13 147	56.03 148	79.22 107	78.19 111	85.37 134	82.45 111

thisisatest0515		67.40 134	68.78 124	65.80 138	70.02 168	75.24 148	69.36 152	57.37 154	54.94 159	53.67 116	55.53 127	54.85 141	58.00 133	78.19 119	78.91 103	86.39 111	83.78 103

USDC		67.36 135	67.90 135	66.74 134	71.72 153	75.23 149	71.58 143	60.28 129	67.45 73	50.54 135	60.93 91	45.20 192	62.08 106	76.56 137	74.50 150	84.25 145	75.38 166

DWT-MVSNet_training		67.24 136	65.96 148	68.74 103	76.15 110	74.36 155	74.37 119	56.66 157	61.82 112	60.51 80	58.23 114	49.76 180	65.07 90	70.04 179	70.39 167	79.70 164	77.11 154

EG-PatchMatch MVS		67.24 136	66.94 143	67.60 117	78.73 91	81.35 83	73.28 137	59.49 138	46.89 189	51.42 130	43.65 182	53.49 153	55.50 155	81.38 74	80.66 78	87.15 85	81.17 125

UniMVSNet_ETH3D		67.18 138	67.03 142	67.36 121	74.44 129	78.12 116	74.07 125	66.38 63	52.22 172	46.87 151	48.64 171	51.84 169	56.96 141	77.29 127	78.53 106	85.42 133	82.59 110

v7n		67.05 139	66.94 143	67.17 125	72.35 148	78.97 106	73.26 138	58.88 147	51.16 178	50.90 132	48.21 173	50.11 178	60.96 119	77.70 123	77.38 124	86.68 104	85.05 89

IterMVS-SCA-FT		66.89 140	69.22 119	64.17 146	71.30 161	75.64 143	71.33 144	53.17 167	57.63 138	49.08 141	60.72 93	60.05 116	63.09 99	74.99 145	73.92 152	77.07 175	81.57 122

IterMVS		66.36 141	68.30 132	64.10 147	69.48 173	74.61 153	73.41 136	50.79 181	57.30 140	48.28 145	60.64 94	59.92 117	60.85 123	74.14 150	72.66 159	81.80 155	78.82 143
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
TDRefinement		66.09 142	65.03 159	67.31 122	69.73 170	76.75 134	75.33 101	64.55 79	60.28 124	49.72 139	45.63 179	42.83 195	60.46 124	75.75 140	75.95 141	84.08 146	78.04 146

pm-mvs1		65.62 143	67.42 139	63.53 152	73.66 139	76.39 137	69.66 149	60.87 123	49.73 182	43.97 165	51.24 163	57.00 132	48.16 172	79.89 98	77.84 115	84.85 143	79.82 136

tpm cat1		65.41 144	63.81 167	67.28 124	75.61 118	72.88 159	75.32 102	52.85 169	62.97 102	63.66 73	53.24 145	53.29 160	61.83 113	65.54 190	64.14 192	74.43 187	74.60 169

SCA		65.40 145	66.58 147	64.02 148	70.65 164	73.37 158	67.35 159	53.46 165	63.66 97	54.14 110	60.84 92	60.20 115	61.50 116	69.96 180	68.14 181	77.01 176	69.91 181

anonymousdsp		65.28 146	67.98 134	62.13 156	58.73 197	73.98 156	67.10 162	50.69 182	48.41 185	47.66 150	54.27 133	52.75 164	61.45 118	76.71 136	80.20 85	87.13 89	89.53 50

PMMVS		65.06 147	69.17 120	60.26 165	55.25 203	63.43 191	66.71 166	43.01 199	62.41 106	50.64 133	69.44 58	67.04 95	63.29 98	74.36 149	73.54 155	82.68 153	73.99 174

CR-MVSNet		64.83 148	65.54 153	64.01 149	70.64 165	69.41 170	65.97 170	52.74 170	57.81 134	52.65 122	54.27 133	56.31 134	60.92 120	72.20 161	73.09 157	81.12 159	75.69 163

TransMVSNet (Re)		64.74 149	65.66 152	63.66 151	77.40 104	75.33 147	69.86 148	62.67 104	47.63 187	41.21 172	50.01 167	52.33 165	45.31 177	79.57 102	77.69 118	85.49 131	77.07 155

test-LLR		64.42 150	64.36 163	64.49 145	75.02 122	63.93 188	66.61 167	61.96 113	54.41 161	47.77 147	57.46 117	60.25 113	55.20 156	70.80 172	69.33 171	80.40 162	74.38 171

MDTV_nov1_ep13		64.37 151	65.24 155	63.37 154	68.94 175	70.81 166	72.40 142	50.29 184	60.10 125	53.91 114	60.07 98	59.15 120	57.21 139	69.43 183	67.30 183	77.47 172	69.78 183

tfpnnormal		64.27 152	63.64 168	65.02 141	75.84 115	75.61 144	71.24 146	62.52 106	47.79 186	42.97 168	42.65 184	44.49 193	52.66 165	78.77 113	76.86 131	84.88 142	79.29 139

PatchmatchNet		64.21 153	64.65 161	63.69 150	71.29 162	68.66 174	69.63 150	51.70 177	63.04 101	53.77 115	59.83 101	58.34 125	60.23 125	68.54 186	66.06 188	75.56 182	68.08 187
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
dps		64.00 154	62.99 170	65.18 139	73.29 140	72.07 162	68.98 154	53.07 168	57.74 136	58.41 90	55.55 126	47.74 186	60.89 122	69.53 182	67.14 185	76.44 179	71.19 179

pmmvs-eth3d		63.52 155	62.44 177	64.77 143	66.82 181	70.12 169	69.41 151	59.48 139	54.34 164	52.71 121	46.24 178	44.35 194	56.93 142	72.37 156	73.77 154	83.30 150	75.91 160

WR-MVS		63.03 156	67.40 140	57.92 175	75.14 121	77.60 128	60.56 188	66.10 66	54.11 165	23.88 195	53.94 138	53.58 150	34.50 192	73.93 151	77.71 117	87.35 83	80.94 126

PEN-MVS		62.96 157	65.77 151	59.70 168	73.98 134	75.45 145	63.39 181	67.61 58	52.49 170	25.49 194	53.39 141	49.12 182	40.85 186	71.94 163	77.26 127	86.86 96	80.72 128

TinyColmap		62.84 158	61.03 183	64.96 142	69.61 171	71.69 163	68.48 156	59.76 137	55.41 152	47.69 149	47.33 175	34.20 203	62.76 102	74.52 147	72.59 160	81.44 157	71.47 178

CP-MVSNet		62.68 159	65.49 154	59.40 171	71.84 151	75.34 146	62.87 183	67.04 61	52.64 169	27.19 192	53.38 142	48.15 184	41.40 184	71.26 166	75.68 142	86.07 117	82.00 116

gg-mvs-nofinetune		62.55 160	65.05 158	59.62 169	78.72 92	77.61 127	70.83 147	53.63 162	39.71 201	22.04 201	36.36 194	64.32 102	47.53 173	81.16 80	79.03 101	85.00 140	77.17 152

CVMVSNet		62.55 160	65.89 149	58.64 173	66.95 179	69.15 172	66.49 169	56.29 160	52.46 171	32.70 185	59.27 103	58.21 126	50.09 169	71.77 164	71.39 164	79.31 166	78.99 142

CMPMVS	47.78 17	62.49 162	62.52 175	62.46 155	70.01 169	70.66 168	62.97 182	51.84 176	51.98 174	56.71 100	42.87 183	53.62 149	57.80 134	72.23 159	70.37 168	75.45 184	75.91 160
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
pmmvs6		62.41 163	62.88 171	61.87 157	71.38 159	75.18 151	67.76 158	59.45 140	41.64 197	42.52 170	37.33 192	52.91 161	46.87 174	77.67 124	76.26 139	83.23 151	79.18 141

tpm		62.41 163	63.15 169	61.55 159	72.24 149	63.79 190	71.31 145	46.12 197	57.82 133	55.33 105	59.90 100	54.74 142	53.63 161	67.24 189	64.29 191	70.65 197	74.25 173

PS-CasMVS		62.38 165	65.06 157	59.25 172	71.73 152	75.21 150	62.77 184	66.99 62	51.94 176	26.96 193	52.00 159	47.52 187	41.06 185	71.16 169	75.60 143	85.97 124	81.97 118

pmmvs5		62.37 166	64.04 165	60.42 163	65.03 184	71.67 164	67.17 161	52.70 172	50.30 179	44.80 162	54.23 136	51.19 173	49.37 170	72.88 155	73.48 156	83.45 149	74.55 170

tpmrst		62.00 167	62.35 178	61.58 158	71.62 156	64.14 187	69.07 153	48.22 193	62.21 108	53.93 113	58.26 113	55.30 139	55.81 151	63.22 194	62.62 194	70.85 196	70.70 180

PatchT		61.97 168	64.04 165	59.55 170	60.49 193	67.40 178	56.54 194	48.65 189	56.69 142	52.65 122	51.10 164	52.14 168	60.92 120	72.20 161	73.09 157	78.03 170	75.69 163

DTE-MVSNet		61.85 169	64.96 160	58.22 174	74.32 130	74.39 154	61.01 187	67.85 57	51.76 177	21.91 202	53.28 143	48.17 183	37.74 189	72.22 160	76.44 137	86.52 109	78.49 144

SixPastTwentyTwo		61.84 170	62.45 176	61.12 161	69.20 174	72.20 161	62.03 185	57.40 153	46.54 190	38.03 179	57.14 120	41.72 197	58.12 132	69.67 181	71.58 163	81.94 154	78.30 145

WR-MVS_H		61.83 171	65.87 150	57.12 178	71.72 153	76.87 132	61.45 186	66.19 64	51.97 175	22.92 199	53.13 148	52.30 167	33.80 193	71.03 170	75.00 147	86.65 105	80.78 127

LTVRE_ROB	59.44 16	61.82 172	62.64 174	60.87 162	72.83 147	77.19 130	64.37 177	58.97 144	33.56 205	28.00 191	52.59 156	42.21 196	63.93 96	74.52 147	76.28 138	77.15 174	82.13 112
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
RPMNet		61.71 173	62.88 171	60.34 164	69.51 172	69.41 170	63.48 180	49.23 185	57.81 134	45.64 160	50.51 165	50.12 177	53.13 164	68.17 188	68.49 179	81.07 160	75.62 165

TESTMET0.1,1		61.10 174	64.36 163	57.29 177	57.53 198	63.93 188	66.61 167	36.22 203	54.41 161	47.77 147	57.46 117	60.25 113	55.20 156	70.80 172	69.33 171	80.40 162	74.38 171

test-mter		60.84 175	64.62 162	56.42 180	55.99 201	64.18 186	65.39 172	34.23 204	54.39 163	46.21 156	57.40 119	59.49 119	55.86 150	71.02 171	69.65 170	80.87 161	76.20 159

PM-MVS		60.48 176	60.94 184	59.94 166	58.85 196	66.83 181	64.27 178	51.39 178	55.03 157	48.03 146	50.00 169	40.79 199	58.26 131	69.20 184	67.13 186	78.84 168	77.60 149

MDTV_nov1_ep13_2view		60.16 177	60.51 185	59.75 167	65.39 183	69.05 173	68.00 157	48.29 191	51.99 173	45.95 158	48.01 174	49.64 181	53.39 162	68.83 185	66.52 187	77.47 172	69.55 184

EPMVS		60.00 178	61.97 179	57.71 176	68.46 176	63.17 194	64.54 176	48.23 192	63.30 99	44.72 163	60.19 96	56.05 136	50.85 168	65.27 192	62.02 195	69.44 199	63.81 193

TAMVS		59.58 179	62.81 173	55.81 182	66.03 182	65.64 185	63.86 179	48.74 188	49.95 181	37.07 181	54.77 131	58.54 123	44.44 178	72.29 158	71.79 161	74.70 186	66.66 189

test0.0.03 1		58.80 180	61.58 181	55.56 183	75.02 122	68.45 176	59.58 192	61.96 113	52.74 168	29.57 188	49.75 170	54.56 143	31.46 195	71.19 167	69.77 169	75.75 180	64.57 191

CHOSEN 280x420		58.70 181	61.88 180	54.98 185	55.45 202	50.55 205	64.92 174	40.36 200	55.21 153	38.13 178	48.31 172	63.76 104	63.03 101	73.73 153	68.58 178	68.00 201	73.04 176

MIMVSNet		58.52 182	61.34 182	55.22 184	60.76 192	67.01 180	66.81 164	49.02 187	56.43 145	38.90 176	40.59 189	54.54 144	40.57 187	73.16 154	71.65 162	75.30 185	66.00 190

FMVSNet5		57.24 183	60.02 186	53.99 188	56.45 200	62.74 195	65.27 173	47.03 194	55.14 154	39.55 175	40.88 187	53.42 157	41.83 181	72.35 157	71.10 166	73.79 189	64.50 192

gm-plane-assit		57.00 184	57.62 191	56.28 181	76.10 111	62.43 197	47.62 204	46.57 195	33.84 204	23.24 197	37.52 191	40.19 200	59.61 126	79.81 99	77.55 121	84.55 144	72.03 177

FC-MVSNet-test		56.90 185	65.20 156	47.21 195	66.98 178	63.20 193	49.11 203	58.60 149	59.38 128	11.50 208	65.60 75	56.68 133	24.66 202	71.17 168	71.36 165	72.38 193	69.02 185

Anonymous20231206		56.36 186	57.80 190	54.67 186	70.08 167	66.39 182	60.46 189	57.54 152	49.50 184	29.30 189	33.86 197	46.64 188	35.18 191	70.44 176	68.88 175	75.47 183	68.88 186

ADS-MVSNet		55.94 187	58.01 188	53.54 190	62.48 190	58.48 199	59.12 193	46.20 196	59.65 127	42.88 169	52.34 158	53.31 159	46.31 175	62.00 196	60.02 197	64.23 203	60.24 200

EU-MVSNet		54.63 188	58.69 187	49.90 193	56.99 199	62.70 196	56.41 195	50.64 183	45.95 192	23.14 198	50.42 166	46.51 189	36.63 190	65.51 191	64.85 190	75.57 181	74.91 168

MVS-HIRNet		54.41 189	52.10 196	57.11 179	58.99 195	56.10 202	49.68 202	49.10 186	46.18 191	52.15 126	33.18 198	46.11 190	56.10 147	63.19 195	59.70 198	76.64 178	60.25 199

testgi		54.39 190	57.86 189	50.35 192	71.59 158	67.24 179	54.95 196	53.25 166	43.36 194	23.78 196	44.64 180	47.87 185	24.96 200	70.45 175	68.66 177	73.60 190	62.78 196

test20.03		53.93 191	56.28 192	51.19 191	72.19 150	65.83 183	53.20 198	61.08 120	42.74 195	22.08 200	37.07 193	45.76 191	24.29 203	70.44 176	69.04 173	74.31 188	63.05 195

MDA-MVSNet-bldmvs		53.37 192	53.01 195	53.79 189	43.67 207	67.95 177	59.69 191	57.92 151	43.69 193	32.41 186	41.47 186	27.89 208	52.38 166	56.97 202	65.99 189	76.68 177	67.13 188

FPMVS		51.87 193	50.00 198	54.07 187	66.83 180	57.25 200	60.25 190	50.91 179	50.25 180	34.36 183	36.04 195	32.02 205	41.49 183	58.98 200	56.07 199	70.56 198	59.36 201

MIMVSNet1		49.27 194	53.25 194	44.62 197	44.61 205	61.52 198	53.61 197	52.18 173	41.62 198	18.68 204	28.14 203	41.58 198	25.50 198	68.46 187	69.04 173	73.15 191	62.37 197

pmmvs3		47.65 195	49.08 200	45.99 196	44.61 205	54.79 203	50.04 200	31.95 207	33.91 203	29.90 187	30.37 199	33.53 204	46.31 175	63.50 193	63.67 193	73.14 192	63.77 194

N_pmnet		47.35 196	50.13 197	44.11 198	59.98 194	51.64 204	51.86 199	44.80 198	49.58 183	20.76 203	40.65 188	40.05 201	29.64 196	59.84 198	55.15 200	57.63 204	54.00 203

new-patchmatchnet		46.97 197	49.47 199	44.05 199	62.82 188	56.55 201	45.35 205	52.01 174	42.47 196	17.04 206	35.73 196	35.21 202	21.84 206	61.27 197	54.83 201	65.26 202	60.26 198

GG-mvs-BLEND		46.86 198	67.51 138	22.75 204	0.05 214	76.21 139	64.69 175	0.04 211	61.90 110	0.09 214	55.57 125	71.32 69	0.08 210	70.54 174	67.19 184	71.58 194	69.86 182

PMVS	39.38 18	46.06 199	43.30 201	49.28 194	62.93 187	38.75 207	41.88 206	53.50 164	33.33 206	35.46 182	28.90 202	31.01 206	33.04 194	58.61 201	54.63 202	68.86 200	57.88 202
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
new_pmnet		38.40 200	42.64 202	33.44 201	37.54 210	45.00 206	36.60 207	32.72 206	40.27 199	12.72 207	29.89 200	28.90 207	24.78 201	53.17 203	52.90 203	56.31 205	48.34 204

Gipuma		36.38 201	35.80 203	37.07 200	45.76 204	33.90 208	29.81 208	48.47 190	39.91 200	18.02 205	8.00 210	8.14 213	25.14 199	59.29 199	61.02 196	55.19 206	40.31 205
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
PMMVS2		25.60 202	29.75 204	20.76 205	28.00 211	30.93 209	23.10 209	29.18 208	23.14 208	1.46 213	18.23 206	16.54 210	5.08 208	40.22 204	41.40 205	37.76 207	37.79 207

E-PMN		21.77 203	18.24 206	25.89 202	40.22 208	19.58 211	12.46 212	39.87 201	18.68 210	6.71 210	9.57 207	4.31 216	22.36 205	19.89 208	27.28 207	33.73 208	28.34 209

EMVS		20.98 204	17.15 207	25.44 203	39.51 209	19.37 212	12.66 211	39.59 202	19.10 209	6.62 211	9.27 208	4.40 215	22.43 204	17.99 209	24.40 208	31.81 209	25.53 210

MVE	19.12 19	20.47 205	23.27 205	17.20 206	12.66 213	25.41 210	10.52 213	34.14 205	14.79 211	6.53 212	8.79 209	4.68 214	16.64 207	29.49 206	41.63 204	22.73 211	38.11 206
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
testmvs		0.09 206	0.15 208	0.02 208	0.01 215	0.02 215	0.05 216	0.01 212	0.11 212	0.01 215	0.26 212	0.01 217	0.06 212	0.10 210	0.10 209	0.01 213	0.43 212

test123		0.09 206	0.14 209	0.02 208	0.00 216	0.02 215	0.02 217	0.01 212	0.09 213	0.00 216	0.30 211	0.00 218	0.08 210	0.03 211	0.09 210	0.01 213	0.45 211

sosnet-low-res		0.00 208	0.00 210	0.00 210	0.00 216	0.00 217	0.00 218	0.00 214	0.00 214	0.00 216	0.00 213	0.00 218	0.00 213	0.00 212	0.00 211	0.00 215	0.00 213

sosnet		0.00 208	0.00 210	0.00 210	0.00 216	0.00 217	0.00 218	0.00 214	0.00 214	0.00 216	0.00 213	0.00 218	0.00 213	0.00 212	0.00 211	0.00 215	0.00 213

SR-MVS					88.99 31			73.57 22				87.54 11

Anonymous202405211			72.16 95		80.85 75	81.85 79	76.88 97	65.40 72	62.89 104		46.35 177	67.99 92	62.05 107	81.15 81	80.38 83	85.97 124	84.50 96

our_test_3					67.93 177	70.99 165	66.89 163

test_part1																	95.35 3

ambc			53.42 193		64.99 185	63.36 192	49.96 201		47.07 188	37.12 180	28.97 201	16.36 211	41.82 182	75.10 144	67.34 182	71.55 195	75.72 162

MTAPA										83.48 1		86.45 15

MTMP										82.66 4		84.91 23

Patchmatch-RL test							2.85 215

tmp_tt				14.50 207	14.68 212	7.17 214	10.46 214	2.21 210	37.73 202	28.71 190	25.26 204	16.98 209	4.37 209	31.49 205	29.77 206	26.56 210

XVS					86.63 43	88.68 25	85.00 44			71.81 43		81.92 34				90.47 19

X-MVStestdata					86.63 43	88.68 25	85.00 44			71.81 43		81.92 34				90.47 19

abl_6				79.05 40	87.27 39	88.85 23	83.62 52	68.25 52	81.68 38	72.94 37	73.79 42	84.45 25	72.55 47			89.66 42	90.64 41

mPP-MVS					89.90 22							81.29 39

NP-MVS									80.10 43

Patchmtry						65.80 184	65.97 170	52.74 170		52.65 122

DeepMVS_CX						18.74 213	18.55 210	8.02 209	26.96 207	7.33 209	23.81 205	13.05 212	25.99 197	25.17 207		22.45 212	36.25 208

High-resolution multi-view benchmark