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.

MCST-MVS		85.13 24	86.62 24	83.39 19	90.55 15	89.82 17	89.29 23	73.89 24	84.38 32	76.03 30	79.01 32	85.90 22	78.47 13	87.81 16	86.11 35	92.11 1	93.29 23

CSCG		85.28 23	87.68 19	82.49 26	89.95 25	91.99 5	88.82 26	71.20 39	86.41 23	79.63 18	79.26 30	88.36 10	73.94 43	86.64 33	86.67 26	91.40 2	94.41 8

xxxxxxxxxxxxxcwj		85.35 20	85.76 31	84.86 8	91.26 6	91.10 8	90.90 6	75.65 8	89.21 9	81.25 8	91.12 8	61.35 120	78.82 10	87.42 20	86.23 31	91.28 3	93.90 13

SF-MVS		87.47 8	89.70 8	84.86 8	91.26 6	91.10 8	90.90 6	75.65 8	89.21 9	81.25 8	91.12 8	88.93 7	78.82 10	87.42 20	86.23 31	91.28 3	93.90 13

SteuartSystems-ACMMP		85.99 16	88.31 16	83.27 22	90.73 11	89.84 15	90.27 15	74.31 16	84.56 31	75.88 31	87.32 15	85.04 25	77.31 25	89.01 7	88.46 3	91.14 5	93.96 12
Skip Steuart: Steuart Systems R&D Blog.
APDe-MVS		88.00 6	90.50 6	85.08 5	90.95 8	91.58 7	92.03 1	75.53 13	91.15 5	80.10 16	92.27 5	88.34 11	80.80 5	88.00 14	86.99 19	91.09 6	95.16 6

DeepC-MVS	78.47 2	84.81 27	86.03 29	83.37 20	89.29 33	90.38 12	88.61 28	76.50 1	86.25 24	77.22 25	75.12 40	80.28 46	77.59 23	88.39 10	88.17 6	91.02 7	93.66 19
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
DVP-MVS++		89.14 1	91.86 1	85.97 1	92.55 2	92.38 1	91.69 4	76.31 3	93.31 1	83.11 3	92.44 4	91.18 1	81.17 2	89.55 2	87.93 8	91.01 8	96.21 1

SED-MVS		88.85 2	91.59 3	85.67 2	90.54 16	92.29 3	91.71 3	76.40 2	92.41 3	83.24 2	92.50 3	90.64 4	81.10 3	89.53 3	88.02 7	91.00 9	95.73 3

DPE-MVS		88.63 4	91.29 4	85.53 3	90.87 9	92.20 4	91.98 2	76.00 6	90.55 8	82.09 7	93.85 1	90.75 2	81.25 1	88.62 8	87.59 14	90.96 10	95.48 4
Kehua Chen, Zhenlong Yuan, Tianlu Mao, Zhaoqi Wang: Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization. AAAI2025
SMA-MVS		87.56 7	90.17 7	84.52 10	91.71 3	90.57 10	90.77 9	75.19 14	90.67 7	80.50 15	86.59 18	88.86 8	78.09 17	89.92 1	89.41 1	90.84 11	95.19 5
Yufeng Yin; Xiaoyan Liu; Zichao Zhang: SMA-MVS: Segmentation-Guided Multi-Scale Anchor Deformation Patch Multi-View Stereo. IEEE Transactions on Circuits and Systems for Video Technology
DVP-MVS		88.67 3	91.62 2	85.22 4	90.47 18	92.36 2	90.69 10	76.15 4	93.08 2	82.75 5	92.19 6	90.71 3	80.45 6	89.27 6	87.91 9	90.82 12	95.84 2
Zhenlong Yuan, Jinguo Luo, Fei Shen, Zhaoxin Li, Cong Liu, Tianlu Mao, Zhaoqi Wang: DVP-MVS: Synergize Depth-Edge and Visibility Prior for Multi-View Stereo. AAAI2025
HFP-MVS		86.15 15	87.95 18	84.06 15	90.80 10	89.20 24	89.62 21	74.26 17	87.52 16	80.63 13	86.82 17	84.19 30	78.22 15	87.58 18	87.19 17	90.81 13	93.13 25

ACMMPR		85.52 18	87.53 20	83.17 23	90.13 21	89.27 21	89.30 22	73.97 22	86.89 21	77.14 26	86.09 19	83.18 33	77.74 21	87.42 20	87.20 16	90.77 14	92.63 26

ACMMP		83.42 32	85.27 33	81.26 32	88.47 38	88.49 35	88.31 32	72.09 34	83.42 36	72.77 42	82.65 25	78.22 51	75.18 37	86.24 39	85.76 37	90.74 15	92.13 32
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
PGM-MVS		84.42 29	86.29 28	82.23 27	90.04 23	88.82 28	89.23 24	71.74 37	82.82 38	74.61 34	84.41 24	82.09 36	77.03 29	87.13 26	86.73 25	90.73 16	92.06 33

DeepC-MVS_fast	78.24 3	84.27 30	85.50 32	82.85 24	90.46 19	89.24 22	87.83 34	74.24 18	84.88 27	76.23 29	75.26 39	81.05 44	77.62 22	88.02 13	87.62 13	90.69 17	92.41 29
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
SD-MVS		86.96 10	89.45 9	84.05 16	90.13 21	89.23 23	89.77 19	74.59 15	89.17 11	80.70 12	89.93 12	89.67 5	78.47 13	87.57 19	86.79 23	90.67 18	93.76 17
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
DROMVSNet		79.44 50	81.35 47	77.22 55	82.95 65	84.67 63	81.31 62	63.65 92	72.47 68	68.75 57	73.15 48	78.33 50	75.99 33	86.06 41	83.96 50	90.67 18	90.79 43

TSAR-MVS + MP.		86.88 11	89.23 10	84.14 14	89.78 27	88.67 32	90.59 11	73.46 28	88.99 12	80.52 14	91.26 7	88.65 9	79.91 8	86.96 31	86.22 33	90.59 20	93.83 15
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
MVS_0304		81.73 40	83.86 37	79.26 43	86.22 51	89.18 25	86.41 39	67.15 67	75.28 56	70.75 54	74.59 42	83.49 32	74.42 40	87.05 29	86.34 30	90.58 21	91.08 41

APD-MVS		86.84 12	88.91 14	84.41 11	90.66 12	90.10 13	90.78 8	75.64 10	87.38 18	78.72 20	90.68 11	86.82 17	80.15 7	87.13 26	86.45 29	90.51 22	93.83 15
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
CP-MVS		84.74 28	86.43 27	82.77 25	89.48 31	88.13 40	88.64 27	73.93 23	84.92 26	76.77 27	81.94 27	83.50 31	77.29 27	86.92 32	86.49 28	90.49 23	93.14 24

XVS					86.63 47	88.68 29	85.00 48			71.81 47		81.92 38				90.47 24

X-MVStestdata					86.63 47	88.68 29	85.00 48			71.81 47		81.92 38				90.47 24

X-MVS		83.23 34	85.20 34	80.92 35	89.71 28	88.68 29	88.21 33	73.60 25	82.57 39	71.81 47	77.07 34	81.92 38	71.72 61	86.98 30	86.86 21	90.47 24	92.36 30

NCCC		85.34 21	86.59 25	83.88 17	91.48 4	88.88 26	89.79 18	75.54 12	86.67 22	77.94 24	76.55 36	84.99 26	78.07 18	88.04 12	87.68 12	90.46 27	93.31 22

CNVR-MVS		86.36 14	88.19 17	84.23 13	91.33 5	89.84 15	90.34 12	75.56 11	87.36 19	78.97 19	81.19 29	86.76 18	78.74 12	89.30 5	88.58 2	90.45 28	94.33 10

MP-MVS		85.50 19	87.40 21	83.28 21	90.65 13	89.51 20	89.16 25	74.11 20	83.70 35	78.06 23	85.54 21	84.89 28	77.31 25	87.40 23	87.14 18	90.41 29	93.65 20
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
3Dnovator+	75.73 4	82.40 36	82.76 41	81.97 30	88.02 39	89.67 18	86.60 38	71.48 38	81.28 44	78.18 22	64.78 86	77.96 53	77.13 28	87.32 24	86.83 22	90.41 29	91.48 37

ACMMP_NAP		86.52 13	89.01 11	83.62 18	90.28 20	90.09 14	90.32 14	74.05 21	88.32 15	79.74 17	87.04 16	85.59 24	76.97 30	89.35 4	88.44 4	90.35 31	94.27 11

zzz-MVS		85.71 17	86.88 23	84.34 12	90.54 16	87.11 45	89.77 19	74.17 19	88.54 14	83.08 4	78.60 33	86.10 20	78.11 16	87.80 17	87.46 15	90.35 31	92.56 27

CS-MVS-test		78.10 61	79.79 60	76.13 60	80.59 85	81.68 89	81.31 62	63.65 92	68.34 76	64.91 74	72.52 51	76.25 58	75.99 33	86.06 41	83.55 56	90.31 33	90.16 50

CDPH-MVS		82.64 35	85.03 35	79.86 40	89.41 32	88.31 37	88.32 31	71.84 36	80.11 46	67.47 64	82.09 26	81.44 42	71.85 59	85.89 43	86.15 34	90.24 34	91.25 39

CS-MVS		77.88 62	80.66 51	74.64 73	79.87 94	82.07 86	78.63 86	59.93 146	72.57 67	63.22 83	73.66 47	77.72 54	75.97 35	85.12 49	83.58 54	90.23 35	89.93 51

LGP-MVS_train		79.83 45	81.22 49	78.22 51	86.28 50	85.36 59	86.76 37	69.59 48	77.34 51	65.14 72	75.68 38	70.79 79	71.37 65	84.60 53	84.01 48	90.18 36	90.74 44

HPM-MVS++		87.09 9	88.92 13	84.95 6	92.61 1	87.91 41	90.23 16	76.06 5	88.85 13	81.20 11	87.33 14	87.93 12	79.47 9	88.59 9	88.23 5	90.15 37	93.60 21

3Dnovator	73.76 5	79.75 47	80.52 54	78.84 46	84.94 61	87.35 42	84.43 53	65.54 78	78.29 50	73.97 36	63.00 94	75.62 61	74.07 42	85.00 50	85.34 41	90.11 38	89.04 57

DPM-MVS		83.30 33	84.33 36	82.11 28	89.56 29	88.49 35	90.33 13	73.24 29	83.85 34	76.46 28	72.43 52	82.65 34	73.02 50	86.37 37	86.91 20	90.03 39	89.62 55

ETV-MVS		77.32 64	78.81 62	75.58 63	82.24 72	83.64 72	79.98 70	64.02 90	69.64 74	63.90 78	70.89 60	69.94 85	73.41 46	85.39 47	83.91 51	89.92 40	88.31 62

ACMP	73.23 7	79.79 46	80.53 53	78.94 45	85.61 54	85.68 54	85.61 44	69.59 48	77.33 52	71.00 53	74.45 43	69.16 91	71.88 57	83.15 67	83.37 57	89.92 40	90.57 47
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
train_agg		84.86 26	87.21 22	82.11 28	90.59 14	85.47 56	89.81 17	73.55 27	83.95 33	73.30 39	89.84 13	87.23 15	75.61 36	86.47 35	85.46 40	89.78 42	92.06 33

TSAR-MVS + GP.		83.69 31	86.58 26	80.32 37	85.14 56	86.96 46	84.91 51	70.25 43	84.71 30	73.91 37	85.16 22	85.63 23	77.92 19	85.44 44	85.71 38	89.77 43	92.45 28

ACMM	72.26 8	78.86 57	78.13 65	79.71 41	86.89 46	83.40 74	86.02 41	70.50 41	75.28 56	71.49 51	63.01 93	69.26 90	73.57 45	84.11 58	83.98 49	89.76 44	87.84 66
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
OPM-MVS		79.68 49	79.28 61	80.15 39	87.99 40	86.77 48	88.52 30	72.72 31	64.55 99	67.65 63	67.87 75	74.33 65	74.31 41	86.37 37	85.25 42	89.73 45	89.81 53
Ray L. Khuboni, Hongjun Xu: Octagram Propagation Matching for Multi-Scale View Stereopsis (OPM-MVS).
MSP-MVS		88.09 5	90.84 5	84.88 7	90.00 24	91.80 6	91.63 5	75.80 7	91.99 4	81.23 10	92.54 2	89.18 6	80.89 4	87.99 15	87.91 9	89.70 46	94.51 7
Zhenlong Yuan, Cong Liu, Fei Shen, Zhaoxin Li, Jingguo luo, Tianlu Mao and Zhaoqi Wang: MSP-MVS: Multi-granularity Segmentation Prior Guided Multi-View Stereo. AAAI2025
CANet		81.62 41	83.41 38	79.53 42	87.06 44	88.59 33	85.47 46	67.96 60	76.59 54	74.05 35	74.69 41	81.98 37	72.98 51	86.14 40	85.47 39	89.68 47	90.42 48

abl_6				79.05 44	87.27 43	88.85 27	83.62 57	68.25 56	81.68 42	72.94 41	73.79 46	84.45 29	72.55 53			89.66 48	90.64 45

MVS_111021_HR		80.13 44	81.46 46	78.58 48	85.77 53	85.17 60	83.45 58	69.28 51	74.08 62	70.31 55	74.31 44	75.26 62	73.13 48	86.46 36	85.15 43	89.53 49	89.81 53

IS_MVSNet		73.33 83	77.34 74	68.65 116	81.29 77	83.47 73	74.45 125	63.58 95	65.75 91	48.49 151	67.11 79	70.61 80	54.63 168	84.51 55	83.58 54	89.48 50	86.34 79

DELS-MVS		79.15 55	81.07 50	76.91 57	83.54 63	87.31 43	84.45 52	64.92 83	69.98 70	69.34 56	71.62 56	76.26 57	69.84 70	86.57 34	85.90 36	89.39 51	89.88 52
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
UniMVSNet_NR-MVSNet		70.59 105	72.19 104	68.72 114	77.72 111	80.72 103	73.81 140	69.65 47	61.99 119	43.23 177	60.54 102	57.50 138	58.57 137	79.56 114	81.07 76	89.34 52	83.97 106

EIA-MVS		75.64 72	76.60 79	74.53 74	82.43 70	83.84 68	78.32 91	62.28 120	65.96 89	63.28 82	68.95 67	67.54 101	71.61 63	82.55 72	81.63 70	89.24 53	85.72 83

PHI-MVS		82.36 37	85.89 30	78.24 50	86.40 49	89.52 19	85.52 45	69.52 50	82.38 41	65.67 70	81.35 28	82.36 35	73.07 49	87.31 25	86.76 24	89.24 53	91.56 36

canonicalmvs		79.16 54	82.37 44	75.41 64	82.33 71	86.38 52	80.80 66	63.18 98	82.90 37	67.34 65	72.79 50	76.07 59	69.62 71	83.46 66	84.41 47	89.20 55	90.60 46

MSLP-MVS++		82.09 38	82.66 42	81.42 31	87.03 45	87.22 44	85.82 43	70.04 44	80.30 45	78.66 21	68.67 71	81.04 45	77.81 20	85.19 48	84.88 45	89.19 56	91.31 38

HQP-MVS		81.19 42	83.27 39	78.76 47	87.40 42	85.45 57	86.95 36	70.47 42	81.31 43	66.91 67	79.24 31	76.63 56	71.67 62	84.43 56	83.78 52	89.19 56	92.05 35

EPP-MVSNet		74.00 81	77.41 72	70.02 101	80.53 87	83.91 67	74.99 120	62.68 112	65.06 94	49.77 146	68.68 70	72.09 73	63.06 107	82.49 74	80.73 81	89.12 58	88.91 58

NR-MVSNet		68.79 126	70.56 114	66.71 145	77.48 114	79.54 112	73.52 144	69.20 52	61.20 127	39.76 184	58.52 114	50.11 189	51.37 177	80.26 105	80.71 86	88.97 59	83.59 112

PVSNet_Blended_VisFu		76.57 67	77.90 66	75.02 67	80.56 86	86.58 50	79.24 80	66.18 72	64.81 96	68.18 61	65.61 80	71.45 74	67.05 82	84.16 57	81.80 68	88.90 60	90.92 42

TranMVSNet+NR-MVSNet		69.25 121	70.81 113	67.43 129	77.23 116	79.46 114	73.48 145	69.66 46	60.43 133	39.56 185	58.82 113	53.48 165	55.74 162	79.59 112	81.21 74	88.89 61	82.70 116

QAPM		78.47 58	80.22 57	76.43 59	85.03 58	86.75 49	80.62 68	66.00 75	73.77 63	65.35 71	65.54 82	78.02 52	72.69 52	83.71 61	83.36 58	88.87 62	90.41 49

casdiffmvs		76.76 66	78.46 64	74.77 70	80.32 90	83.73 71	80.65 67	63.24 97	73.58 64	66.11 69	69.39 66	74.09 66	69.49 73	82.52 73	79.35 110	88.84 63	86.52 77

CPTT-MVS		81.77 39	83.10 40	80.21 38	85.93 52	86.45 51	87.72 35	70.98 40	82.54 40	71.53 50	74.23 45	81.49 41	76.31 32	82.85 70	81.87 67	88.79 64	92.26 31

Effi-MVS+		75.28 74	76.20 80	74.20 76	81.15 79	83.24 77	81.11 64	63.13 100	66.37 85	60.27 89	64.30 90	68.88 95	70.93 68	81.56 79	81.69 69	88.61 65	87.35 69

UniMVSNet (Re)		69.53 117	71.90 107	66.76 143	76.42 120	80.93 99	72.59 150	68.03 59	61.75 122	41.68 182	58.34 120	57.23 140	53.27 173	79.53 115	80.62 90	88.57 66	84.90 98

PCF-MVS	73.28 6	79.42 51	80.41 55	78.26 49	84.88 62	88.17 38	86.08 40	69.85 45	75.23 58	68.43 59	68.03 74	78.38 49	71.76 60	81.26 88	80.65 89	88.56 67	91.18 40
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
test_part1		74.24 78	73.44 92	75.18 66	82.02 74	82.34 85	83.88 55	62.40 118	60.93 129	68.68 58	49.25 181	69.71 87	65.73 97	81.26 88	81.98 66	88.35 68	88.60 61

GeoE		74.23 79	74.84 85	73.52 78	80.42 89	81.46 93	79.77 74	61.06 130	67.23 82	63.67 79	59.56 109	68.74 97	67.90 80	80.25 106	79.37 109	88.31 69	87.26 72

test2506		71.72 94	72.95 98	70.29 96	81.49 75	83.27 75	75.74 109	67.59 64	68.19 78	49.81 145	61.15 97	49.73 191	58.82 135	84.76 51	82.94 59	88.27 70	80.63 137

ECVR-MVS		72.20 90	73.91 89	70.20 98	81.49 75	83.27 75	75.74 109	67.59 64	68.19 78	49.31 149	55.77 131	62.00 118	58.82 135	84.76 51	82.94 59	88.27 70	80.41 141

MAR-MVS		79.21 53	80.32 56	77.92 52	87.46 41	88.15 39	83.95 54	67.48 66	74.28 60	68.25 60	64.70 87	77.04 55	72.17 55	85.42 45	85.00 44	88.22 72	87.62 68
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
Fast-Effi-MVS+		73.11 85	73.66 90	72.48 82	77.72 111	80.88 102	78.55 88	58.83 159	65.19 93	60.36 88	59.98 106	62.42 117	71.22 66	81.66 76	80.61 91	88.20 73	84.88 99

ET-MVSNet_ETH3D		72.46 89	74.19 87	70.44 94	62.50 200	81.17 97	79.90 73	62.46 117	64.52 100	57.52 102	71.49 58	59.15 131	72.08 56	78.61 126	81.11 75	88.16 74	83.29 114

OMC-MVS		80.26 43	82.59 43	77.54 53	83.04 64	85.54 55	83.25 59	65.05 82	87.32 20	72.42 43	72.04 54	78.97 48	73.30 47	83.86 59	81.60 71	88.15 75	88.83 59

AdaColmap		79.74 48	78.62 63	81.05 34	89.23 34	86.06 53	84.95 50	71.96 35	79.39 49	75.51 32	63.16 92	68.84 96	76.51 31	83.55 63	82.85 61	88.13 76	86.46 78

test1111		71.56 96	73.44 92	69.38 109	81.16 78	82.95 80	74.99 120	67.68 62	66.89 83	46.33 165	55.19 137	60.91 122	57.99 143	84.59 54	82.70 62	88.12 77	80.85 134

OpenMVS	70.44 10	76.15 70	76.82 78	75.37 65	85.01 59	84.79 62	78.99 84	62.07 121	71.27 69	67.88 62	57.91 122	72.36 72	70.15 69	82.23 75	81.41 72	88.12 77	87.78 67

UA-Net		74.47 77	77.80 67	70.59 93	85.33 55	85.40 58	73.54 143	65.98 76	60.65 131	56.00 110	72.11 53	79.15 47	54.63 168	83.13 68	82.25 64	88.04 79	81.92 126

DU-MVS		69.63 116	70.91 112	68.13 120	75.99 122	79.54 112	73.81 140	69.20 52	61.20 127	43.23 177	58.52 114	53.50 163	58.57 137	79.22 118	80.45 92	87.97 80	83.97 106

FC-MVSNet-train		72.60 88	75.07 84	69.71 104	81.10 81	78.79 122	73.74 142	65.23 81	66.10 88	53.34 125	70.36 62	63.40 114	56.92 153	81.44 81	80.96 78	87.93 81	84.46 104

IB-MVS	66.94 12	71.21 101	71.66 109	70.68 90	79.18 99	82.83 82	72.61 149	61.77 125	59.66 136	63.44 81	53.26 153	59.65 129	59.16 134	76.78 146	82.11 65	87.90 82	87.33 70
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
PVSNet_BlendedMVS		76.21 68	77.52 70	74.69 71	79.46 97	83.79 69	77.50 98	64.34 88	69.88 71	71.88 45	68.54 72	70.42 81	67.05 82	83.48 64	79.63 101	87.89 83	86.87 74

PVSNet_Blended		76.21 68	77.52 70	74.69 71	79.46 97	83.79 69	77.50 98	64.34 88	69.88 71	71.88 45	68.54 72	70.42 81	67.05 82	83.48 64	79.63 101	87.89 83	86.87 74

DeepPCF-MVS	79.04 1	85.30 22	88.93 12	81.06 33	88.77 37	90.48 11	85.46 47	73.08 30	90.97 6	73.77 38	84.81 23	85.95 21	77.43 24	88.22 11	87.73 11	87.85 85	94.34 9

thisisatest0530		71.48 98	73.01 97	69.70 105	73.83 146	78.62 124	74.53 124	59.12 153	64.13 102	58.63 95	64.60 88	58.63 133	64.27 100	80.28 104	80.17 97	87.82 86	84.64 102

TAPA-MVS	71.42 9	77.69 63	80.05 58	74.94 68	80.68 84	84.52 64	81.36 61	63.14 99	84.77 28	64.82 75	68.72 69	75.91 60	71.86 58	81.62 77	79.55 105	87.80 87	85.24 92
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
tttt0517		71.41 99	72.95 98	69.60 106	73.70 148	78.70 123	74.42 128	59.12 153	63.89 106	58.35 98	64.56 89	58.39 135	64.27 100	80.29 103	80.17 97	87.74 88	84.69 101

MVS_Test		75.37 73	77.13 76	73.31 80	79.07 100	81.32 95	79.98 70	60.12 143	69.72 73	64.11 77	70.53 61	73.22 68	68.90 75	80.14 108	79.48 107	87.67 89	85.50 87

DI_MVS_plusplus_trai		75.13 75	76.12 81	73.96 77	78.18 105	81.55 90	80.97 65	62.54 114	68.59 75	65.13 73	61.43 96	74.81 63	69.32 74	81.01 94	79.59 103	87.64 90	85.89 81

MVSTER		72.06 91	74.24 86	69.51 107	70.39 176	75.97 152	76.91 104	57.36 166	64.64 98	61.39 87	68.86 68	63.76 112	63.46 104	81.44 81	79.70 100	87.56 91	85.31 91

TSAR-MVS + ACMM		85.10 25	88.81 15	80.77 36	89.55 30	88.53 34	88.59 29	72.55 32	87.39 17	71.90 44	90.95 10	87.55 13	74.57 38	87.08 28	86.54 27	87.47 92	93.67 18

CLD-MVS		79.35 52	81.23 48	77.16 56	85.01 59	86.92 47	85.87 42	60.89 132	80.07 48	75.35 33	72.96 49	73.21 69	68.43 79	85.41 46	84.63 46	87.41 93	85.44 89
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
Effi-MVS+-dtu		71.82 93	71.86 108	71.78 84	78.77 101	80.47 105	78.55 88	61.67 128	60.68 130	55.49 111	58.48 116	65.48 107	68.85 76	76.92 143	75.55 155	87.35 94	85.46 88

WR-MVS		63.03 166	67.40 151	57.92 185	75.14 131	77.60 139	60.56 198	66.10 73	54.11 175	23.88 207	53.94 147	53.58 161	34.50 203	73.93 162	77.71 128	87.35 94	80.94 133

v144192		69.34 120	68.68 138	70.12 99	74.06 142	80.54 104	78.08 94	60.54 136	54.99 168	54.13 118	52.92 160	52.80 174	66.73 89	77.13 141	76.72 144	87.15 96	85.63 84

EPNet		79.08 56	80.62 52	77.28 54	88.90 36	83.17 79	83.65 56	72.41 33	74.41 59	67.15 66	76.78 35	74.37 64	64.43 99	83.70 62	83.69 53	87.15 96	88.19 63
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
EG-PatchMatch MVS		67.24 147	66.94 154	67.60 127	78.73 102	81.35 94	73.28 147	59.49 149	46.89 200	51.42 137	43.65 193	53.49 164	55.50 165	81.38 83	80.66 88	87.15 96	81.17 132

v1144		69.93 114	69.36 128	70.61 92	74.89 134	80.93 99	79.11 82	60.64 134	55.97 160	55.31 113	53.85 148	54.14 156	66.54 91	78.10 131	77.44 134	87.14 99	85.09 94

anonymousdsp		65.28 156	67.98 145	62.13 166	58.73 208	73.98 166	67.10 172	50.69 192	48.41 196	47.66 159	54.27 142	52.75 175	61.45 125	76.71 147	80.20 95	87.13 100	89.53 56

PLC	68.99 11	75.68 71	75.31 83	76.12 62	82.94 66	81.26 96	79.94 72	66.10 73	77.15 53	66.86 68	59.13 112	68.53 98	73.73 44	80.38 101	79.04 111	87.13 100	81.68 128
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
v10		70.22 110	69.76 123	70.74 88	74.79 135	80.30 109	79.22 81	59.81 147	57.71 147	56.58 108	54.22 146	55.31 149	66.95 85	78.28 129	77.47 133	87.12 102	85.07 95

v1192		69.50 118	68.83 134	70.29 96	74.49 138	80.92 101	78.55 88	60.54 136	55.04 166	54.21 116	52.79 162	52.33 176	66.92 86	77.88 133	77.35 137	87.04 103	85.51 86

v1921920		69.03 123	68.32 142	69.86 102	74.03 143	80.37 106	77.55 96	60.25 140	54.62 170	53.59 124	52.36 166	51.50 182	66.75 88	77.17 140	76.69 146	86.96 104	85.56 85

v2v482		70.05 113	69.46 127	70.74 88	74.62 137	80.32 108	79.00 83	60.62 135	57.41 149	56.89 105	55.43 136	55.14 151	66.39 93	77.25 139	77.14 139	86.90 105	83.57 113

Vis-MVSNet		72.77 87	77.20 75	67.59 128	74.19 141	84.01 66	76.61 108	61.69 126	60.62 132	50.61 141	70.25 63	71.31 77	55.57 164	83.85 60	82.28 63	86.90 105	88.08 64
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
PEN-MVS		62.96 167	65.77 161	59.70 178	73.98 144	75.45 156	63.39 191	67.61 63	52.49 181	25.49 206	53.39 150	49.12 193	40.85 196	71.94 174	77.26 138	86.86 107	80.72 136

v8		70.23 109	69.86 121	70.67 91	74.69 136	79.82 111	78.79 85	59.18 152	58.80 140	58.20 99	55.00 138	57.33 139	66.31 94	77.51 136	76.71 145	86.82 108	83.88 109

ACMH+	66.54 13	71.36 100	70.09 118	72.85 81	82.59 68	81.13 98	78.56 87	68.04 58	61.55 123	52.52 132	51.50 170	54.14 156	68.56 78	78.85 123	79.50 106	86.82 108	83.94 108

v1240		68.64 128	67.89 147	69.51 107	73.89 145	80.26 110	76.73 106	59.97 145	53.43 178	53.08 127	51.82 169	50.84 185	66.62 90	76.79 145	76.77 143	86.78 110	85.34 90

DCV-MVSNet		73.65 82	75.78 82	71.16 87	80.19 91	79.27 116	77.45 100	61.68 127	66.73 84	58.72 94	65.31 83	69.96 84	62.19 112	81.29 87	80.97 77	86.74 111	86.91 73

GBi-Net		70.78 102	73.37 95	67.76 121	72.95 153	78.00 129	75.15 115	62.72 107	64.13 102	51.44 134	58.37 117	69.02 92	57.59 145	81.33 84	80.72 82	86.70 112	82.02 120

test1		70.78 102	73.37 95	67.76 121	72.95 153	78.00 129	75.15 115	62.72 107	64.13 102	51.44 134	58.37 117	69.02 92	57.59 145	81.33 84	80.72 82	86.70 112	82.02 120

FMVSNet2		70.39 108	72.67 102	67.72 124	72.95 153	78.00 129	75.15 115	62.69 111	63.29 110	51.25 138	55.64 132	68.49 99	57.59 145	80.91 95	80.35 94	86.70 112	82.02 120

v7n		67.05 149	66.94 154	67.17 135	72.35 158	78.97 117	73.26 148	58.88 158	51.16 189	50.90 139	48.21 184	50.11 189	60.96 126	77.70 134	77.38 135	86.68 115	85.05 96

WR-MVS_H		61.83 181	65.87 160	57.12 188	71.72 163	76.87 143	61.45 196	66.19 71	51.97 186	22.92 211	53.13 157	52.30 178	33.80 204	71.03 181	75.00 158	86.65 116	80.78 135

MSDG		71.52 97	69.87 120	73.44 79	82.21 73	79.35 115	79.52 77	64.59 85	66.15 87	61.87 84	53.21 155	56.09 146	65.85 96	78.94 122	78.50 118	86.60 117	76.85 165

FMVSNet3		70.49 106	72.90 100	67.67 126	72.88 156	77.98 132	74.96 122	62.72 107	64.13 102	51.44 134	58.37 117	69.02 92	57.43 148	79.43 116	79.57 104	86.59 118	81.81 127

MVS_111021_LR		78.13 60	79.85 59	76.13 60	81.12 80	81.50 92	80.28 69	65.25 80	76.09 55	71.32 52	76.49 37	72.87 71	72.21 54	82.79 71	81.29 73	86.59 118	87.91 65

DTE-MVSNet		61.85 179	64.96 170	58.22 184	74.32 140	74.39 165	61.01 197	67.85 61	51.76 188	21.91 214	53.28 152	48.17 194	37.74 200	72.22 171	76.44 148	86.52 120	78.49 154

baseline2		69.69 115	70.27 117	69.01 112	75.72 126	77.13 142	73.82 139	58.94 157	61.35 125	57.09 104	61.68 95	57.17 141	61.99 116	78.10 131	76.58 147	86.48 121	79.85 145

thisisatest0515		67.40 145	68.78 135	65.80 148	70.02 178	75.24 159	69.36 162	57.37 165	54.94 169	53.67 123	55.53 135	54.85 152	58.00 142	78.19 130	78.91 114	86.39 122	83.78 110

FMVSNet1		68.84 125	70.47 116	66.94 140	71.35 170	77.68 137	74.71 123	62.35 119	56.93 151	49.94 144	50.01 176	64.59 109	57.07 150	81.33 84	80.72 82	86.25 123	82.00 123

UGNet		72.78 86	77.67 68	67.07 138	71.65 165	83.24 77	75.20 114	63.62 94	64.93 95	56.72 106	71.82 55	73.30 67	49.02 181	81.02 93	80.70 87	86.22 124	88.67 60
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
Anonymous20231211		71.90 92	72.48 103	71.21 86	80.14 92	81.53 91	76.92 103	62.89 103	64.46 101	58.94 91	43.80 192	70.98 78	62.22 111	80.70 96	80.19 96	86.18 125	85.73 82

tfpn200view9		68.11 131	68.72 137	67.40 130	77.83 109	78.93 118	74.28 130	62.81 104	56.64 153	46.82 161	52.65 163	53.47 166	56.59 154	80.41 98	78.43 119	86.11 126	80.52 139

CANet_DTU		73.29 84	76.96 77	69.00 113	77.04 117	82.06 87	79.49 78	56.30 169	67.85 80	53.29 126	71.12 59	70.37 83	61.81 121	81.59 78	80.96 78	86.09 127	84.73 100

CP-MVSNet		62.68 169	65.49 164	59.40 181	71.84 161	75.34 157	62.87 193	67.04 68	52.64 180	27.19 204	53.38 151	48.15 195	41.40 194	71.26 177	75.68 153	86.07 128	82.00 123

ACMH	65.37 14	70.71 104	70.00 119	71.54 85	82.51 69	82.47 84	77.78 95	68.13 57	56.19 158	46.06 168	54.30 141	51.20 183	68.68 77	80.66 97	80.72 82	86.07 128	84.45 105
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
thres600view7		67.68 139	68.43 141	66.80 142	77.90 106	78.86 120	73.84 138	62.75 105	56.07 159	44.70 175	52.85 161	52.81 173	55.58 163	80.41 98	77.77 127	86.05 130	80.28 142

thres200		67.98 133	68.55 140	67.30 133	77.89 108	78.86 120	74.18 134	62.75 105	56.35 156	46.48 164	52.98 159	53.54 162	56.46 155	80.41 98	77.97 125	86.05 130	79.78 147

CNLPA		77.20 65	77.54 69	76.80 58	82.63 67	84.31 65	79.77 74	64.64 84	85.17 25	73.18 40	56.37 129	69.81 86	74.53 39	81.12 92	78.69 116	86.04 132	87.29 71

TSAR-MVS + COLMAP		78.34 59	81.64 45	74.48 75	80.13 93	85.01 61	81.73 60	65.93 77	84.75 29	61.68 85	85.79 20	66.27 105	71.39 64	82.91 69	80.78 80	86.01 133	85.98 80

LS3D		74.08 80	73.39 94	74.88 69	85.05 57	82.62 83	79.71 76	68.66 54	72.82 65	58.80 93	57.61 123	61.31 121	71.07 67	80.32 102	78.87 115	86.00 134	80.18 143

Anonymous202405211			72.16 106		80.85 83	81.85 88	76.88 105	65.40 79	62.89 114		46.35 188	67.99 100	62.05 114	81.15 91	80.38 93	85.97 135	84.50 103

PS-CasMVS		62.38 175	65.06 167	59.25 182	71.73 162	75.21 161	62.77 194	66.99 69	51.94 187	26.96 205	52.00 168	47.52 198	41.06 195	71.16 180	75.60 154	85.97 135	81.97 125

thres400		67.95 134	68.62 139	67.17 135	77.90 106	78.59 125	74.27 131	62.72 107	56.34 157	45.77 170	53.00 158	53.35 169	56.46 155	80.21 107	78.43 119	85.91 137	80.43 140

thres100view900		67.60 143	68.02 144	67.12 137	77.83 109	77.75 136	73.90 137	62.52 115	56.64 153	46.82 161	52.65 163	53.47 166	55.92 159	78.77 124	77.62 130	85.72 138	79.23 150

Vis-MVSNet (Re-imp)		67.83 137	73.52 91	61.19 170	78.37 104	76.72 146	66.80 175	62.96 101	65.50 92	34.17 196	67.19 78	69.68 88	39.20 199	79.39 117	79.44 108	85.68 139	76.73 166

baseline1		70.10 112	72.17 105	67.69 125	79.74 95	76.80 144	73.91 136	64.38 87	62.74 115	48.30 153	64.94 84	64.08 111	54.17 170	81.46 80	78.92 113	85.66 140	76.22 167

V42		68.76 127	69.63 124	67.74 123	64.93 196	78.01 128	78.30 92	56.48 168	58.65 141	56.30 109	54.26 144	57.03 142	64.85 98	77.47 137	77.01 141	85.60 141	84.96 97

TransMVSNet (Re)		64.74 159	65.66 162	63.66 161	77.40 115	75.33 158	69.86 158	62.67 113	47.63 198	41.21 183	50.01 176	52.33 176	45.31 187	79.57 113	77.69 129	85.49 142	77.07 164

IterMVS-LS		71.69 95	72.82 101	70.37 95	77.54 113	76.34 149	75.13 118	60.46 138	61.53 124	57.57 101	64.89 85	67.33 102	66.04 95	77.09 142	77.37 136	85.48 143	85.18 93
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
UniMVSNet_ETH3D		67.18 148	67.03 153	67.36 131	74.44 139	78.12 127	74.07 135	66.38 70	52.22 183	46.87 160	48.64 182	51.84 180	56.96 151	77.29 138	78.53 117	85.42 144	82.59 117

Baseline_NR-MVSNet		67.53 144	68.77 136	66.09 147	75.99 122	74.75 163	72.43 151	68.41 55	61.33 126	38.33 189	51.31 171	54.13 158	56.03 158	79.22 118	78.19 122	85.37 145	82.45 118

Fast-Effi-MVS+-dtu		68.34 129	69.47 126	67.01 139	75.15 130	77.97 134	77.12 102	55.40 171	57.87 142	46.68 163	56.17 130	60.39 123	62.36 110	76.32 150	76.25 151	85.35 146	81.34 130

diffmvs		74.86 76	77.37 73	71.93 83	75.62 127	80.35 107	79.42 79	60.15 142	72.81 66	64.63 76	71.51 57	73.11 70	66.53 92	79.02 121	77.98 124	85.25 147	86.83 76

GA-MVS		68.14 130	69.17 131	66.93 141	73.77 147	78.50 126	74.45 125	58.28 161	55.11 165	48.44 152	60.08 104	53.99 159	61.50 123	78.43 128	77.57 131	85.13 148	80.54 138

v148		67.85 136	67.53 148	68.23 118	73.25 151	77.57 140	74.26 132	57.36 166	55.70 161	57.45 103	53.53 149	55.42 148	61.96 117	75.23 154	73.92 163	85.08 149	81.32 131

HyFIR lowres test		69.47 119	68.94 133	70.09 100	76.77 119	82.93 81	76.63 107	60.17 141	59.00 139	54.03 119	40.54 201	65.23 108	67.89 81	76.54 149	78.30 121	85.03 150	80.07 144

gg-mvs-nofinetune		62.55 170	65.05 168	59.62 179	78.72 103	77.61 138	70.83 157	53.63 172	39.71 212	22.04 213	36.36 205	64.32 110	47.53 183	81.16 90	79.03 112	85.00 151	77.17 162

COLMAP_ROB	62.73 15	67.66 140	66.76 156	68.70 115	80.49 88	77.98 132	75.29 113	62.95 102	63.62 108	49.96 143	47.32 187	50.72 186	58.57 137	76.87 144	75.50 156	84.94 152	75.33 176
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
tfpnnormal		64.27 162	63.64 178	65.02 151	75.84 125	75.61 155	71.24 156	62.52 115	47.79 197	42.97 179	42.65 195	44.49 205	52.66 175	78.77 124	76.86 142	84.88 153	79.29 149

pm-mvs1		65.62 153	67.42 150	63.53 162	73.66 149	76.39 148	69.66 159	60.87 133	49.73 193	43.97 176	51.24 172	57.00 143	48.16 182	79.89 109	77.84 126	84.85 154	79.82 146

gm-plane-assit		57.00 194	57.62 201	56.28 191	76.10 121	62.43 207	47.62 215	46.57 206	33.84 216	23.24 209	37.52 202	40.19 212	59.61 133	79.81 110	77.55 132	84.55 155	72.03 186

USDC		67.36 146	67.90 146	66.74 144	71.72 163	75.23 160	71.58 153	60.28 139	67.45 81	50.54 142	60.93 98	45.20 204	62.08 113	76.56 148	74.50 161	84.25 156	75.38 175

MS-PatchMatch		70.17 111	70.49 115	69.79 103	80.98 82	77.97 134	77.51 97	58.95 156	62.33 117	55.22 114	53.14 156	65.90 106	62.03 115	79.08 120	77.11 140	84.08 157	77.91 157

TDRefinement		66.09 152	65.03 169	67.31 132	69.73 180	76.75 145	75.33 111	64.55 86	60.28 134	49.72 147	45.63 190	42.83 207	60.46 131	75.75 151	75.95 152	84.08 157	78.04 156

pmmvs4		67.89 135	67.39 152	68.48 117	71.60 167	73.57 167	74.45 125	60.98 131	64.65 97	57.97 100	54.95 139	51.73 181	61.88 118	73.78 163	75.11 157	83.99 159	77.91 157

pmmvs5		62.37 176	64.04 175	60.42 173	65.03 194	71.67 174	67.17 171	52.70 182	50.30 190	44.80 173	54.23 145	51.19 184	49.37 180	72.88 166	73.48 167	83.45 160	74.55 179

pmmvs-eth3d		63.52 165	62.44 187	64.77 153	66.82 191	70.12 179	69.41 161	59.48 150	54.34 174	52.71 128	46.24 189	44.35 206	56.93 152	72.37 167	73.77 165	83.30 161	75.91 169

pmmvs6		62.41 173	62.88 181	61.87 167	71.38 169	75.18 162	67.76 168	59.45 151	41.64 208	42.52 181	37.33 203	52.91 172	46.87 184	77.67 135	76.26 150	83.23 162	79.18 151

CDS-MVSNet		67.65 141	69.83 122	65.09 150	75.39 129	76.55 147	74.42 128	63.75 91	53.55 176	49.37 148	59.41 110	62.45 116	44.44 188	79.71 111	79.82 99	83.17 163	77.36 161
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
PMMVS		65.06 157	69.17 131	60.26 175	55.25 214	63.43 201	66.71 176	43.01 210	62.41 116	50.64 140	69.44 65	67.04 103	63.29 105	74.36 160	73.54 166	82.68 164	73.99 183

SixPastTwentyTwo		61.84 180	62.45 186	61.12 171	69.20 184	72.20 171	62.03 195	57.40 164	46.54 201	38.03 191	57.14 127	41.72 209	58.12 141	69.67 191	71.58 174	81.94 165	78.30 155

IterMVS		66.36 151	68.30 143	64.10 157	69.48 183	74.61 164	73.41 146	50.79 191	57.30 150	48.28 154	60.64 101	59.92 128	60.85 130	74.14 161	72.66 170	81.80 166	78.82 153
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
PatchMatch-RL		67.78 138	66.65 157	69.10 111	73.01 152	72.69 170	68.49 165	61.85 124	62.93 113	60.20 90	56.83 128	50.42 187	69.52 72	75.62 152	74.46 162	81.51 167	73.62 184

TinyColmap		62.84 168	61.03 193	64.96 152	69.61 181	71.69 173	68.48 166	59.76 148	55.41 162	47.69 158	47.33 186	34.20 216	62.76 109	74.52 158	72.59 171	81.44 168	71.47 187

EPNet_dtu		68.08 132	71.00 111	64.67 154	79.64 96	68.62 185	75.05 119	63.30 96	66.36 86	45.27 172	67.40 77	66.84 104	43.64 190	75.37 153	74.98 159	81.15 169	77.44 160
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
CR-MVSNet		64.83 158	65.54 163	64.01 159	70.64 175	69.41 180	65.97 180	52.74 180	57.81 144	52.65 129	54.27 142	56.31 145	60.92 127	72.20 172	73.09 168	81.12 170	75.69 172

RPMNet		61.71 183	62.88 181	60.34 174	69.51 182	69.41 180	63.48 190	49.23 196	57.81 144	45.64 171	50.51 174	50.12 188	53.13 174	68.17 198	68.49 189	81.07 171	75.62 174

test-mter		60.84 185	64.62 172	56.42 190	55.99 212	64.18 196	65.39 182	34.23 215	54.39 173	46.21 167	57.40 126	59.49 130	55.86 160	71.02 182	69.65 180	80.87 172	76.20 168

test-LLR		64.42 160	64.36 173	64.49 155	75.02 132	63.93 198	66.61 177	61.96 122	54.41 171	47.77 156	57.46 124	60.25 124	55.20 166	70.80 183	69.33 181	80.40 173	74.38 180

TESTMET0.1,1		61.10 184	64.36 173	57.29 187	57.53 209	63.93 198	66.61 177	36.22 214	54.41 171	47.77 156	57.46 124	60.25 124	55.20 166	70.80 183	69.33 181	80.40 173	74.38 180

CostFormer		68.92 124	69.58 125	68.15 119	75.98 124	76.17 151	78.22 93	51.86 185	65.80 90	61.56 86	63.57 91	62.83 115	61.85 119	70.40 189	68.67 186	79.42 175	79.62 148

CVMVSNet		62.55 170	65.89 159	58.64 183	66.95 189	69.15 182	66.49 179	56.29 170	52.46 182	32.70 197	59.27 111	58.21 137	50.09 179	71.77 175	71.39 175	79.31 176	78.99 152

CHOSEN 1792x2688		69.20 122	69.26 129	69.13 110	76.86 118	78.93 118	77.27 101	60.12 143	61.86 121	54.42 115	42.54 196	61.61 119	66.91 87	78.55 127	78.14 123	79.23 177	83.23 115

PM-MVS		60.48 186	60.94 194	59.94 176	58.85 207	66.83 191	64.27 188	51.39 188	55.03 167	48.03 155	50.00 178	40.79 211	58.26 140	69.20 194	67.13 196	78.84 178	77.60 159

baseline		70.45 107	74.09 88	66.20 146	70.95 173	75.67 153	74.26 132	53.57 173	68.33 77	58.42 96	69.87 64	71.45 74	61.55 122	74.84 157	74.76 160	78.42 179	83.72 111

PatchT		61.97 178	64.04 175	59.55 180	60.49 204	67.40 188	56.54 205	48.65 200	56.69 152	52.65 129	51.10 173	52.14 179	60.92 127	72.20 172	73.09 168	78.03 180	75.69 172

RPSCF		67.64 142	71.25 110	63.43 163	61.86 202	70.73 177	67.26 170	50.86 190	74.20 61	58.91 92	67.49 76	69.33 89	64.10 102	71.41 176	68.45 190	77.61 181	77.17 162

MDTV_nov1_ep13_2view		60.16 187	60.51 195	59.75 177	65.39 193	69.05 183	68.00 167	48.29 202	51.99 184	45.95 169	48.01 185	49.64 192	53.39 172	68.83 195	66.52 197	77.47 182	69.55 193

MDTV_nov1_ep13		64.37 161	65.24 165	63.37 164	68.94 185	70.81 176	72.40 152	50.29 194	60.10 135	53.91 121	60.07 105	59.15 131	57.21 149	69.43 193	67.30 193	77.47 182	69.78 192

LTVRE_ROB	59.44 16	61.82 182	62.64 184	60.87 172	72.83 157	77.19 141	64.37 187	58.97 155	33.56 217	28.00 203	52.59 165	42.21 208	63.93 103	74.52 158	76.28 149	77.15 184	82.13 119
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
IterMVS-SCA-FT		66.89 150	69.22 130	64.17 156	71.30 171	75.64 154	71.33 154	53.17 177	57.63 148	49.08 150	60.72 100	60.05 127	63.09 106	74.99 156	73.92 163	77.07 185	81.57 129

SCA		65.40 155	66.58 158	64.02 158	70.65 174	73.37 168	67.35 169	53.46 175	63.66 107	54.14 117	60.84 99	60.20 126	61.50 123	69.96 190	68.14 191	77.01 186	69.91 190

MDA-MVSNet-bldmvs		53.37 203	53.01 206	53.79 199	43.67 218	67.95 187	59.69 201	57.92 162	43.69 204	32.41 198	41.47 197	27.89 221	52.38 176	56.97 213	65.99 199	76.68 187	67.13 197

MVS-HIRNet		54.41 200	52.10 207	57.11 189	58.99 206	56.10 213	49.68 213	49.10 197	46.18 202	52.15 133	33.18 209	46.11 201	56.10 157	63.19 206	59.70 209	76.64 188	60.25 209

dps		64.00 164	62.99 180	65.18 149	73.29 150	72.07 172	68.98 164	53.07 178	57.74 146	58.41 97	55.55 134	47.74 197	60.89 129	69.53 192	67.14 195	76.44 189	71.19 188

test0.0.03 1		58.80 190	61.58 191	55.56 193	75.02 132	68.45 186	59.58 202	61.96 122	52.74 179	29.57 200	49.75 179	54.56 154	31.46 206	71.19 178	69.77 179	75.75 190	64.57 201

EU-MVSNet		54.63 199	58.69 197	49.90 204	56.99 210	62.70 206	56.41 206	50.64 193	45.95 203	23.14 210	50.42 175	46.51 200	36.63 201	65.51 201	64.85 200	75.57 191	74.91 177

PatchmatchNet		64.21 163	64.65 171	63.69 160	71.29 172	68.66 184	69.63 160	51.70 187	63.04 111	53.77 122	59.83 108	58.34 136	60.23 132	68.54 196	66.06 198	75.56 192	68.08 196
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
Anonymous20231206		56.36 196	57.80 200	54.67 196	70.08 177	66.39 192	60.46 199	57.54 163	49.50 195	29.30 201	33.86 208	46.64 199	35.18 202	70.44 187	68.88 185	75.47 193	68.88 195

CMPMVS	47.78 17	62.49 172	62.52 185	62.46 165	70.01 179	70.66 178	62.97 192	51.84 186	51.98 185	56.71 107	42.87 194	53.62 160	57.80 144	72.23 170	70.37 178	75.45 194	75.91 169
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
MIMVSNet		58.52 192	61.34 192	55.22 194	60.76 203	67.01 190	66.81 174	49.02 198	56.43 155	38.90 187	40.59 200	54.54 155	40.57 197	73.16 165	71.65 173	75.30 195	66.00 199

TAMVS		59.58 189	62.81 183	55.81 192	66.03 192	65.64 195	63.86 189	48.74 199	49.95 192	37.07 193	54.77 140	58.54 134	44.44 188	72.29 169	71.79 172	74.70 196	66.66 198

tpm cat1		65.41 154	63.81 177	67.28 134	75.61 128	72.88 169	75.32 112	52.85 179	62.97 112	63.66 80	53.24 154	53.29 171	61.83 120	65.54 200	64.14 202	74.43 197	74.60 178

test20.03		53.93 202	56.28 203	51.19 202	72.19 160	65.83 193	53.20 209	61.08 129	42.74 206	22.08 212	37.07 204	45.76 203	24.29 214	70.44 187	69.04 183	74.31 198	63.05 205

FMVSNet5		57.24 193	60.02 196	53.99 198	56.45 211	62.74 205	65.27 183	47.03 205	55.14 164	39.55 186	40.88 198	53.42 168	41.83 191	72.35 168	71.10 177	73.79 199	64.50 202

testgi		54.39 201	57.86 199	50.35 203	71.59 168	67.24 189	54.95 207	53.25 176	43.36 205	23.78 208	44.64 191	47.87 196	24.96 211	70.45 186	68.66 187	73.60 200	62.78 206

MIMVSNet1		49.27 205	53.25 205	44.62 208	44.61 216	61.52 208	53.61 208	52.18 183	41.62 209	18.68 217	28.14 214	41.58 210	25.50 209	68.46 197	69.04 183	73.15 201	62.37 207

pmmvs3		47.65 206	49.08 211	45.99 207	44.61 216	54.79 214	50.04 211	31.95 218	33.91 215	29.90 199	30.37 210	33.53 217	46.31 185	63.50 204	63.67 203	73.14 202	63.77 204

FC-MVSNet-test		56.90 195	65.20 166	47.21 206	66.98 188	63.20 203	49.11 214	58.60 160	59.38 138	11.50 221	65.60 81	56.68 144	24.66 213	71.17 179	71.36 176	72.38 203	69.02 194

GG-mvs-BLEND		46.86 209	67.51 149	22.75 215	0.05 226	76.21 150	64.69 185	0.04 223	61.90 120	0.09 227	55.57 133	71.32 76	0.08 222	70.54 185	67.19 194	71.58 204	69.86 191

ambc			53.42 204		64.99 195	63.36 202	49.96 212		47.07 199	37.12 192	28.97 212	16.36 224	41.82 192	75.10 155	67.34 192	71.55 205	75.72 171

tpmrst		62.00 177	62.35 188	61.58 168	71.62 166	64.14 197	69.07 163	48.22 204	62.21 118	53.93 120	58.26 121	55.30 150	55.81 161	63.22 205	62.62 204	70.85 206	70.70 189

tpm		62.41 173	63.15 179	61.55 169	72.24 159	63.79 200	71.31 155	46.12 208	57.82 143	55.33 112	59.90 107	54.74 153	53.63 171	67.24 199	64.29 201	70.65 207	74.25 182

FPMVS		51.87 204	50.00 209	54.07 197	66.83 190	57.25 211	60.25 200	50.91 189	50.25 191	34.36 195	36.04 206	32.02 218	41.49 193	58.98 211	56.07 210	70.56 208	59.36 211

EPMVS		60.00 188	61.97 189	57.71 186	68.46 186	63.17 204	64.54 186	48.23 203	63.30 109	44.72 174	60.19 103	56.05 147	50.85 178	65.27 203	62.02 205	69.44 209	63.81 203

PMVS	39.38 18	46.06 210	43.30 212	49.28 205	62.93 198	38.75 218	41.88 217	53.50 174	33.33 218	35.46 194	28.90 213	31.01 219	33.04 205	58.61 212	54.63 213	68.86 210	57.88 212
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
pmnet_mix02		55.30 198	57.01 202	53.30 201	64.14 197	59.09 209	58.39 204	50.24 195	53.47 177	38.68 188	49.75 179	45.86 202	40.14 198	65.38 202	60.22 207	68.19 211	65.33 200

CHOSEN 280x420		58.70 191	61.88 190	54.98 195	55.45 213	50.55 216	64.92 184	40.36 211	55.21 163	38.13 190	48.31 183	63.76 112	63.03 108	73.73 164	68.58 188	68.00 212	73.04 185

new-patchmatchnet		46.97 208	49.47 210	44.05 210	62.82 199	56.55 212	45.35 216	52.01 184	42.47 207	17.04 219	35.73 207	35.21 215	21.84 217	61.27 208	54.83 212	65.26 213	60.26 208

ADS-MVSNet		55.94 197	58.01 198	53.54 200	62.48 201	58.48 210	59.12 203	46.20 207	59.65 137	42.88 180	52.34 167	53.31 170	46.31 185	62.00 207	60.02 208	64.23 214	60.24 210

N_pmnet		47.35 207	50.13 208	44.11 209	59.98 205	51.64 215	51.86 210	44.80 209	49.58 194	20.76 215	40.65 199	40.05 213	29.64 207	59.84 209	55.15 211	57.63 215	54.00 213

new_pmnet		38.40 211	42.64 213	33.44 212	37.54 221	45.00 217	36.60 218	32.72 217	40.27 210	12.72 220	29.89 211	28.90 220	24.78 212	53.17 214	52.90 214	56.31 216	48.34 214

Gipuma		36.38 212	35.80 214	37.07 211	45.76 215	33.90 219	29.81 219	48.47 201	39.91 211	18.02 218	8.00 222	8.14 226	25.14 210	59.29 210	61.02 206	55.19 217	40.31 215
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
PMMVS2		25.60 213	29.75 215	20.76 216	28.00 222	30.93 220	23.10 221	29.18 219	23.14 220	1.46 226	18.23 218	16.54 223	5.08 220	40.22 215	41.40 216	37.76 218	37.79 217

test_method		22.26 214	25.94 216	17.95 217	3.24 225	7.17 225	23.83 220	7.27 221	37.35 214	20.44 216	21.87 217	39.16 214	18.67 218	34.56 216	20.84 220	34.28 219	20.64 221

E-PMN		21.77 215	18.24 218	25.89 213	40.22 219	19.58 222	12.46 224	39.87 212	18.68 222	6.71 223	9.57 219	4.31 229	22.36 216	19.89 220	27.28 218	33.73 220	28.34 219

EMVS		20.98 216	17.15 219	25.44 214	39.51 220	19.37 223	12.66 223	39.59 213	19.10 221	6.62 224	9.27 220	4.40 228	22.43 215	17.99 221	24.40 219	31.81 221	25.53 220

tmp_tt				14.50 219	14.68 223	7.17 225	10.46 226	2.21 222	37.73 213	28.71 202	25.26 215	16.98 222	4.37 221	31.49 217	29.77 217	26.56 222

MVE	19.12 19	20.47 217	23.27 217	17.20 218	12.66 224	25.41 221	10.52 225	34.14 216	14.79 223	6.53 225	8.79 221	4.68 227	16.64 219	29.49 218	41.63 215	22.73 223	38.11 216
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
DeepMVS_CX						18.74 224	18.55 222	8.02 220	26.96 219	7.33 222	23.81 216	13.05 225	25.99 208	25.17 219		22.45 224	36.25 218

testmvs		0.09 218	0.15 220	0.02 220	0.01 227	0.02 227	0.05 228	0.01 224	0.11 224	0.01 228	0.26 224	0.01 230	0.06 224	0.10 222	0.10 221	0.01 225	0.43 223

test123		0.09 218	0.14 221	0.02 220	0.00 228	0.02 227	0.02 229	0.01 224	0.09 225	0.00 229	0.30 223	0.00 231	0.08 222	0.03 223	0.09 222	0.01 225	0.45 222

uanet_test		0.00 220	0.00 222	0.00 222	0.00 228	0.00 229	0.00 230	0.00 226	0.00 226	0.00 229	0.00 225	0.00 231	0.00 225	0.00 224	0.00 223	0.00 227	0.00 224

sosnet-low-res		0.00 220	0.00 222	0.00 222	0.00 228	0.00 229	0.00 230	0.00 226	0.00 226	0.00 229	0.00 225	0.00 231	0.00 225	0.00 224	0.00 223	0.00 227	0.00 224

sosnet		0.00 220	0.00 222	0.00 222	0.00 228	0.00 229	0.00 230	0.00 226	0.00 226	0.00 229	0.00 225	0.00 231	0.00 225	0.00 224	0.00 223	0.00 227	0.00 224

RE-MVS-def										46.24 166

9.14												86.88 16

SR-MVS					88.99 35			73.57 26				87.54 14

our_test_3					67.93 187	70.99 175	66.89 173

MTAPA										83.48 1		86.45 19

MTMP										82.66 6		84.91 27

Patchmatch-RL test							2.85 227

mPP-MVS					89.90 26							81.29 43

NP-MVS									80.10 47

Patchmtry						65.80 194	65.97 180	52.74 180		52.65 129

High-resolution multi-view benchmark