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 22	86.62 23	83.39 17	90.55 14	89.82 16	89.29 21	73.89 22	84.38 30	76.03 28	79.01 31	85.90 21	78.47 12	87.81 16	86.11 33	92.11 1	93.29 22

CSCG		85.28 21	87.68 19	82.49 24	89.95 23	91.99 5	88.82 24	71.20 37	86.41 21	79.63 16	79.26 29	88.36 10	73.94 41	86.64 31	86.67 25	91.40 2	94.41 8

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 7	91.12 8	88.93 7	78.82 10	87.42 19	86.23 30	91.28 3	93.90 13

SteuartSystems-ACMMP		85.99 16	88.31 16	83.27 20	90.73 10	89.84 14	90.27 14	74.31 15	84.56 29	75.88 29	87.32 14	85.04 24	77.31 23	89.01 7	88.46 3	91.14 4	93.96 12
Skip Steuart: Steuart Systems R&D Blog.
APDe-MVS		88.00 6	90.50 6	85.08 5	90.95 7	91.58 7	92.03 1	75.53 12	91.15 5	80.10 14	92.27 5	88.34 11	80.80 5	88.00 14	86.99 18	91.09 5	95.16 6

DeepC-MVS	78.47 2	84.81 25	86.03 28	83.37 18	89.29 31	90.38 11	88.61 26	76.50 1	86.25 22	77.22 23	75.12 39	80.28 44	77.59 21	88.39 10	88.17 6	91.02 6	93.66 18
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 7	96.21 1

SED-MVS		88.85 2	91.59 3	85.67 2	90.54 15	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 8	95.73 3

DPE-MVS		88.63 4	91.29 4	85.53 3	90.87 8	92.20 4	91.98 2	76.00 6	90.55 8	82.09 6	93.85 1	90.75 2	81.25 1	88.62 8	87.59 14	90.96 9	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 9	91.71 3	90.57 9	90.77 8	75.19 13	90.67 7	80.50 13	86.59 17	88.86 8	78.09 15	89.92 1	89.41 1	90.84 10	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 16	92.36 2	90.69 9	76.15 4	93.08 2	82.75 4	92.19 6	90.71 3	80.45 6	89.27 6	87.91 9	90.82 11	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 13	90.80 9	89.20 23	89.62 19	74.26 16	87.52 14	80.63 11	86.82 16	84.19 28	78.22 14	87.58 17	87.19 16	90.81 12	93.13 24

ACMMPR		85.52 17	87.53 20	83.17 21	90.13 19	89.27 20	89.30 20	73.97 20	86.89 19	77.14 24	86.09 18	83.18 31	77.74 19	87.42 19	87.20 15	90.77 13	92.63 25

ACMMP		83.42 30	85.27 31	81.26 30	88.47 36	88.49 33	88.31 30	72.09 32	83.42 34	72.77 39	82.65 24	78.22 49	75.18 34	86.24 38	85.76 35	90.74 14	92.13 30
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 27	86.29 27	82.23 25	90.04 21	88.82 26	89.23 22	71.74 35	82.82 36	74.61 32	84.41 23	82.09 34	77.03 27	87.13 24	86.73 24	90.73 15	92.06 31

DeepC-MVS_fast	78.24 3	84.27 28	85.50 30	82.85 22	90.46 17	89.24 21	87.83 32	74.24 17	84.88 25	76.23 27	75.26 38	81.05 42	77.62 20	88.02 13	87.62 13	90.69 16	92.41 27
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 14	90.13 19	89.23 22	89.77 18	74.59 14	89.17 10	80.70 10	89.93 11	89.67 5	78.47 12	87.57 18	86.79 22	90.67 17	93.76 16
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 48	81.35 45	77.22 52	82.95 62	84.67 61	81.31 59	63.65 91	72.47 67	68.75 56	73.15 46	78.33 48	75.99 32	86.06 40	83.96 48	90.67 17	90.79 41

TSAR-MVS + MP.		86.88 11	89.23 10	84.14 12	89.78 25	88.67 30	90.59 10	73.46 26	88.99 11	80.52 12	91.26 7	88.65 9	79.91 8	86.96 29	86.22 31	90.59 19	93.83 14
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 38	83.86 35	79.26 41	86.22 48	89.18 24	86.41 37	67.15 64	75.28 53	70.75 51	74.59 41	83.49 30	74.42 38	87.05 27	86.34 29	90.58 20	91.08 39

APD-MVS		86.84 12	88.91 14	84.41 10	90.66 11	90.10 12	90.78 7	75.64 9	87.38 16	78.72 18	90.68 10	86.82 17	80.15 7	87.13 24	86.45 28	90.51 21	93.83 14
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
CP-MVS		84.74 26	86.43 26	82.77 23	89.48 29	88.13 38	88.64 25	73.93 21	84.92 24	76.77 25	81.94 26	83.50 29	77.29 25	86.92 30	86.49 27	90.49 22	93.14 23

XVS					86.63 44	88.68 27	85.00 46			71.81 44		81.92 36				90.47 23

X-MVStestdata					86.63 44	88.68 27	85.00 46			71.81 44		81.92 36				90.47 23

X-MVS		83.23 32	85.20 32	80.92 33	89.71 26	88.68 27	88.21 31	73.60 23	82.57 37	71.81 44	77.07 32	81.92 36	71.72 58	86.98 28	86.86 20	90.47 23	92.36 28

NCCC		85.34 19	86.59 24	83.88 15	91.48 4	88.88 25	89.79 17	75.54 11	86.67 20	77.94 22	76.55 34	84.99 25	78.07 16	88.04 12	87.68 12	90.46 26	93.31 21

CNVR-MVS		86.36 14	88.19 17	84.23 11	91.33 5	89.84 14	90.34 11	75.56 10	87.36 17	78.97 17	81.19 28	86.76 18	78.74 11	89.30 5	88.58 2	90.45 27	94.33 10

MP-MVS		85.50 18	87.40 21	83.28 19	90.65 12	89.51 19	89.16 23	74.11 18	83.70 33	78.06 21	85.54 20	84.89 27	77.31 23	87.40 21	87.14 17	90.41 28	93.65 19
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
3Dnovator+	75.73 4	82.40 34	82.76 39	81.97 28	88.02 37	89.67 17	86.60 36	71.48 36	81.28 41	78.18 20	64.78 84	77.96 51	77.13 26	87.32 22	86.83 21	90.41 28	91.48 35

ACMMP_NAP		86.52 13	89.01 11	83.62 16	90.28 18	90.09 13	90.32 13	74.05 19	88.32 13	79.74 15	87.04 15	85.59 23	76.97 28	89.35 4	88.44 4	90.35 30	94.27 11

CS-MVS		79.22 51	81.11 48	77.01 54	81.36 74	84.03 64	80.35 65	63.25 95	73.43 64	70.37 52	74.10 45	76.03 57	76.40 30	86.32 37	83.95 49	90.34 31	89.93 47

CDPH-MVS		82.64 33	85.03 33	79.86 38	89.41 30	88.31 35	88.32 29	71.84 34	80.11 43	67.47 63	82.09 25	81.44 40	71.85 56	85.89 41	86.15 32	90.24 32	91.25 37

LGP-MVS_train		79.83 43	81.22 47	78.22 48	86.28 47	85.36 56	86.76 35	69.59 46	77.34 48	65.14 72	75.68 36	70.79 78	71.37 62	84.60 50	84.01 46	90.18 33	90.74 42

HPM-MVS++		87.09 9	88.92 13	84.95 6	92.61 1	87.91 39	90.23 15	76.06 5	88.85 12	81.20 9	87.33 13	87.93 12	79.47 9	88.59 9	88.23 5	90.15 34	93.60 20

3Dnovator	73.76 5	79.75 45	80.52 53	78.84 43	84.94 58	87.35 40	84.43 51	65.54 75	78.29 47	73.97 34	63.00 92	75.62 59	74.07 40	85.00 47	85.34 39	90.11 35	89.04 53

DPM-MVS		83.30 31	84.33 34	82.11 26	89.56 27	88.49 33	90.33 12	73.24 27	83.85 32	76.46 26	72.43 49	82.65 32	73.02 48	86.37 35	86.91 19	90.03 36	89.62 51

ETV-MVS		77.32 63	78.81 61	75.58 62	82.24 69	83.64 72	79.98 67	64.02 88	69.64 74	63.90 77	70.89 57	69.94 84	73.41 44	85.39 45	83.91 50	89.92 37	88.31 58

ACMP	73.23 7	79.79 44	80.53 52	78.94 42	85.61 51	85.68 51	85.61 42	69.59 46	77.33 49	71.00 50	74.45 42	69.16 89	71.88 54	83.15 66	83.37 54	89.92 37	90.57 44
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
CS-MVS-test		78.79 57	80.72 50	76.53 57	81.11 79	83.88 67	79.69 74	63.72 90	73.80 61	69.95 54	75.40 37	76.17 55	74.85 35	84.50 53	82.78 59	89.87 39	88.54 57

train_agg		84.86 24	87.21 22	82.11 26	90.59 13	85.47 53	89.81 16	73.55 25	83.95 31	73.30 37	89.84 12	87.23 15	75.61 33	86.47 33	85.46 38	89.78 40	92.06 31

TSAR-MVS + GP.		83.69 29	86.58 25	80.32 35	85.14 53	86.96 43	84.91 49	70.25 41	84.71 28	73.91 35	85.16 21	85.63 22	77.92 17	85.44 42	85.71 36	89.77 41	92.45 26

ACMM	72.26 8	78.86 56	78.13 64	79.71 39	86.89 43	83.40 74	86.02 39	70.50 39	75.28 53	71.49 48	63.01 91	69.26 88	73.57 43	84.11 56	83.98 47	89.76 42	87.84 62
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
OPM-MVS		79.68 47	79.28 60	80.15 37	87.99 38	86.77 45	88.52 28	72.72 29	64.55 98	67.65 62	67.87 73	74.33 64	74.31 39	86.37 35	85.25 40	89.73 43	89.81 49
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 22	91.80 6	91.63 5	75.80 7	91.99 4	81.23 8	92.54 2	89.18 6	80.89 4	87.99 15	87.91 9	89.70 44	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 39	83.41 36	79.53 40	87.06 41	88.59 31	85.47 44	67.96 57	76.59 51	74.05 33	74.69 40	81.98 35	72.98 49	86.14 39	85.47 37	89.68 45	90.42 45

MVS_111021_HR		80.13 42	81.46 44	78.58 45	85.77 50	85.17 57	83.45 54	69.28 49	74.08 60	70.31 53	74.31 43	75.26 60	73.13 46	86.46 34	85.15 41	89.53 46	89.81 49

IS_MVSNet		73.33 82	77.34 73	68.65 114	81.29 75	83.47 73	74.45 123	63.58 93	65.75 90	48.49 149	67.11 77	70.61 79	54.63 166	84.51 52	83.58 53	89.48 47	86.34 76

DELS-MVS		79.15 54	81.07 49	76.91 55	83.54 60	87.31 41	84.45 50	64.92 80	69.98 69	69.34 55	71.62 53	76.26 54	69.84 67	86.57 32	85.90 34	89.39 48	89.88 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
UniMVSNet_NR-MVSNet		70.59 104	72.19 103	68.72 112	77.72 109	80.72 100	73.81 138	69.65 45	61.99 118	43.23 175	60.54 101	57.50 136	58.57 135	79.56 113	81.07 74	89.34 49	83.97 104

casdiffmvs_mvg		77.79 61	79.55 59	75.73 61	81.56 71	84.70 60	82.12 56	64.26 87	74.27 58	67.93 60	70.83 58	74.66 62	69.19 72	83.33 65	81.94 64	89.29 50	87.14 69
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
EIA-MVS		75.64 71	76.60 78	74.53 71	82.43 67	83.84 68	78.32 89	62.28 118	65.96 88	63.28 81	68.95 65	67.54 99	71.61 60	82.55 71	81.63 68	89.24 51	85.72 80

PHI-MVS		82.36 35	85.89 29	78.24 47	86.40 46	89.52 18	85.52 43	69.52 48	82.38 39	65.67 69	81.35 27	82.36 33	73.07 47	87.31 23	86.76 23	89.24 51	91.56 34

canonicalmvs		79.16 53	82.37 42	75.41 63	82.33 68	86.38 49	80.80 62	63.18 97	82.90 35	67.34 64	72.79 48	76.07 56	69.62 68	83.46 64	84.41 45	89.20 53	90.60 43

MSLP-MVS++		82.09 36	82.66 40	81.42 29	87.03 42	87.22 42	85.82 41	70.04 42	80.30 42	78.66 19	68.67 69	81.04 43	77.81 18	85.19 46	84.88 43	89.19 54	91.31 36

HQP-MVS		81.19 40	83.27 37	78.76 44	87.40 40	85.45 54	86.95 34	70.47 40	81.31 40	66.91 66	79.24 30	76.63 53	71.67 59	84.43 54	83.78 51	89.19 54	92.05 33

EPP-MVSNet		74.00 79	77.41 71	70.02 99	80.53 85	83.91 66	74.99 118	62.68 111	65.06 93	49.77 144	68.68 68	72.09 72	63.06 105	82.49 73	80.73 79	89.12 56	88.91 54

NR-MVSNet		68.79 125	70.56 113	66.71 143	77.48 112	79.54 110	73.52 142	69.20 50	61.20 126	39.76 182	58.52 113	50.11 187	51.37 175	80.26 104	80.71 84	88.97 57	83.59 110

PVSNet_Blended_VisFu		76.57 66	77.90 65	75.02 65	80.56 84	86.58 47	79.24 78	66.18 69	64.81 95	68.18 59	65.61 78	71.45 73	67.05 81	84.16 55	81.80 66	88.90 58	90.92 40

TranMVSNet+NR-MVSNet		69.25 120	70.81 112	67.43 127	77.23 114	79.46 112	73.48 143	69.66 44	60.43 131	39.56 183	58.82 112	53.48 163	55.74 160	79.59 111	81.21 72	88.89 59	82.70 114

QAPM		78.47 58	80.22 56	76.43 58	85.03 55	86.75 46	80.62 64	66.00 72	73.77 62	65.35 71	65.54 80	78.02 50	72.69 50	83.71 59	83.36 55	88.87 60	90.41 46

casdiffmvs		76.76 65	78.46 63	74.77 68	80.32 88	83.73 71	80.65 63	63.24 96	73.58 63	66.11 68	69.39 64	74.09 65	69.49 70	82.52 72	79.35 109	88.84 61	86.52 74
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
CPTT-MVS		81.77 37	83.10 38	80.21 36	85.93 49	86.45 48	87.72 33	70.98 38	82.54 38	71.53 47	74.23 44	81.49 39	76.31 31	82.85 69	81.87 65	88.79 62	92.26 29

Effi-MVS+		75.28 73	76.20 79	74.20 73	81.15 77	83.24 77	81.11 60	63.13 99	66.37 84	60.27 87	64.30 88	68.88 93	70.93 65	81.56 78	81.69 67	88.61 63	87.35 65

UniMVSNet (Re)		69.53 116	71.90 106	66.76 141	76.42 118	80.93 96	72.59 148	68.03 56	61.75 121	41.68 180	58.34 119	57.23 138	53.27 171	79.53 114	80.62 88	88.57 64	84.90 96

PCF-MVS	73.28 6	79.42 49	80.41 54	78.26 46	84.88 59	88.17 36	86.08 38	69.85 43	75.23 55	68.43 57	68.03 72	78.38 47	71.76 57	81.26 87	80.65 87	88.56 65	91.18 38
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
GeoE		74.23 77	74.84 85	73.52 75	80.42 87	81.46 90	79.77 71	61.06 128	67.23 81	63.67 78	59.56 108	68.74 95	67.90 78	80.25 105	79.37 108	88.31 66	87.26 68

test2506		71.72 93	72.95 97	70.29 94	81.49 72	83.27 75	75.74 107	67.59 61	68.19 77	49.81 143	61.15 96	49.73 189	58.82 133	84.76 48	82.94 56	88.27 67	80.63 135

ECVR-MVS		72.20 89	73.91 89	70.20 96	81.49 72	83.27 75	75.74 107	67.59 61	68.19 77	49.31 147	55.77 130	62.00 117	58.82 133	84.76 48	82.94 56	88.27 67	80.41 139

MAR-MVS		79.21 52	80.32 55	77.92 49	87.46 39	88.15 37	83.95 52	67.48 63	74.28 57	68.25 58	64.70 85	77.04 52	72.17 52	85.42 43	85.00 42	88.22 69	87.62 64
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 84	73.66 90	72.48 79	77.72 109	80.88 99	78.55 86	58.83 156	65.19 92	60.36 86	59.98 105	62.42 116	71.22 63	81.66 75	80.61 89	88.20 70	84.88 97

ET-MVSNet_ETH3D		72.46 88	74.19 87	70.44 92	62.50 198	81.17 94	79.90 70	62.46 116	64.52 99	57.52 100	71.49 55	59.15 129	72.08 53	78.61 125	81.11 73	88.16 71	83.29 112

OMC-MVS		80.26 41	82.59 41	77.54 50	83.04 61	85.54 52	83.25 55	65.05 79	87.32 18	72.42 40	72.04 51	78.97 46	73.30 45	83.86 57	81.60 69	88.15 72	88.83 55

AdaColmap		79.74 46	78.62 62	81.05 32	89.23 32	86.06 50	84.95 48	71.96 33	79.39 46	75.51 30	63.16 90	68.84 94	76.51 29	83.55 61	82.85 58	88.13 73	86.46 75

test1111		71.56 95	73.44 92	69.38 107	81.16 76	82.95 80	74.99 118	67.68 59	66.89 82	46.33 163	55.19 136	60.91 120	57.99 141	84.59 51	82.70 60	88.12 74	80.85 132

OpenMVS	70.44 10	76.15 69	76.82 77	75.37 64	85.01 56	84.79 59	78.99 82	62.07 119	71.27 68	67.88 61	57.91 121	72.36 71	70.15 66	82.23 74	81.41 70	88.12 74	87.78 63

FA-MVS(training)		73.66 80	74.95 84	72.15 80	78.63 101	80.46 103	78.92 83	54.79 169	69.71 73	65.37 70	62.04 93	66.89 102	67.10 80	80.72 94	79.87 97	88.10 76	84.97 94

UA-Net		74.47 76	77.80 66	70.59 91	85.33 52	85.40 55	73.54 141	65.98 73	60.65 129	56.00 108	72.11 50	79.15 45	54.63 166	83.13 67	82.25 62	88.04 77	81.92 124

DU-MVS		69.63 115	70.91 111	68.13 118	75.99 120	79.54 110	73.81 138	69.20 50	61.20 126	43.23 175	58.52 113	53.50 161	58.57 135	79.22 117	80.45 90	87.97 78	83.97 104

FC-MVSNet-train		72.60 87	75.07 83	69.71 102	81.10 80	78.79 120	73.74 140	65.23 78	66.10 87	53.34 123	70.36 60	63.40 113	56.92 151	81.44 80	80.96 76	87.93 79	84.46 102

IB-MVS	66.94 12	71.21 100	71.66 108	70.68 88	79.18 96	82.83 82	72.61 147	61.77 123	59.66 134	63.44 80	53.26 152	59.65 127	59.16 132	76.78 145	82.11 63	87.90 80	87.33 66
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 67	77.52 69	74.69 69	79.46 94	83.79 69	77.50 96	64.34 85	69.88 70	71.88 42	68.54 70	70.42 80	67.05 81	83.48 62	79.63 100	87.89 81	86.87 71

PVSNet_Blended		76.21 67	77.52 69	74.69 69	79.46 94	83.79 69	77.50 96	64.34 85	69.88 70	71.88 42	68.54 70	70.42 80	67.05 81	83.48 62	79.63 100	87.89 81	86.87 71

DeepPCF-MVS	79.04 1	85.30 20	88.93 12	81.06 31	88.77 35	90.48 10	85.46 45	73.08 28	90.97 6	73.77 36	84.81 22	85.95 20	77.43 22	88.22 11	87.73 11	87.85 83	94.34 9

thisisatest0530		71.48 97	73.01 96	69.70 103	73.83 144	78.62 122	74.53 122	59.12 150	64.13 101	58.63 93	64.60 86	58.63 131	64.27 98	80.28 103	80.17 95	87.82 84	84.64 100

TAPA-MVS	71.42 9	77.69 62	80.05 57	74.94 66	80.68 83	84.52 62	81.36 58	63.14 98	84.77 26	64.82 74	68.72 67	75.91 58	71.86 55	81.62 76	79.55 104	87.80 85	85.24 89
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
tttt0517		71.41 98	72.95 97	69.60 104	73.70 146	78.70 121	74.42 126	59.12 150	63.89 105	58.35 96	64.56 87	58.39 133	64.27 98	80.29 102	80.17 95	87.74 86	84.69 99

MVS_Test		75.37 72	77.13 75	73.31 77	79.07 97	81.32 92	79.98 67	60.12 141	69.72 72	64.11 76	70.53 59	73.22 67	68.90 73	80.14 107	79.48 106	87.67 87	85.50 84

DI_MVS_plusplus_trai		75.13 74	76.12 80	73.96 74	78.18 103	81.55 87	80.97 61	62.54 113	68.59 75	65.13 73	61.43 95	74.81 61	69.32 71	81.01 92	79.59 102	87.64 88	85.89 78

MVSTER		72.06 90	74.24 86	69.51 105	70.39 174	75.97 150	76.91 102	57.36 163	64.64 97	61.39 85	68.86 66	63.76 111	63.46 102	81.44 80	79.70 99	87.56 89	85.31 88

TSAR-MVS + ACMM		85.10 23	88.81 15	80.77 34	89.55 28	88.53 32	88.59 27	72.55 30	87.39 15	71.90 41	90.95 9	87.55 13	74.57 36	87.08 26	86.54 26	87.47 90	93.67 17

CLD-MVS		79.35 50	81.23 46	77.16 53	85.01 56	86.92 44	85.87 40	60.89 130	80.07 45	75.35 31	72.96 47	73.21 68	68.43 77	85.41 44	84.63 44	87.41 91	85.44 86
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 92	71.86 107	71.78 82	78.77 98	80.47 102	78.55 86	61.67 126	60.68 128	55.49 109	58.48 115	65.48 106	68.85 74	76.92 142	75.55 154	87.35 92	85.46 85

WR-MVS		63.03 165	67.40 150	57.92 183	75.14 129	77.60 137	60.56 196	66.10 70	54.11 173	23.88 205	53.94 146	53.58 159	34.50 201	73.93 161	77.71 127	87.35 92	80.94 131

v144192		69.34 119	68.68 137	70.12 97	74.06 140	80.54 101	78.08 92	60.54 134	54.99 166	54.13 116	52.92 159	52.80 172	66.73 88	77.13 140	76.72 143	87.15 94	85.63 81

EPNet		79.08 55	80.62 51	77.28 51	88.90 34	83.17 79	83.65 53	72.41 31	74.41 56	67.15 65	76.78 33	74.37 63	64.43 97	83.70 60	83.69 52	87.15 94	88.19 59
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
EG-PatchMatch MVS		67.24 146	66.94 153	67.60 125	78.73 99	81.35 91	73.28 145	59.49 146	46.89 198	51.42 135	43.65 191	53.49 162	55.50 163	81.38 82	80.66 86	87.15 94	81.17 130

v1144		69.93 113	69.36 127	70.61 90	74.89 132	80.93 96	79.11 80	60.64 132	55.97 158	55.31 111	53.85 147	54.14 154	66.54 90	78.10 130	77.44 133	87.14 97	85.09 91

anonymousdsp		65.28 155	67.98 144	62.13 164	58.73 206	73.98 164	67.10 170	50.69 190	48.41 194	47.66 157	54.27 141	52.75 173	61.45 123	76.71 146	80.20 93	87.13 98	89.53 52

PLC	68.99 11	75.68 70	75.31 82	76.12 60	82.94 63	81.26 93	79.94 69	66.10 70	77.15 50	66.86 67	59.13 111	68.53 96	73.73 42	80.38 100	79.04 110	87.13 98	81.68 126
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
v10		70.22 109	69.76 122	70.74 86	74.79 133	80.30 107	79.22 79	59.81 144	57.71 145	56.58 106	54.22 145	55.31 147	66.95 84	78.28 128	77.47 132	87.12 100	85.07 92

v1192		69.50 117	68.83 133	70.29 94	74.49 136	80.92 98	78.55 86	60.54 134	55.04 164	54.21 114	52.79 161	52.33 174	66.92 85	77.88 132	77.35 136	87.04 101	85.51 83

v1921920		69.03 122	68.32 141	69.86 100	74.03 141	80.37 104	77.55 94	60.25 138	54.62 168	53.59 122	52.36 165	51.50 180	66.75 87	77.17 139	76.69 145	86.96 102	85.56 82

v2v482		70.05 112	69.46 126	70.74 86	74.62 135	80.32 106	79.00 81	60.62 133	57.41 147	56.89 103	55.43 135	55.14 149	66.39 92	77.25 138	77.14 138	86.90 103	83.57 111

Vis-MVSNet		72.77 86	77.20 74	67.59 126	74.19 139	84.01 65	76.61 106	61.69 124	60.62 130	50.61 139	70.25 61	71.31 76	55.57 162	83.85 58	82.28 61	86.90 103	88.08 60
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
PEN-MVS		62.96 166	65.77 160	59.70 176	73.98 142	75.45 154	63.39 189	67.61 60	52.49 179	25.49 204	53.39 149	49.12 191	40.85 194	71.94 173	77.26 137	86.86 105	80.72 134

v8		70.23 108	69.86 120	70.67 89	74.69 134	79.82 109	78.79 84	59.18 149	58.80 138	58.20 97	55.00 137	57.33 137	66.31 93	77.51 135	76.71 144	86.82 106	83.88 107

ACMH+	66.54 13	71.36 99	70.09 117	72.85 78	82.59 65	81.13 95	78.56 85	68.04 55	61.55 122	52.52 130	51.50 169	54.14 154	68.56 76	78.85 122	79.50 105	86.82 106	83.94 106

v1240		68.64 127	67.89 146	69.51 105	73.89 143	80.26 108	76.73 104	59.97 143	53.43 176	53.08 125	51.82 168	50.84 183	66.62 89	76.79 144	76.77 142	86.78 108	85.34 87

DCV-MVSNet		73.65 81	75.78 81	71.16 85	80.19 89	79.27 114	77.45 98	61.68 125	66.73 83	58.72 92	65.31 81	69.96 83	62.19 110	81.29 86	80.97 75	86.74 109	86.91 70

GBi-Net		70.78 101	73.37 94	67.76 119	72.95 151	78.00 127	75.15 113	62.72 106	64.13 101	51.44 132	58.37 116	69.02 90	57.59 143	81.33 83	80.72 80	86.70 110	82.02 118

test1		70.78 101	73.37 94	67.76 119	72.95 151	78.00 127	75.15 113	62.72 106	64.13 101	51.44 132	58.37 116	69.02 90	57.59 143	81.33 83	80.72 80	86.70 110	82.02 118

FMVSNet2		70.39 107	72.67 101	67.72 122	72.95 151	78.00 127	75.15 113	62.69 110	63.29 109	51.25 136	55.64 131	68.49 97	57.59 143	80.91 93	80.35 92	86.70 110	82.02 118

v7n		67.05 148	66.94 153	67.17 133	72.35 156	78.97 115	73.26 146	58.88 155	51.16 187	50.90 137	48.21 182	50.11 187	60.96 124	77.70 133	77.38 134	86.68 113	85.05 93

WR-MVS_H		61.83 180	65.87 159	57.12 186	71.72 161	76.87 141	61.45 194	66.19 68	51.97 184	22.92 209	53.13 156	52.30 176	33.80 202	71.03 180	75.00 157	86.65 114	80.78 133

MSDG		71.52 96	69.87 119	73.44 76	82.21 70	79.35 113	79.52 75	64.59 82	66.15 86	61.87 82	53.21 154	56.09 144	65.85 95	78.94 121	78.50 117	86.60 115	76.85 163

FMVSNet3		70.49 105	72.90 99	67.67 124	72.88 154	77.98 130	74.96 120	62.72 106	64.13 101	51.44 132	58.37 116	69.02 90	57.43 146	79.43 115	79.57 103	86.59 116	81.81 125

MVS_111021_LR		78.13 60	79.85 58	76.13 59	81.12 78	81.50 89	80.28 66	65.25 77	76.09 52	71.32 49	76.49 35	72.87 70	72.21 51	82.79 70	81.29 71	86.59 116	87.91 61

DTE-MVSNet		61.85 178	64.96 169	58.22 182	74.32 138	74.39 163	61.01 195	67.85 58	51.76 186	21.91 212	53.28 151	48.17 192	37.74 198	72.22 170	76.44 147	86.52 118	78.49 152

baseline2		69.69 114	70.27 116	69.01 110	75.72 124	77.13 140	73.82 137	58.94 154	61.35 124	57.09 102	61.68 94	57.17 139	61.99 114	78.10 130	76.58 146	86.48 119	79.85 143

thisisatest0515		67.40 144	68.78 134	65.80 146	70.02 176	75.24 157	69.36 160	57.37 162	54.94 167	53.67 121	55.53 134	54.85 150	58.00 140	78.19 129	78.91 113	86.39 120	83.78 108

FMVSNet1		68.84 124	70.47 115	66.94 138	71.35 168	77.68 135	74.71 121	62.35 117	56.93 149	49.94 142	50.01 175	64.59 108	57.07 148	81.33 83	80.72 80	86.25 121	82.00 121

UGNet		72.78 85	77.67 67	67.07 136	71.65 163	83.24 77	75.20 112	63.62 92	64.93 94	56.72 104	71.82 52	73.30 66	49.02 179	81.02 91	80.70 85	86.22 122	88.67 56
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 91	72.48 102	71.21 84	80.14 90	81.53 88	76.92 101	62.89 102	64.46 100	58.94 89	43.80 190	70.98 77	62.22 109	80.70 95	80.19 94	86.18 123	85.73 79

tfpn200view9		68.11 130	68.72 136	67.40 128	77.83 107	78.93 116	74.28 128	62.81 103	56.64 151	46.82 159	52.65 162	53.47 164	56.59 152	80.41 97	78.43 118	86.11 124	80.52 137

CANet_DTU		73.29 83	76.96 76	69.00 111	77.04 115	82.06 85	79.49 76	56.30 166	67.85 79	53.29 124	71.12 56	70.37 82	61.81 119	81.59 77	80.96 76	86.09 125	84.73 98

CP-MVSNet		62.68 168	65.49 163	59.40 179	71.84 159	75.34 155	62.87 191	67.04 65	52.64 178	27.19 202	53.38 150	48.15 193	41.40 192	71.26 176	75.68 152	86.07 126	82.00 121

ACMH	65.37 14	70.71 103	70.00 118	71.54 83	82.51 66	82.47 84	77.78 93	68.13 54	56.19 156	46.06 166	54.30 140	51.20 181	68.68 75	80.66 96	80.72 80	86.07 126	84.45 103
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
thres600view7		67.68 138	68.43 140	66.80 140	77.90 104	78.86 118	73.84 136	62.75 104	56.07 157	44.70 173	52.85 160	52.81 171	55.58 161	80.41 97	77.77 126	86.05 128	80.28 140

thres200		67.98 132	68.55 139	67.30 131	77.89 106	78.86 118	74.18 132	62.75 104	56.35 154	46.48 162	52.98 158	53.54 160	56.46 153	80.41 97	77.97 124	86.05 128	79.78 145

CNLPA		77.20 64	77.54 68	76.80 56	82.63 64	84.31 63	79.77 71	64.64 81	85.17 23	73.18 38	56.37 128	69.81 85	74.53 37	81.12 90	78.69 115	86.04 130	87.29 67

TSAR-MVS + COLMAP		78.34 59	81.64 43	74.48 72	80.13 91	85.01 58	81.73 57	65.93 74	84.75 27	61.68 83	85.79 19	66.27 104	71.39 61	82.91 68	80.78 78	86.01 131	85.98 77

LS3D		74.08 78	73.39 93	74.88 67	85.05 54	82.62 83	79.71 73	68.66 52	72.82 65	58.80 91	57.61 122	61.31 119	71.07 64	80.32 101	78.87 114	86.00 132	80.18 141

Anonymous202405211			72.16 105		80.85 82	81.85 86	76.88 103	65.40 76	62.89 113		46.35 186	67.99 98	62.05 112	81.15 89	80.38 91	85.97 133	84.50 101

PS-CasMVS		62.38 174	65.06 166	59.25 180	71.73 160	75.21 159	62.77 192	66.99 66	51.94 185	26.96 203	52.00 167	47.52 196	41.06 193	71.16 179	75.60 153	85.97 133	81.97 123

thres400		67.95 133	68.62 138	67.17 133	77.90 104	78.59 123	74.27 129	62.72 106	56.34 155	45.77 168	53.00 157	53.35 167	56.46 153	80.21 106	78.43 118	85.91 135	80.43 138

thres100view900		67.60 142	68.02 143	67.12 135	77.83 107	77.75 134	73.90 135	62.52 114	56.64 151	46.82 159	52.65 162	53.47 164	55.92 157	78.77 123	77.62 129	85.72 136	79.23 148

Vis-MVSNet (Re-imp)		67.83 136	73.52 91	61.19 168	78.37 102	76.72 144	66.80 173	62.96 100	65.50 91	34.17 194	67.19 76	69.68 86	39.20 197	79.39 116	79.44 107	85.68 137	76.73 164

baseline1		70.10 111	72.17 104	67.69 123	79.74 92	76.80 142	73.91 134	64.38 84	62.74 114	48.30 151	64.94 82	64.08 110	54.17 168	81.46 79	78.92 112	85.66 138	76.22 165

V42		68.76 126	69.63 123	67.74 121	64.93 194	78.01 126	78.30 90	56.48 165	58.65 139	56.30 107	54.26 143	57.03 140	64.85 96	77.47 136	77.01 140	85.60 139	84.96 95

TransMVSNet (Re)		64.74 158	65.66 161	63.66 159	77.40 113	75.33 156	69.86 156	62.67 112	47.63 196	41.21 181	50.01 175	52.33 174	45.31 185	79.57 112	77.69 128	85.49 140	77.07 162

IterMVS-LS		71.69 94	72.82 100	70.37 93	77.54 111	76.34 147	75.13 116	60.46 136	61.53 123	57.57 99	64.89 83	67.33 100	66.04 94	77.09 141	77.37 135	85.48 141	85.18 90
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
UniMVSNet_ETH3D		67.18 147	67.03 152	67.36 129	74.44 137	78.12 125	74.07 133	66.38 67	52.22 181	46.87 158	48.64 180	51.84 178	56.96 149	77.29 137	78.53 116	85.42 142	82.59 115

Baseline_NR-MVSNet		67.53 143	68.77 135	66.09 145	75.99 120	74.75 161	72.43 149	68.41 53	61.33 125	38.33 187	51.31 170	54.13 156	56.03 156	79.22 117	78.19 121	85.37 143	82.45 116

Fast-Effi-MVS+-dtu		68.34 128	69.47 125	67.01 137	75.15 128	77.97 132	77.12 100	55.40 168	57.87 140	46.68 161	56.17 129	60.39 121	62.36 108	76.32 149	76.25 150	85.35 144	81.34 128

diffmvs		74.86 75	77.37 72	71.93 81	75.62 125	80.35 105	79.42 77	60.15 140	72.81 66	64.63 75	71.51 54	73.11 69	66.53 91	79.02 120	77.98 123	85.25 145	86.83 73
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
GA-MVS		68.14 129	69.17 130	66.93 139	73.77 145	78.50 124	74.45 123	58.28 158	55.11 163	48.44 150	60.08 103	53.99 157	61.50 121	78.43 127	77.57 130	85.13 146	80.54 136

v148		67.85 135	67.53 147	68.23 116	73.25 149	77.57 138	74.26 130	57.36 163	55.70 159	57.45 101	53.53 148	55.42 146	61.96 115	75.23 153	73.92 162	85.08 147	81.32 129

HyFIR lowres test		69.47 118	68.94 132	70.09 98	76.77 117	82.93 81	76.63 105	60.17 139	59.00 137	54.03 117	40.54 199	65.23 107	67.89 79	76.54 148	78.30 120	85.03 148	80.07 142

gg-mvs-nofinetune		62.55 169	65.05 167	59.62 177	78.72 100	77.61 136	70.83 155	53.63 170	39.71 210	22.04 211	36.36 203	64.32 109	47.53 181	81.16 88	79.03 111	85.00 149	77.17 160

COLMAP_ROB	62.73 15	67.66 139	66.76 155	68.70 113	80.49 86	77.98 130	75.29 111	62.95 101	63.62 107	49.96 141	47.32 185	50.72 184	58.57 135	76.87 143	75.50 155	84.94 150	75.33 174
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
tfpnnormal		64.27 161	63.64 177	65.02 149	75.84 123	75.61 153	71.24 154	62.52 114	47.79 195	42.97 177	42.65 193	44.49 203	52.66 173	78.77 123	76.86 141	84.88 151	79.29 147

pm-mvs1		65.62 152	67.42 149	63.53 160	73.66 147	76.39 146	69.66 157	60.87 131	49.73 191	43.97 174	51.24 171	57.00 141	48.16 180	79.89 108	77.84 125	84.85 152	79.82 144

gm-plane-assit		57.00 193	57.62 200	56.28 189	76.10 119	62.43 205	47.62 213	46.57 204	33.84 214	23.24 207	37.52 200	40.19 210	59.61 131	79.81 109	77.55 131	84.55 153	72.03 184

USDC		67.36 145	67.90 145	66.74 142	71.72 161	75.23 158	71.58 151	60.28 137	67.45 80	50.54 140	60.93 97	45.20 202	62.08 111	76.56 147	74.50 160	84.25 154	75.38 173

MS-PatchMatch		70.17 110	70.49 114	69.79 101	80.98 81	77.97 132	77.51 95	58.95 153	62.33 116	55.22 112	53.14 155	65.90 105	62.03 113	79.08 119	77.11 139	84.08 155	77.91 155

TDRefinement		66.09 151	65.03 168	67.31 130	69.73 178	76.75 143	75.33 109	64.55 83	60.28 132	49.72 145	45.63 188	42.83 205	60.46 129	75.75 150	75.95 151	84.08 155	78.04 154

pmmvs4		67.89 134	67.39 151	68.48 115	71.60 165	73.57 165	74.45 123	60.98 129	64.65 96	57.97 98	54.95 138	51.73 179	61.88 116	73.78 162	75.11 156	83.99 157	77.91 155

pmmvs5		62.37 175	64.04 174	60.42 171	65.03 192	71.67 172	67.17 169	52.70 180	50.30 188	44.80 171	54.23 144	51.19 182	49.37 178	72.88 165	73.48 166	83.45 158	74.55 177

pmmvs-eth3d		63.52 164	62.44 186	64.77 151	66.82 189	70.12 177	69.41 159	59.48 147	54.34 172	52.71 126	46.24 187	44.35 204	56.93 150	72.37 166	73.77 164	83.30 159	75.91 167

pmmvs6		62.41 172	62.88 180	61.87 165	71.38 167	75.18 160	67.76 166	59.45 148	41.64 206	42.52 179	37.33 201	52.91 170	46.87 182	77.67 134	76.26 149	83.23 160	79.18 149

CDS-MVSNet		67.65 140	69.83 121	65.09 148	75.39 127	76.55 145	74.42 126	63.75 89	53.55 174	49.37 146	59.41 109	62.45 115	44.44 186	79.71 110	79.82 98	83.17 161	77.36 159
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
PMMVS		65.06 156	69.17 130	60.26 173	55.25 212	63.43 199	66.71 174	43.01 208	62.41 115	50.64 138	69.44 63	67.04 101	63.29 103	74.36 159	73.54 165	82.68 162	73.99 181

SixPastTwentyTwo		61.84 179	62.45 185	61.12 169	69.20 182	72.20 169	62.03 193	57.40 161	46.54 199	38.03 189	57.14 126	41.72 207	58.12 139	69.67 190	71.58 173	81.94 163	78.30 153

IterMVS		66.36 150	68.30 142	64.10 155	69.48 181	74.61 162	73.41 144	50.79 189	57.30 148	48.28 152	60.64 100	59.92 126	60.85 128	74.14 160	72.66 169	81.80 164	78.82 151
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
PatchMatch-RL		67.78 137	66.65 156	69.10 109	73.01 150	72.69 168	68.49 163	61.85 122	62.93 112	60.20 88	56.83 127	50.42 185	69.52 69	75.62 151	74.46 161	81.51 165	73.62 182

TinyColmap		62.84 167	61.03 192	64.96 150	69.61 179	71.69 171	68.48 164	59.76 145	55.41 160	47.69 156	47.33 184	34.20 214	62.76 107	74.52 157	72.59 170	81.44 166	71.47 185

EPNet_dtu		68.08 131	71.00 110	64.67 152	79.64 93	68.62 183	75.05 117	63.30 94	66.36 85	45.27 170	67.40 75	66.84 103	43.64 188	75.37 152	74.98 158	81.15 167	77.44 158
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
CR-MVSNet		64.83 157	65.54 162	64.01 157	70.64 173	69.41 178	65.97 178	52.74 178	57.81 142	52.65 127	54.27 141	56.31 143	60.92 125	72.20 171	73.09 167	81.12 168	75.69 170

RPMNet		61.71 182	62.88 180	60.34 172	69.51 180	69.41 178	63.48 188	49.23 194	57.81 142	45.64 169	50.51 173	50.12 186	53.13 172	68.17 197	68.49 188	81.07 169	75.62 172

test-mter		60.84 184	64.62 171	56.42 188	55.99 210	64.18 194	65.39 180	34.23 213	54.39 171	46.21 165	57.40 125	59.49 128	55.86 158	71.02 181	69.65 179	80.87 170	76.20 166

test-LLR		64.42 159	64.36 172	64.49 153	75.02 130	63.93 196	66.61 175	61.96 120	54.41 169	47.77 154	57.46 123	60.25 122	55.20 164	70.80 182	69.33 180	80.40 171	74.38 178

TESTMET0.1,1		61.10 183	64.36 172	57.29 185	57.53 207	63.93 196	66.61 175	36.22 212	54.41 169	47.77 154	57.46 123	60.25 122	55.20 164	70.80 182	69.33 180	80.40 171	74.38 178

CostFormer		68.92 123	69.58 124	68.15 117	75.98 122	76.17 149	78.22 91	51.86 183	65.80 89	61.56 84	63.57 89	62.83 114	61.85 117	70.40 188	68.67 185	79.42 173	79.62 146

CVMVSNet		62.55 169	65.89 158	58.64 181	66.95 187	69.15 180	66.49 177	56.29 167	52.46 180	32.70 195	59.27 110	58.21 135	50.09 177	71.77 174	71.39 174	79.31 174	78.99 150

CHOSEN 1792x2688		69.20 121	69.26 128	69.13 108	76.86 116	78.93 116	77.27 99	60.12 141	61.86 120	54.42 113	42.54 194	61.61 118	66.91 86	78.55 126	78.14 122	79.23 175	83.23 113

PM-MVS		60.48 185	60.94 193	59.94 174	58.85 205	66.83 189	64.27 186	51.39 186	55.03 165	48.03 153	50.00 177	40.79 209	58.26 138	69.20 193	67.13 195	78.84 176	77.60 157

baseline		70.45 106	74.09 88	66.20 144	70.95 171	75.67 151	74.26 130	53.57 171	68.33 76	58.42 94	69.87 62	71.45 73	61.55 120	74.84 156	74.76 159	78.42 177	83.72 109

PatchT		61.97 177	64.04 174	59.55 178	60.49 202	67.40 186	56.54 203	48.65 198	56.69 150	52.65 127	51.10 172	52.14 177	60.92 125	72.20 171	73.09 167	78.03 178	75.69 170

RPSCF		67.64 141	71.25 109	63.43 161	61.86 200	70.73 175	67.26 168	50.86 188	74.20 59	58.91 90	67.49 74	69.33 87	64.10 100	71.41 175	68.45 189	77.61 179	77.17 160

MDTV_nov1_ep13_2view		60.16 186	60.51 194	59.75 175	65.39 191	69.05 181	68.00 165	48.29 200	51.99 182	45.95 167	48.01 183	49.64 190	53.39 170	68.83 194	66.52 196	77.47 180	69.55 191

MDTV_nov1_ep13		64.37 160	65.24 164	63.37 162	68.94 183	70.81 174	72.40 150	50.29 192	60.10 133	53.91 119	60.07 104	59.15 129	57.21 147	69.43 192	67.30 192	77.47 180	69.78 190

LTVRE_ROB	59.44 16	61.82 181	62.64 183	60.87 170	72.83 155	77.19 139	64.37 185	58.97 152	33.56 215	28.00 201	52.59 164	42.21 206	63.93 101	74.52 157	76.28 148	77.15 182	82.13 117
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		66.89 149	69.22 129	64.17 154	71.30 169	75.64 152	71.33 152	53.17 175	57.63 146	49.08 148	60.72 99	60.05 125	63.09 104	74.99 155	73.92 162	77.07 183	81.57 127

SCA		65.40 154	66.58 157	64.02 156	70.65 172	73.37 166	67.35 167	53.46 173	63.66 106	54.14 115	60.84 98	60.20 124	61.50 121	69.96 189	68.14 190	77.01 184	69.91 188

MDA-MVSNet-bldmvs		53.37 202	53.01 205	53.79 197	43.67 216	67.95 185	59.69 199	57.92 159	43.69 202	32.41 196	41.47 195	27.89 219	52.38 174	56.97 212	65.99 198	76.68 185	67.13 195

MVS-HIRNet		54.41 199	52.10 206	57.11 187	58.99 204	56.10 211	49.68 211	49.10 195	46.18 200	52.15 131	33.18 207	46.11 199	56.10 155	63.19 205	59.70 208	76.64 186	60.25 207

dps		64.00 163	62.99 179	65.18 147	73.29 148	72.07 170	68.98 162	53.07 176	57.74 144	58.41 95	55.55 133	47.74 195	60.89 127	69.53 191	67.14 194	76.44 187	71.19 186

test0.0.03 1		58.80 189	61.58 190	55.56 191	75.02 130	68.45 184	59.58 200	61.96 120	52.74 177	29.57 198	49.75 178	54.56 152	31.46 204	71.19 177	69.77 178	75.75 188	64.57 199

EU-MVSNet		54.63 198	58.69 196	49.90 202	56.99 208	62.70 204	56.41 204	50.64 191	45.95 201	23.14 208	50.42 174	46.51 198	36.63 199	65.51 200	64.85 199	75.57 189	74.91 175

PatchmatchNet		64.21 162	64.65 170	63.69 158	71.29 170	68.66 182	69.63 158	51.70 185	63.04 110	53.77 120	59.83 107	58.34 134	60.23 130	68.54 195	66.06 197	75.56 190	68.08 194
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
Anonymous20231206		56.36 195	57.80 199	54.67 194	70.08 175	66.39 190	60.46 197	57.54 160	49.50 193	29.30 199	33.86 206	46.64 197	35.18 200	70.44 186	68.88 184	75.47 191	68.88 193

CMPMVS	47.78 17	62.49 171	62.52 184	62.46 163	70.01 177	70.66 176	62.97 190	51.84 184	51.98 183	56.71 105	42.87 192	53.62 158	57.80 142	72.23 169	70.37 177	75.45 192	75.91 167
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
MIMVSNet		58.52 191	61.34 191	55.22 192	60.76 201	67.01 188	66.81 172	49.02 196	56.43 153	38.90 185	40.59 198	54.54 153	40.57 195	73.16 164	71.65 172	75.30 193	66.00 197

TAMVS		59.58 188	62.81 182	55.81 190	66.03 190	65.64 193	63.86 187	48.74 197	49.95 190	37.07 191	54.77 139	58.54 132	44.44 186	72.29 168	71.79 171	74.70 194	66.66 196

tpm cat1		65.41 153	63.81 176	67.28 132	75.61 126	72.88 167	75.32 110	52.85 177	62.97 111	63.66 79	53.24 153	53.29 169	61.83 118	65.54 199	64.14 201	74.43 195	74.60 176

test20.03		53.93 201	56.28 202	51.19 200	72.19 158	65.83 191	53.20 207	61.08 127	42.74 204	22.08 210	37.07 202	45.76 201	24.29 212	70.44 186	69.04 182	74.31 196	63.05 203

FMVSNet5		57.24 192	60.02 195	53.99 196	56.45 209	62.74 203	65.27 181	47.03 203	55.14 162	39.55 184	40.88 196	53.42 166	41.83 189	72.35 167	71.10 176	73.79 197	64.50 200

testgi		54.39 200	57.86 198	50.35 201	71.59 166	67.24 187	54.95 205	53.25 174	43.36 203	23.78 206	44.64 189	47.87 194	24.96 209	70.45 185	68.66 186	73.60 198	62.78 204

MIMVSNet1		49.27 204	53.25 204	44.62 206	44.61 214	61.52 206	53.61 206	52.18 181	41.62 207	18.68 215	28.14 212	41.58 208	25.50 207	68.46 196	69.04 182	73.15 199	62.37 205

pmmvs3		47.65 205	49.08 210	45.99 205	44.61 214	54.79 212	50.04 209	31.95 216	33.91 213	29.90 197	30.37 208	33.53 215	46.31 183	63.50 203	63.67 202	73.14 200	63.77 202

FC-MVSNet-test		56.90 194	65.20 165	47.21 204	66.98 186	63.20 201	49.11 212	58.60 157	59.38 136	11.50 219	65.60 79	56.68 142	24.66 211	71.17 178	71.36 175	72.38 201	69.02 192

GG-mvs-BLEND		46.86 208	67.51 148	22.75 213	0.05 224	76.21 148	64.69 183	0.04 221	61.90 119	0.09 225	55.57 132	71.32 75	0.08 220	70.54 184	67.19 193	71.58 202	69.86 189

ambc			53.42 203		64.99 193	63.36 200	49.96 210		47.07 197	37.12 190	28.97 210	16.36 222	41.82 190	75.10 154	67.34 191	71.55 203	75.72 169

tpmrst		62.00 176	62.35 187	61.58 166	71.62 164	64.14 195	69.07 161	48.22 202	62.21 117	53.93 118	58.26 120	55.30 148	55.81 159	63.22 204	62.62 203	70.85 204	70.70 187

tpm		62.41 172	63.15 178	61.55 167	72.24 157	63.79 198	71.31 153	46.12 206	57.82 141	55.33 110	59.90 106	54.74 151	53.63 169	67.24 198	64.29 200	70.65 205	74.25 180

FPMVS		51.87 203	50.00 208	54.07 195	66.83 188	57.25 209	60.25 198	50.91 187	50.25 189	34.36 193	36.04 204	32.02 216	41.49 191	58.98 210	56.07 209	70.56 206	59.36 209

EPMVS		60.00 187	61.97 188	57.71 184	68.46 184	63.17 202	64.54 184	48.23 201	63.30 108	44.72 172	60.19 102	56.05 145	50.85 176	65.27 202	62.02 204	69.44 207	63.81 201

PMVS	39.38 18	46.06 209	43.30 211	49.28 203	62.93 196	38.75 216	41.88 215	53.50 172	33.33 216	35.46 192	28.90 211	31.01 217	33.04 203	58.61 211	54.63 212	68.86 208	57.88 210
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
pmnet_mix02		55.30 197	57.01 201	53.30 199	64.14 195	59.09 207	58.39 202	50.24 193	53.47 175	38.68 186	49.75 178	45.86 200	40.14 196	65.38 201	60.22 206	68.19 209	65.33 198

CHOSEN 280x420		58.70 190	61.88 189	54.98 193	55.45 211	50.55 214	64.92 182	40.36 209	55.21 161	38.13 188	48.31 181	63.76 111	63.03 106	73.73 163	68.58 187	68.00 210	73.04 183

new-patchmatchnet		46.97 207	49.47 209	44.05 208	62.82 197	56.55 210	45.35 214	52.01 182	42.47 205	17.04 217	35.73 205	35.21 213	21.84 215	61.27 207	54.83 211	65.26 211	60.26 206

ADS-MVSNet		55.94 196	58.01 197	53.54 198	62.48 199	58.48 208	59.12 201	46.20 205	59.65 135	42.88 178	52.34 166	53.31 168	46.31 183	62.00 206	60.02 207	64.23 212	60.24 208

N_pmnet		47.35 206	50.13 207	44.11 207	59.98 203	51.64 213	51.86 208	44.80 207	49.58 192	20.76 213	40.65 197	40.05 211	29.64 205	59.84 208	55.15 210	57.63 213	54.00 211

new_pmnet		38.40 210	42.64 212	33.44 210	37.54 219	45.00 215	36.60 216	32.72 215	40.27 208	12.72 218	29.89 209	28.90 218	24.78 210	53.17 213	52.90 213	56.31 214	48.34 212

Gipuma		36.38 211	35.80 213	37.07 209	45.76 213	33.90 217	29.81 217	48.47 199	39.91 209	18.02 216	8.00 220	8.14 224	25.14 208	59.29 209	61.02 205	55.19 215	40.31 213
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
PMMVS2		25.60 212	29.75 214	20.76 214	28.00 220	30.93 218	23.10 219	29.18 217	23.14 218	1.46 224	18.23 216	16.54 221	5.08 218	40.22 214	41.40 215	37.76 216	37.79 215

test_method		22.26 213	25.94 215	17.95 215	3.24 223	7.17 223	23.83 218	7.27 219	37.35 212	20.44 214	21.87 215	39.16 212	18.67 216	34.56 215	20.84 219	34.28 217	20.64 219

E-PMN		21.77 214	18.24 217	25.89 211	40.22 217	19.58 220	12.46 222	39.87 210	18.68 220	6.71 221	9.57 217	4.31 227	22.36 214	19.89 219	27.28 217	33.73 218	28.34 217

EMVS		20.98 215	17.15 218	25.44 212	39.51 218	19.37 221	12.66 221	39.59 211	19.10 219	6.62 222	9.27 218	4.40 226	22.43 213	17.99 220	24.40 218	31.81 219	25.53 218

tmp_tt				14.50 217	14.68 221	7.17 223	10.46 224	2.21 220	37.73 211	28.71 200	25.26 213	16.98 220	4.37 219	31.49 216	29.77 216	26.56 220

MVE	19.12 19	20.47 216	23.27 216	17.20 216	12.66 222	25.41 219	10.52 223	34.14 214	14.79 221	6.53 223	8.79 219	4.68 225	16.64 217	29.49 217	41.63 214	22.73 221	38.11 214
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
DeepMVS_CX						18.74 222	18.55 220	8.02 218	26.96 217	7.33 220	23.81 214	13.05 223	25.99 206	25.17 218		22.45 222	36.25 216

testmvs		0.09 217	0.15 219	0.02 218	0.01 225	0.02 225	0.05 226	0.01 222	0.11 222	0.01 226	0.26 222	0.01 228	0.06 222	0.10 221	0.10 220	0.01 223	0.43 221

test123		0.09 217	0.14 220	0.02 218	0.00 226	0.02 225	0.02 227	0.01 222	0.09 223	0.00 227	0.30 221	0.00 229	0.08 220	0.03 222	0.09 221	0.01 223	0.45 220

uanet_test		0.00 219	0.00 221	0.00 220	0.00 226	0.00 227	0.00 228	0.00 224	0.00 224	0.00 227	0.00 223	0.00 229	0.00 223	0.00 223	0.00 222	0.00 225	0.00 222

sosnet-low-res		0.00 219	0.00 221	0.00 220	0.00 226	0.00 227	0.00 228	0.00 224	0.00 224	0.00 227	0.00 223	0.00 229	0.00 223	0.00 223	0.00 222	0.00 225	0.00 222

sosnet		0.00 219	0.00 221	0.00 220	0.00 226	0.00 227	0.00 228	0.00 224	0.00 224	0.00 227	0.00 223	0.00 229	0.00 223	0.00 223	0.00 222	0.00 225	0.00 222

RE-MVS-def										46.24 164

9.14												86.88 16

SR-MVS					88.99 33			73.57 24				87.54 14

our_test_3					67.93 185	70.99 173	66.89 171

MTAPA										83.48 1		86.45 19

MTMP										82.66 5		84.91 26

Patchmatch-RL test							2.85 225

mPP-MVS					89.90 24							81.29 41

NP-MVS									80.10 44

Patchmtry						65.80 192	65.97 178	52.74 178		52.65 127

High-resolution multi-view benchmark