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.

SED-MVS		95.53 1	95.79 1	95.23 1	97.60 10	98.92 1	95.99 5	92.05 7	97.14 1	94.19 1	94.71 6	93.25 2	95.08 1	94.32 11	92.59 15	96.49 17	99.58 3

DPE-MVS		95.10 2	95.53 2	94.60 5	97.77 8	98.64 5	96.60 4	92.45 5	96.34 6	91.41 6	96.70 2	92.26 6	93.56 6	93.68 18	91.73 30	95.79 40	99.37 7
Kehua Chen, Zhenlong Yuan, Tianlu Mao, Zhaoqi Wang: Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization. AAAI2025
DVP-MVS		95.06 3	95.37 4	94.70 3	97.59 11	98.89 2	95.37 12	92.04 8	96.85 3	94.00 2	92.81 14	93.02 3	92.93 7	94.22 14	92.15 21	96.30 25	99.61 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
DVP-MVS++		95.03 4	95.03 5	95.03 2	97.91 6	98.84 3	95.80 6	91.88 10	96.65 5	93.15 3	93.79 8	90.11 12	95.03 2	94.20 16	92.39 16	96.44 21	99.22 10

MSP-MVS		95.00 5	95.47 3	94.45 6	96.78 19	98.11 11	95.72 8	90.91 14	96.68 4	91.57 5	96.98 1	89.47 15	94.76 3	95.24 3	92.15 21	96.98 7	99.64 1
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
CNVR-MVS		94.53 6	94.85 7	94.15 8	98.03 4	98.59 7	95.56 9	92.91 2	94.86 13	88.46 14	91.32 21	90.83 10	94.03 5	95.20 4	94.16 6	95.89 35	99.01 17

SF-MVS		94.40 7	94.15 13	94.70 3	98.25 3	98.24 9	96.86 3	93.46 1	94.87 12	90.26 9	95.96 3	88.42 18	92.76 10	92.29 31	90.84 43	96.62 12	98.44 27

APDe-MVS		94.31 8	94.30 10	94.33 7	97.57 12	98.06 13	95.79 7	91.98 9	95.50 9	92.19 4	95.25 4	87.97 21	92.93 7	93.01 24	91.02 41	95.52 43	99.29 8
Zhaojie Zeng, Yuesong Wang, Tao Guan: Matching Ambiguity-Resilient Multi-View Stereo via Adaptive Patch Deformation. Pattern Recognition
MCST-MVS		94.10 9	94.77 8	93.31 10	98.31 2	98.34 8	95.43 10	92.54 4	94.41 17	83.05 32	91.38 19	90.97 9	92.24 14	95.05 7	94.02 7	98.31 1	99.20 11

HPM-MVS++		94.04 10	94.96 6	92.96 12	97.93 5	97.71 19	94.65 15	91.01 13	95.91 7	87.43 16	93.52 11	92.63 5	92.29 13	94.22 14	92.34 18	94.47 68	98.37 28

NCCC		93.59 11	94.00 15	93.10 11	97.90 7	97.93 15	95.40 11	92.39 6	94.47 15	84.94 22	91.21 22	89.32 16	92.53 11	93.90 17	92.98 12	95.44 45	98.22 31

SMA-MVS		93.47 12	94.29 11	92.52 14	97.72 9	97.77 18	94.46 18	90.19 17	94.96 11	87.15 17	90.15 26	90.99 8	91.49 17	94.31 12	93.33 10	94.10 74	98.53 25
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
APD-MVS		93.47 12	93.44 18	93.50 9	97.06 15	97.09 27	95.27 13	91.47 11	95.71 8	89.57 11	93.66 9	86.28 27	92.81 9	92.06 34	90.70 44	94.83 65	98.60 22
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
SD-MVS		93.36 14	94.33 9	92.22 16	94.68 44	97.89 17	94.56 16	90.89 15	94.80 14	90.04 10	93.53 10	90.14 11	89.78 24	92.74 27	92.17 19	93.35 114	99.07 15
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.		93.07 15	93.53 17	92.53 13	94.23 47	97.54 21	94.75 14	89.87 18	95.26 10	89.20 13	93.16 12	88.19 20	92.15 15	91.79 39	89.65 64	94.99 60	99.16 13
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
DPM-MVS		92.86 16	93.19 20	92.47 15	95.78 35	97.40 22	97.39 1	92.56 3	92.88 25	81.84 39	81.31 40	92.95 4	91.21 18	96.54 1	97.33 1	96.01 33	93.94 120

MVS_0304		92.61 17	94.18 12	90.77 24	95.62 38	98.60 6	93.09 25	83.78 45	94.44 16	85.52 21	87.49 32	89.90 13	90.25 21	95.14 5	94.49 5	96.37 24	99.19 12

SteuartSystems-ACMMP		92.31 18	93.31 19	91.15 22	96.88 17	97.36 23	93.95 22	89.44 20	92.62 26	83.20 29	94.34 7	85.55 29	88.95 31	93.07 23	91.90 26	94.51 67	98.30 29
Skip Steuart: Steuart Systems R&D Blog.
ACMMP_NAP		92.16 19	92.91 23	91.28 21	96.95 16	97.36 23	93.66 23	89.23 22	93.33 20	83.71 27	90.53 23	86.84 24	90.39 20	93.30 22	91.56 32	93.74 86	97.43 48

HFP-MVS		92.02 20	92.13 25	91.89 19	97.16 14	96.46 39	93.57 24	87.60 25	93.79 19	88.17 15	93.15 13	83.94 39	91.19 19	90.81 49	89.83 59	93.66 90	96.94 64

train_agg		91.99 21	93.71 16	89.98 28	96.42 27	97.03 29	94.31 20	89.05 23	93.33 20	77.75 47	95.06 5	88.27 19	88.38 38	92.02 36	91.41 35	94.00 78	98.84 20

DeepC-MVS_fast	86.59 2	91.69 22	91.39 28	92.05 18	97.43 13	96.92 32	94.05 21	90.23 16	93.31 23	83.19 30	77.91 46	84.23 35	92.42 12	94.62 9	94.83 3	95.00 59	97.88 38
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.		91.29 23	93.11 22	89.18 33	87.81 91	96.21 45	92.51 33	83.83 44	94.24 18	83.77 26	91.87 18	89.62 14	90.07 22	90.40 54	90.31 49	97.09 6	99.10 14

ACMMPR		91.15 24	91.44 27	90.81 23	96.61 21	96.25 43	93.09 25	87.08 28	93.32 22	84.78 23	92.08 17	82.10 44	89.71 25	90.24 55	89.82 60	93.61 95	96.30 80

DeepPCF-MVS	86.71 1	91.00 25	94.05 14	87.43 44	95.58 39	98.17 10	86.22 80	88.59 24	97.01 2	76.77 57	85.11 36	88.90 17	87.29 46	95.02 8	94.69 4	90.15 190	99.48 6

TSAR-MVS + ACMM		90.98 26	93.18 21	88.42 38	95.69 36	96.73 34	94.52 17	86.97 31	92.99 24	76.32 59	92.31 16	86.64 25	84.40 73	92.97 25	92.02 23	92.62 140	98.59 23

MP-MVS		90.81 27	91.45 26	90.06 27	96.59 22	96.33 42	92.46 34	87.19 27	90.27 40	82.54 35	91.38 19	84.88 32	88.27 39	90.58 52	89.30 69	93.30 116	97.44 46
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
CP-MVS		90.57 28	90.68 30	90.44 25	96.13 29	95.90 51	92.77 31	86.86 32	92.12 30	84.19 24	89.18 29	82.37 42	89.43 28	89.65 67	88.43 81	93.27 117	97.13 56

MSLP-MVS++		90.33 29	88.82 39	92.10 17	96.52 25	95.93 47	94.35 19	86.26 33	88.37 55	89.24 12	75.94 54	82.60 41	89.71 25	89.45 72	92.17 19	96.51 16	97.24 53

CANet		89.98 30	90.42 34	89.47 32	94.13 48	98.05 14	91.76 39	83.27 48	90.87 37	81.90 38	72.32 62	84.82 33	88.42 36	94.52 10	93.78 9	97.34 4	98.58 24

PGM-MVS		89.97 31	90.64 32	89.18 33	96.53 24	95.90 51	93.06 27	82.48 56	90.04 42	80.37 41	92.75 15	80.96 49	88.93 32	89.88 63	89.08 73	93.69 89	95.86 86

PHI-MVS		89.88 32	92.75 24	86.52 53	94.97 41	97.57 20	89.99 50	84.56 40	92.52 28	69.72 98	90.35 25	87.11 23	84.89 64	91.82 38	92.37 17	95.02 58	97.51 44

CSCG		89.81 33	89.69 35	89.96 29	96.55 23	97.90 16	92.89 29	87.06 29	88.74 52	86.17 18	78.24 45	86.53 26	84.75 67	87.82 94	90.59 46	92.32 145	98.01 34

X-MVS		89.73 34	90.65 31	88.66 36	96.44 26	95.93 47	92.26 36	86.98 30	90.73 38	76.32 59	89.56 28	82.05 45	86.51 51	89.98 61	89.60 65	93.43 109	96.72 73

EPNet		89.30 35	90.89 29	87.44 43	95.67 37	96.81 33	91.13 42	83.12 50	91.14 34	76.31 63	87.60 31	80.40 53	84.45 70	92.13 33	91.12 40	93.96 79	97.01 60
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
DeepC-MVS	84.14 3	88.80 36	88.03 45	89.71 31	94.83 42	96.56 35	92.57 32	89.38 21	89.25 48	79.59 43	70.02 70	77.05 65	88.24 40	92.44 29	92.79 13	93.65 93	98.10 33
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
CDPH-MVS		88.76 37	90.43 33	86.81 49	96.04 31	96.53 38	92.95 28	85.95 35	90.36 39	67.93 104	85.80 35	80.69 50	83.82 78	90.81 49	91.85 29	94.18 72	96.99 61

3Dnovator+	81.14 5	88.59 38	87.49 48	89.88 30	95.83 34	96.45 41	91.94 38	82.41 57	87.09 61	85.94 20	62.80 103	85.37 30	89.46 27	91.51 41	91.89 28	93.72 87	97.30 51

ACMMP		88.48 39	88.71 40	88.22 40	94.61 45	95.53 57	90.64 46	85.60 37	90.97 35	78.62 45	89.88 27	74.20 79	86.29 53	88.16 91	86.37 102	93.57 96	95.86 86
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
AdaColmap		88.46 40	85.75 64	91.62 20	96.25 28	95.35 62	90.71 44	91.08 12	90.22 41	86.17 18	74.33 58	73.67 82	92.00 16	86.31 114	85.82 111	93.52 99	94.53 107

3Dnovator	80.58 8	88.20 41	86.53 54	90.15 26	96.86 18	96.46 39	91.97 37	83.06 51	85.16 66	83.66 28	62.28 108	82.15 43	88.98 30	90.99 46	92.65 14	96.38 23	96.03 83

CPTT-MVS		88.17 42	87.84 46	88.55 37	93.33 50	93.75 89	92.33 35	84.75 39	89.87 44	81.72 40	83.93 37	81.12 48	88.45 35	85.42 123	84.07 130	90.72 180	96.72 73

MVS_111021_HR		87.82 43	88.84 38	86.62 51	94.42 46	97.36 23	88.21 61	83.26 49	83.42 69	72.52 84	82.63 38	76.93 66	84.95 63	91.93 37	91.15 39	96.39 22	98.49 26

DELS-MVS		87.75 44	86.92 52	88.71 35	94.69 43	97.34 26	92.78 30	84.50 41	77.87 94	81.94 37	67.17 78	75.49 74	82.84 86	95.38 2	95.93 2	95.55 42	99.27 9
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
MVSTER		87.68 45	89.12 37	86.01 55	88.11 89	90.05 129	89.28 54	77.05 95	91.37 31	79.97 42	76.70 50	85.25 31	84.89 64	93.53 19	91.41 35	96.73 10	95.55 93

MVS_111021_LR		87.58 46	88.67 41	86.31 54	92.58 54	95.89 53	86.20 81	82.49 55	89.08 50	77.47 53	86.20 34	74.22 78	85.49 58	90.03 60	88.52 79	93.66 90	96.74 71

QAPM		87.06 47	86.46 55	87.75 41	96.63 20	97.09 27	91.71 40	82.62 54	80.58 82	71.28 90	66.04 85	84.24 34	87.01 47	89.93 62	89.91 58	97.26 5	97.44 46

PVSNet_BlendedMVS		86.98 48	87.05 50	86.90 46	93.03 51	96.98 30	86.57 74	81.82 59	89.78 45	82.78 33	71.54 64	66.07 115	80.73 99	93.46 20	91.97 24	96.45 19	99.53 4

PVSNet_Blended		86.98 48	87.05 50	86.90 46	93.03 51	96.98 30	86.57 74	81.82 59	89.78 45	82.78 33	71.54 64	66.07 115	80.73 99	93.46 20	91.97 24	96.45 19	99.53 4

ETV-MVS		86.94 50	89.49 36	83.95 75	87.28 98	95.61 56	83.58 111	76.37 102	92.59 27	73.20 76	80.35 41	76.42 69	87.38 45	92.20 32	90.45 48	95.90 34	98.83 21

SPE-MVS-test		86.72 51	88.35 42	84.83 65	91.78 60	96.03 46	81.71 122	76.71 96	91.19 33	77.12 56	77.64 48	75.63 73	87.59 44	90.82 48	89.11 71	94.06 76	97.99 36

CS-MVS		86.70 52	87.61 47	85.65 56	91.33 64	95.64 55	84.73 99	76.64 98	88.68 53	77.78 46	74.87 55	72.86 86	89.09 29	92.89 26	90.18 53	94.31 71	98.16 32

EC-MVSNet		86.42 53	88.31 43	84.20 71	86.61 109	94.08 82	86.20 81	72.18 135	89.06 51	76.02 64	74.48 57	80.47 52	88.90 33	92.03 35	90.07 56	95.30 47	98.00 35

OMC-MVS		86.38 54	86.21 60	86.57 52	92.30 56	94.35 81	87.60 65	83.51 47	92.32 29	77.37 54	72.27 63	77.83 58	86.59 50	87.62 96	85.95 108	92.08 149	93.11 134

HQP-MVS		86.17 55	87.35 49	84.80 66	91.41 63	92.37 107	91.05 43	84.35 43	88.52 54	64.21 111	87.05 33	68.91 102	84.80 66	89.12 75	88.16 85	92.96 131	97.31 50

sasdasda		85.93 56	86.26 58	85.54 57	88.94 77	95.44 58	89.56 51	76.01 104	87.83 56	77.70 48	76.43 51	68.66 104	87.80 42	87.02 101	91.51 33	93.25 118	96.95 62

canonicalmvs		85.93 56	86.26 58	85.54 57	88.94 77	95.44 58	89.56 51	76.01 104	87.83 56	77.70 48	76.43 51	68.66 104	87.80 42	87.02 101	91.51 33	93.25 118	96.95 62

MAR-MVS		85.65 58	86.30 57	84.88 64	95.51 40	95.89 53	86.50 76	76.71 96	89.23 49	68.59 101	70.93 68	74.49 76	88.55 34	89.40 73	90.30 50	93.42 110	93.88 125
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
PCF-MVS	82.38 4	85.52 59	84.41 69	86.81 49	91.51 62	96.23 44	90.27 47	89.81 19	77.87 94	70.67 94	69.20 72	77.86 56	85.55 57	85.92 119	86.38 101	93.03 128	97.43 48
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
CLD-MVS		85.43 60	84.24 72	86.83 48	87.69 94	93.16 98	90.01 49	82.72 53	87.17 60	79.28 44	71.43 67	65.81 118	86.02 54	87.33 99	86.96 95	95.25 53	97.83 40
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
OpenMVS	77.91 11	85.09 61	83.42 76	87.03 45	96.12 30	96.55 37	89.36 53	81.59 61	79.19 90	75.20 69	55.84 139	79.04 55	84.45 70	88.47 85	89.35 68	95.48 44	95.48 94

MGCFI-Net		85.07 62	85.99 61	83.99 73	88.81 80	95.23 67	89.06 56	75.74 107	87.40 59	70.72 93	75.99 53	68.44 106	86.51 51	86.83 105	91.24 37	93.11 125	96.78 70

TSAR-MVS + COLMAP		84.93 63	85.79 63	83.92 76	90.90 66	93.57 93	89.25 55	82.00 58	91.29 32	61.66 120	88.25 30	59.46 144	86.71 49	89.79 64	87.09 92	93.01 129	91.09 156

TAPA-MVS	80.99 7	84.83 64	84.42 68	85.31 60	91.89 59	93.73 91	88.53 60	82.80 52	89.99 43	69.78 97	71.53 66	75.03 75	85.47 59	86.26 115	84.54 125	93.39 112	89.90 166
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
PLC	81.02 6	84.81 65	81.81 98	88.31 39	93.77 49	90.35 122	88.80 58	84.47 42	86.76 62	82.17 36	66.56 81	71.01 94	88.41 37	85.48 121	84.28 128	92.26 147	88.21 179
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
EIA-MVS		84.75 66	86.43 56	82.79 83	86.88 104	95.36 61	82.84 118	76.39 101	87.61 58	71.03 91	74.33 58	71.12 93	85.16 60	89.69 66	88.70 78	94.40 69	98.23 30

CNLPA		84.72 67	82.14 92	87.73 42	92.85 53	93.83 88	84.70 100	85.07 38	90.90 36	83.16 31	56.28 135	71.53 90	88.14 41	84.19 128	84.00 134	92.48 142	94.26 114

MVS_Test		84.60 68	85.13 67	83.99 73	88.17 87	95.27 66	88.21 61	73.15 126	84.31 68	70.55 95	68.67 76	68.78 103	86.99 48	91.71 40	91.90 26	96.84 9	95.27 99

casdiffmvs_mvg		83.97 69	82.62 85	85.54 57	87.71 92	94.38 79	88.93 57	80.11 66	77.34 98	77.57 52	63.01 102	65.95 117	84.96 62	90.69 51	90.23 52	93.95 80	96.74 71
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
casdiffmvs		83.84 70	82.65 84	85.22 62	87.25 100	94.62 76	86.01 86	79.62 68	79.48 87	77.59 51	61.92 111	64.34 122	85.57 56	90.55 53	90.51 47	95.26 51	97.14 55
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
baseline		83.83 71	84.38 70	83.18 82	86.65 107	94.59 77	85.79 89	73.78 123	85.83 64	72.94 77	69.28 71	70.80 96	83.45 82	86.80 106	87.59 88	96.47 18	95.77 90

diffmvs		83.69 72	83.17 80	84.31 68	85.45 121	93.92 83	86.89 68	78.62 74	82.71 75	75.95 65	66.78 80	63.90 125	83.84 77	87.90 93	89.16 70	95.10 56	97.82 41
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
viewcassd2359sk11		83.64 73	82.24 90	85.27 61	87.13 101	94.82 72	86.47 77	79.81 67	76.49 106	77.69 50	64.03 96	63.90 125	84.40 73	89.49 70	90.26 51	96.12 27	96.68 76

CANet_DTU		83.33 74	86.59 53	79.53 109	88.88 79	94.87 70	86.63 73	68.85 158	85.45 65	50.54 166	77.86 47	69.94 99	85.62 55	92.63 28	90.88 42	96.63 11	94.46 108

DI_MVS_pp		83.32 75	82.53 87	84.25 70	86.26 115	93.66 92	90.23 48	77.16 94	77.05 103	74.06 73	53.74 148	74.33 77	83.61 81	91.40 43	89.82 60	94.17 73	97.73 42

viewmanbaseed2359cas		83.27 76	82.21 91	84.51 67	87.27 99	94.83 71	86.41 78	79.61 69	77.03 104	73.99 74	63.66 99	63.85 127	84.06 76	88.94 78	90.63 45	95.72 41	96.56 78

diffmvs_AUTHOR		83.09 77	82.28 89	84.04 72	85.22 125	93.85 87	86.75 69	78.41 77	79.81 85	75.32 68	64.61 90	63.57 129	83.62 80	87.60 97	88.86 77	94.93 61	97.68 43

viewdifsd2359ckpt13		82.93 78	81.75 101	84.30 69	87.00 103	94.76 73	85.59 91	79.57 70	76.32 108	74.15 72	62.74 105	62.67 131	84.44 72	89.49 70	90.15 54	95.06 57	96.13 82

baseline1		82.63 79	82.02 93	83.34 81	88.30 86	91.89 111	88.03 64	80.86 63	75.05 113	65.96 106	64.27 93	72.20 88	80.01 103	91.32 44	89.56 66	96.90 8	89.85 167

PVSNet_Blended_VisFu		82.55 80	83.70 75	81.21 96	89.66 70	95.15 69	82.41 119	77.36 93	72.53 132	73.64 75	61.15 114	77.19 64	70.35 163	91.31 45	89.72 63	93.84 82	98.85 19

ET-MVSNet_ETH3D		82.37 81	85.68 65	78.51 119	62.90 223	94.66 74	87.06 67	73.57 124	83.13 71	61.52 122	78.37 44	76.01 71	89.99 23	84.14 129	89.03 74	96.03 32	94.42 109

PMMVS		82.26 82	85.48 66	78.51 119	85.92 118	91.92 110	78.30 152	70.77 143	86.30 63	61.11 124	82.46 39	70.88 95	84.70 68	88.05 92	84.78 121	90.24 189	93.98 118

ACMP	79.58 9	82.23 83	81.82 97	82.71 84	88.15 88	90.95 119	85.23 95	78.52 76	81.70 77	72.52 84	78.41 43	60.63 139	80.48 101	82.88 141	83.44 138	91.37 166	94.70 105
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
viewmambaseed2359dif		82.18 84	80.85 103	83.72 79	85.36 123	93.20 97	86.29 79	77.89 83	77.11 102	76.48 58	62.40 107	63.42 130	83.28 84	84.01 131	87.92 86	93.04 127	97.91 37

CHOSEN 280x420		82.15 85	85.87 62	77.80 124	86.54 111	93.42 95	81.74 121	59.96 202	78.99 92	63.99 112	74.50 56	83.95 38	80.99 94	89.53 69	85.01 116	93.56 98	95.71 92

LGP-MVS_train		82.12 86	82.57 86	81.59 90	89.26 74	90.23 125	88.76 59	78.05 78	81.26 79	61.64 121	79.52 42	62.11 134	79.59 107	85.20 124	84.68 123	92.27 146	95.02 103

FMVSNet3		81.93 87	81.98 94	81.88 89	79.49 162	87.02 145	88.15 63	72.57 129	83.02 72	72.63 81	56.55 131	73.48 83	82.34 90	91.49 42	91.20 38	96.07 28	91.13 155

test2506		81.91 88	81.78 100	82.06 88	89.09 75	95.32 63	84.61 102	77.54 87	74.61 117	68.77 100	63.80 98	67.53 109	77.09 116	90.19 57	89.01 75	95.27 48	92.00 147

thisisatest0530		81.67 89	84.27 71	78.63 115	85.53 119	93.88 86	81.77 120	73.84 120	81.35 78	63.85 114	68.79 74	77.64 60	73.02 144	88.73 83	85.73 112	93.76 85	93.80 129

viewmacassd2359aftdt		81.60 90	79.90 110	83.58 80	86.67 106	94.36 80	86.02 85	79.17 71	70.40 139	71.64 88	58.95 120	62.12 133	82.55 88	87.08 100	90.14 55	95.41 46	96.24 81

tttt0517		81.51 91	84.12 74	78.47 121	85.33 124	93.74 90	81.42 125	73.84 120	81.21 80	63.59 115	68.73 75	77.46 63	73.02 144	88.47 85	85.73 112	93.63 94	93.49 133

FA-MVS(training)		81.41 92	81.98 94	80.76 102	87.58 95	94.59 77	83.09 113	61.18 199	79.80 86	74.74 70	58.46 123	69.76 100	82.12 91	88.90 79	87.00 93	95.83 38	95.33 96

OPM-MVS		81.34 93	78.18 119	85.02 63	91.27 65	91.78 112	90.66 45	83.62 46	62.39 164	65.91 107	63.35 100	64.33 123	85.03 61	87.77 95	85.88 110	93.66 90	91.75 151
Ray L. Khuboni, Hongjun Xu: Octagram Propagation Matching for Multi-Scale View Stereopsis (OPM-MVS).
baseline2		81.21 94	83.36 79	78.70 113	83.22 140	92.71 100	80.32 133	74.25 119	80.39 83	63.94 113	68.89 73	68.44 106	74.67 130	89.61 68	86.68 99	95.83 38	96.81 69

IS_MVSNet		80.92 95	84.14 73	77.16 127	87.43 96	93.90 85	80.44 129	74.64 113	75.05 113	61.10 125	65.59 87	76.89 67	67.39 171	90.88 47	90.05 57	91.95 153	96.62 77

ACMM	78.09 10	80.91 96	78.39 116	83.86 77	89.61 73	87.71 142	85.16 96	80.67 65	79.04 91	74.18 71	63.82 97	60.84 138	82.59 87	84.33 126	83.59 137	90.96 174	89.39 172
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
EPP-MVSNet		80.82 97	82.79 82	78.52 117	86.31 114	92.37 107	79.83 136	74.51 114	73.79 124	64.46 110	67.01 79	80.63 51	74.33 133	85.63 120	84.35 127	91.68 159	95.79 89

CostFormer		80.72 98	81.81 98	79.44 111	86.50 112	91.65 113	84.31 104	59.84 203	80.86 81	72.69 79	62.46 106	73.74 80	79.93 104	82.58 145	84.50 126	93.37 113	96.90 67

GBi-Net		80.72 98	80.49 104	81.00 99	78.18 166	86.19 159	86.73 70	72.57 129	83.02 72	72.63 81	56.55 131	73.48 83	80.99 94	86.57 108	86.83 96	94.89 62	90.77 159

test1		80.72 98	80.49 104	81.00 99	78.18 166	86.19 159	86.73 70	72.57 129	83.02 72	72.63 81	56.55 131	73.48 83	80.99 94	86.57 108	86.83 96	94.89 62	90.77 159

UGNet		80.71 101	83.09 81	77.93 123	87.02 102	92.71 100	80.28 134	76.53 99	73.83 123	71.35 89	70.07 69	73.71 81	58.93 191	87.39 98	86.97 94	93.48 106	96.94 64
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
CHOSEN 1792x2688		80.23 102	79.16 113	81.48 92	91.97 57	96.56 35	86.18 83	75.40 110	76.17 109	61.32 123	37.43 209	61.08 137	76.52 122	92.35 30	91.64 31	97.46 3	98.86 18

thres100view900		79.83 103	77.79 123	82.21 85	88.42 83	93.54 94	87.07 66	81.11 62	70.15 140	61.01 126	56.65 129	51.22 163	81.78 92	89.77 65	85.95 108	93.84 82	97.26 52

Effi-MVS+		79.80 104	80.04 106	79.52 110	85.53 119	93.31 96	85.28 93	70.68 145	74.15 119	58.79 135	62.03 110	60.51 140	83.37 83	88.41 87	86.09 107	93.49 105	95.80 88

ECVR-MVS		79.76 105	78.27 117	81.50 91	89.09 75	95.32 63	84.61 102	77.54 87	74.61 117	65.38 108	50.22 160	56.31 157	77.09 116	90.19 57	89.01 75	95.27 48	92.25 142

DCV-MVSNet		79.76 105	79.17 112	80.44 105	84.65 129	84.51 183	84.20 105	72.36 134	75.17 112	70.81 92	66.21 84	66.56 112	80.99 94	82.89 140	84.56 124	89.65 195	94.30 113

FC-MVSNet-train		79.54 107	78.20 118	81.09 98	86.55 110	88.63 138	79.96 135	78.53 75	70.90 137	68.24 102	65.87 86	56.45 156	80.29 102	86.20 117	84.08 129	92.97 130	95.31 98

test-LLR		79.52 108	83.42 76	74.97 136	81.79 145	91.26 114	76.17 173	70.57 146	77.71 96	52.14 153	66.26 82	77.47 61	73.10 140	87.02 101	87.16 90	96.05 30	97.02 58

FMVSNet2		79.24 109	78.14 120	80.53 104	78.18 166	86.19 159	86.73 70	71.91 136	72.97 127	70.48 96	50.63 158	66.56 112	80.99 94	90.10 59	89.77 62	94.89 62	90.77 159

TESTMET0.1,1		79.15 110	83.42 76	74.18 142	79.81 160	91.26 114	76.17 173	67.83 171	77.71 96	52.14 153	66.26 82	77.47 61	73.10 140	87.02 101	87.16 90	96.05 30	97.02 58

tfpn200view9		79.05 111	77.21 127	81.18 97	88.42 83	92.55 105	85.12 97	77.94 80	70.15 140	61.01 126	56.65 129	51.22 163	81.11 93	88.23 88	84.80 120	93.50 104	96.90 67

test1111		78.99 112	77.77 124	80.42 106	88.64 81	95.31 65	83.39 112	77.67 85	72.76 130	61.91 118	49.58 163	55.59 159	75.67 127	90.23 56	89.09 72	95.23 54	91.83 150

viewdifsd2359ckpt11		78.83 113	76.68 131	81.33 94	84.03 137	90.13 127	80.89 127	77.43 91	70.01 142	75.72 66	60.97 115	59.03 148	79.67 105	79.81 171	81.58 160	90.25 187	95.22 100

viewmsd2359difaftdt		78.82 114	76.68 131	81.33 94	84.03 137	90.13 127	80.89 127	77.43 91	70.00 143	75.68 67	60.97 115	59.01 149	79.67 105	79.82 170	81.58 160	90.25 187	95.22 100

PatchMatch-RL		78.75 115	76.47 136	81.41 93	88.53 82	91.10 116	78.09 153	77.51 90	77.33 99	71.98 86	64.38 92	48.10 176	82.55 88	84.06 130	82.35 147	89.78 192	87.97 181

LS3D		78.72 116	75.79 141	82.15 86	91.91 58	89.39 134	83.66 109	85.88 36	76.81 105	59.22 134	57.67 126	58.53 150	83.72 79	82.07 150	81.63 158	88.50 203	84.39 192

thres200		78.69 117	76.71 130	80.99 101	88.35 85	92.56 103	86.03 84	77.94 80	66.27 149	60.66 128	56.08 136	51.11 165	79.45 108	88.23 88	85.54 115	93.52 99	97.20 54

Anonymous20231211		78.61 118	75.57 144	82.15 86	84.43 133	90.26 123	84.08 107	77.68 84	71.09 135	72.90 78	39.24 203	66.21 114	84.23 75	82.15 148	84.04 131	89.61 196	96.03 83

IB-MVS	74.10 12	78.52 119	78.51 115	78.52 117	90.15 68	95.39 60	71.95 193	77.53 89	74.95 115	77.25 55	58.93 121	55.92 158	58.37 193	79.01 177	87.89 87	95.88 36	97.47 45
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
EPNet_dtu		78.49 120	81.96 96	74.45 141	92.57 55	88.74 137	82.98 114	78.83 73	83.28 70	44.64 197	77.40 49	67.73 108	53.98 202	85.44 122	84.91 117	93.71 88	86.22 187
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
thres400		78.39 121	76.39 137	80.73 103	88.02 90	92.94 99	84.77 98	78.88 72	65.20 157	59.70 132	55.20 142	50.85 166	79.45 108	88.81 80	84.81 119	93.57 96	96.91 66

UA-Net		78.30 122	80.92 102	75.25 135	87.42 97	92.48 106	79.54 139	75.49 109	60.47 168	60.52 129	68.44 77	84.08 37	57.54 195	88.54 84	88.45 80	90.96 174	83.97 194

Vis-MVSNet (Re-imp)		78.28 123	82.68 83	73.16 153	86.64 108	92.68 102	78.07 154	74.48 115	74.05 120	53.47 146	64.22 94	76.52 68	54.28 198	88.96 77	88.29 83	92.03 151	94.00 117

MSDG		78.11 124	73.17 157	83.86 77	91.78 60	86.83 147	85.25 94	86.02 34	72.84 129	69.69 99	51.43 155	54.00 161	77.61 112	81.95 153	82.27 149	92.83 136	82.91 199

HyFIR lowres test		78.08 125	76.81 128	79.56 108	90.77 67	94.64 75	82.97 115	69.85 151	69.81 144	59.53 133	33.52 215	64.66 119	78.97 110	88.77 82	88.38 82	95.27 48	97.86 39

GeoE		78.04 126	77.52 126	78.65 114	84.51 131	90.84 120	80.94 126	69.24 156	72.86 128	66.06 105	53.45 149	60.46 141	77.37 113	84.20 127	84.85 118	93.78 84	96.00 85

test-mter		77.90 127	82.44 88	72.60 158	78.52 164	90.24 124	73.85 186	65.31 186	76.37 107	51.29 157	65.58 88	75.94 72	71.36 154	85.98 118	86.26 103	95.26 51	96.71 75

thres600view7		77.66 128	75.67 142	79.98 107	87.71 92	92.56 103	83.79 108	77.94 80	64.41 159	58.69 136	54.32 147	50.54 167	78.23 111	88.23 88	83.06 141	93.52 99	96.55 79

MS-PatchMatch		77.47 129	76.48 135	78.63 115	89.89 69	90.42 121	85.42 92	69.53 153	70.79 138	60.43 130	50.05 161	70.62 98	70.66 160	86.71 107	82.54 144	95.86 37	84.23 193

Fast-Effi-MVS+		77.37 130	76.68 131	78.17 122	82.84 142	89.94 130	81.47 124	68.01 167	72.99 126	60.26 131	55.07 143	53.20 162	82.99 85	86.47 113	86.12 106	93.46 107	92.98 137

dmvs_re		77.25 131	75.86 140	78.86 112	81.08 151	89.36 135	84.15 106	80.73 64	73.02 125	55.58 142	58.33 124	48.97 172	75.32 128	83.92 134	86.25 104	96.29 26	91.20 154

Vis-MVSNet		77.24 132	79.99 109	74.02 143	84.62 130	93.92 83	80.33 132	72.55 132	62.58 163	55.25 144	64.45 91	69.49 101	57.00 196	88.78 81	88.21 84	94.36 70	92.54 139
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
MDTV_nov1_ep13		77.20 133	80.04 106	73.90 145	82.22 143	90.14 126	79.25 143	61.52 197	78.63 93	56.98 137	65.52 89	72.80 87	73.05 142	80.93 161	83.20 139	90.36 184	89.05 175

EPMVS		77.16 134	79.08 114	74.92 137	86.73 105	91.98 109	78.62 148	55.44 211	79.43 88	56.59 139	61.24 113	70.73 97	76.97 119	80.59 164	81.43 166	95.15 55	88.17 180

tpm cat1		76.93 135	76.19 139	77.79 125	85.08 128	88.58 139	82.96 116	59.33 204	75.72 111	72.64 80	51.25 156	64.41 121	75.74 126	77.90 185	80.10 182	90.97 173	95.35 95

PatchmatchNet		76.85 136	80.03 108	73.15 154	84.08 135	91.04 118	77.76 158	55.85 210	79.43 88	52.74 151	62.08 109	76.02 70	74.56 131	79.92 169	81.41 167	93.92 81	90.29 164
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
IterMVS-LS		76.80 137	76.33 138	77.35 126	84.07 136	84.11 184	81.54 123	68.52 160	66.17 150	61.74 119	57.84 125	64.31 124	74.88 129	83.48 137	86.21 105	93.34 115	92.16 144
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
CDS-MVSNet		76.57 138	76.78 129	76.32 130	80.94 153	89.75 131	82.94 117	72.64 128	59.01 174	62.95 117	58.60 122	62.67 131	66.91 173	86.26 115	87.20 89	91.57 161	93.97 119
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
SCA		76.41 139	79.90 110	72.35 162	84.26 134	85.24 174	75.57 180	54.56 213	79.95 84	52.72 152	64.22 94	77.84 57	73.73 137	80.48 165	81.37 168	93.25 118	90.20 165

tpmrst		76.27 140	77.65 125	74.66 139	86.13 117	89.53 133	79.31 142	54.91 212	77.19 101	56.27 140	55.87 138	64.58 120	77.25 114	80.85 162	80.21 179	94.07 75	95.32 97

dps		75.76 141	75.02 146	76.63 129	84.51 131	88.12 140	77.51 159	58.33 206	75.91 110	71.98 86	57.37 127	57.85 151	76.81 121	77.89 186	78.40 191	90.63 181	89.63 169

CR-MVSNet		74.84 142	77.91 121	71.26 175	81.77 147	85.52 170	78.32 150	54.14 215	74.05 120	51.09 160	50.00 162	71.38 92	70.77 158	86.48 111	84.03 132	91.46 165	93.92 122

Effi-MVS+-dtu		74.57 143	74.60 150	74.53 140	81.38 149	86.74 149	80.39 131	67.70 172	67.36 148	53.06 147	59.86 118	57.50 152	75.84 125	80.19 167	78.62 189	88.79 202	91.95 149

RPSCF		74.27 144	73.24 156	75.48 134	81.01 152	80.18 206	76.24 172	72.37 133	74.84 116	68.24 102	72.47 61	67.39 110	73.89 134	71.05 210	69.38 218	81.14 223	77.37 211

FMVSNet1		74.26 145	71.95 162	76.95 128	74.28 198	83.94 186	83.61 110	69.99 149	57.08 180	65.08 109	42.39 192	57.41 153	76.98 118	86.57 108	86.83 96	91.77 158	89.42 170

GA-MVS		73.62 146	74.52 151	72.58 159	79.93 158	89.29 136	78.02 155	71.67 139	60.79 167	42.68 201	54.41 146	49.07 171	70.07 164	89.39 74	86.55 100	93.13 124	92.12 145

Fast-Effi-MVS+-dtu		73.56 147	75.32 145	71.50 171	80.35 155	86.83 147	79.72 137	58.07 207	67.64 147	44.83 194	60.28 117	54.07 160	73.59 139	81.90 155	82.30 148	92.46 143	94.18 115

tpm		73.50 148	74.85 147	71.93 165	83.19 141	86.84 146	78.61 149	55.91 209	65.64 152	48.90 173	56.30 134	61.09 136	72.31 146	79.10 176	80.61 178	92.68 138	94.35 112

RPMNet		73.46 149	77.85 122	68.34 185	81.71 148	85.52 170	73.83 187	50.54 222	74.05 120	46.10 188	53.03 152	71.91 89	66.31 175	83.55 135	82.18 151	91.55 163	94.71 104

USDC		73.43 150	72.31 160	74.73 138	80.86 154	86.21 157	80.42 130	71.83 138	71.69 134	46.94 181	59.60 119	42.58 197	76.47 123	82.66 144	81.22 171	91.88 155	82.24 205

pmmvs4		73.38 151	71.53 165	75.55 133	75.95 184	85.24 174	77.25 163	71.59 140	71.03 136	63.10 116	49.09 168	44.22 187	73.73 137	82.04 151	80.18 180	91.68 159	88.89 177

UniMVSNet_NR-MVSNet		73.11 152	72.59 158	73.71 148	76.90 175	86.58 153	77.01 164	75.82 106	65.59 153	48.82 174	50.97 157	48.42 174	71.61 150	79.19 175	83.03 142	92.11 148	94.37 110

FMVSNet5		72.83 153	73.89 154	71.59 169	67.42 216	76.28 214	75.88 177	63.74 191	77.27 100	54.59 145	53.32 150	71.48 91	73.85 135	81.95 153	81.69 156	94.06 76	75.20 216

PatchT		72.66 154	76.58 134	68.09 187	79.02 163	86.09 163	59.81 215	51.78 220	72.00 133	51.09 160	46.84 172	66.70 111	70.77 158	86.48 111	84.03 132	96.07 28	93.92 122

ACMH	71.22 14	72.65 155	70.13 170	75.59 132	86.19 116	86.14 162	75.76 178	77.63 86	54.79 188	46.16 187	53.28 151	47.28 178	77.24 115	78.91 178	81.18 172	90.57 182	89.33 173
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
IterMVS		72.43 156	74.05 152	70.55 179	80.34 156	81.17 200	77.44 160	61.00 201	63.57 162	46.82 183	55.88 137	59.09 147	65.03 177	83.15 138	83.83 135	92.67 139	91.65 152
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
ACMH+	72.14 13	72.38 157	69.34 177	75.93 131	85.21 126	84.89 178	76.96 167	76.04 103	59.76 169	51.63 156	50.37 159	48.69 173	76.90 120	76.06 194	78.69 187	88.85 201	86.90 185

DU-MVS		72.19 158	71.35 166	73.17 152	75.95 184	86.02 164	77.01 164	74.42 116	65.39 155	48.82 174	49.10 166	42.81 195	71.61 150	78.67 179	83.10 140	91.22 169	94.37 110

IterMVS-SCA-FT		72.18 159	73.96 153	70.11 181	80.15 157	81.11 201	77.42 161	61.09 200	63.67 161	46.73 184	55.77 140	59.15 146	63.95 180	82.83 142	83.70 136	91.31 167	91.49 153

UniMVSNet (Re)		72.12 160	72.28 161	71.93 165	76.77 176	87.38 144	75.73 179	73.51 125	65.76 151	50.24 168	48.65 169	46.49 179	63.85 181	80.10 168	82.47 145	91.49 164	95.13 102

ADS-MVSNet		72.11 161	73.72 155	70.24 180	81.24 150	86.59 152	74.75 183	50.56 221	72.58 131	49.17 171	55.40 141	61.46 135	73.80 136	76.01 195	78.14 192	91.93 154	85.86 188

gg-mvs-nofinetune		72.10 162	74.79 148	68.97 184	83.31 139	95.22 68	85.66 90	48.77 223	35.68 225	22.17 232	30.49 218	77.73 59	76.37 124	94.30 13	93.03 11	97.55 2	97.05 57

TAMVS		72.06 163	71.76 164	72.41 161	76.68 177	88.12 140	74.82 182	68.09 165	53.52 193	56.91 138	52.94 153	56.93 155	66.91 173	81.37 158	82.44 146	91.07 171	86.99 184

v2v482		71.73 164	69.80 172	73.99 144	75.88 188	86.66 151	79.58 138	71.90 137	57.58 178	50.41 167	45.35 176	43.24 193	73.05 142	79.69 172	82.18 151	93.08 126	93.87 126

test0.0.03 1		71.70 165	74.68 149	68.23 186	81.79 145	83.81 187	68.64 197	70.57 146	68.81 146	43.47 198	62.77 104	60.09 143	51.77 209	82.48 146	81.67 157	93.16 122	83.13 197

V42		71.58 166	70.11 171	73.30 151	75.66 191	86.68 150	79.17 145	69.92 150	59.29 173	52.80 150	44.36 180	45.66 181	68.83 165	79.48 174	81.49 163	93.44 108	93.82 128

NR-MVSNet		71.47 167	71.11 167	71.90 167	77.73 171	86.02 164	76.88 168	74.42 116	65.39 155	46.09 189	49.10 166	39.87 210	64.27 179	81.40 157	82.24 150	91.99 152	93.75 130

v8		71.42 168	69.69 173	73.43 150	76.45 180	85.12 177	79.53 140	67.47 175	59.34 172	52.90 149	44.60 178	45.82 180	71.05 156	79.56 173	81.45 165	93.17 121	91.96 148

TranMVSNet+NR-MVSNet		71.12 169	70.24 169	72.15 163	76.01 183	84.80 180	76.55 170	75.65 108	61.99 165	45.29 192	48.42 170	43.07 194	67.55 169	78.28 182	82.83 143	91.85 156	92.29 140

v10		70.97 170	69.44 174	72.75 155	75.90 187	84.58 182	79.43 141	66.45 180	58.07 176	49.93 169	43.87 186	43.68 188	71.91 148	82.04 151	81.70 155	92.89 134	92.11 146

v1144		70.93 171	69.42 176	72.70 156	75.48 192	86.26 155	79.22 144	69.39 155	55.61 185	48.05 179	43.47 187	42.55 198	71.51 152	82.11 149	81.74 154	92.56 141	94.17 116

thisisatest0515		70.62 172	71.94 163	69.07 183	76.48 179	85.59 169	68.03 198	68.02 166	59.70 170	52.94 148	52.19 154	50.36 168	58.10 194	83.15 138	81.63 158	90.87 177	90.99 157

Baseline_NR-MVSNet		70.61 173	68.87 180	72.65 157	75.95 184	80.49 204	75.92 176	74.75 112	65.10 158	48.78 176	41.28 198	44.28 186	68.45 166	78.67 179	79.64 183	92.04 150	92.62 138

v148		70.34 174	68.46 183	72.54 160	76.04 182	86.38 154	74.83 181	72.73 127	55.88 184	55.26 143	43.32 189	43.49 189	64.52 178	76.93 192	80.11 181	91.85 156	93.11 134

v1192		70.32 175	68.77 181	72.12 164	74.76 194	85.62 168	78.73 146	68.53 159	55.08 187	46.34 186	42.39 192	40.67 205	71.90 149	82.27 147	81.53 162	92.43 144	93.86 127

v144192		70.10 176	68.55 182	71.90 167	74.55 195	85.67 167	77.81 156	68.22 164	54.65 189	46.91 182	42.76 190	41.27 202	70.95 157	80.48 165	81.11 176	92.96 131	93.90 124

pmmvs5		70.01 177	69.31 178	70.82 178	75.80 190	86.26 155	72.94 188	67.91 168	53.84 192	47.22 180	47.31 171	41.47 201	67.61 168	83.93 133	81.93 153	93.42 110	90.42 163

COLMAP_ROB	66.31 15	69.91 178	66.61 188	73.76 146	86.44 113	82.76 191	76.59 169	76.46 100	63.82 160	50.92 164	45.60 175	49.13 170	65.87 176	74.96 200	74.45 208	86.30 213	75.57 215
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
v1921920		69.85 179	68.38 184	71.58 170	74.35 196	85.39 172	77.78 157	67.88 170	54.64 190	45.39 191	42.11 195	39.97 209	71.10 155	81.68 156	81.17 174	92.96 131	93.69 132

pm-mvs1		69.62 180	68.07 186	71.44 172	77.21 173	85.32 173	76.11 175	71.05 141	46.55 213	51.17 159	41.83 196	48.20 175	61.81 187	84.00 132	81.14 175	91.28 168	89.42 170

UniMVSNet_ETH3D		69.49 181	65.86 190	73.72 147	76.51 178	85.88 166	78.65 147	70.52 148	48.08 210	55.71 141	37.64 206	40.56 206	71.38 153	75.05 199	81.49 163	89.57 198	92.29 140

tfpnnormal		69.29 182	65.58 191	73.62 149	79.87 159	84.82 179	76.97 166	75.12 111	45.29 214	49.03 172	35.57 213	37.20 218	68.02 167	82.70 143	81.24 170	92.69 137	92.20 143

v1240		69.28 183	67.82 187	71.00 177	74.09 200	85.13 176	76.54 171	67.28 177	53.17 194	44.70 195	41.55 197	39.38 211	70.51 162	81.29 159	81.18 172	92.88 135	93.02 136

CVMVSNet		68.95 184	70.79 168	66.79 193	79.69 161	83.75 188	72.05 192	70.90 142	56.20 182	36.30 213	54.94 145	59.22 145	54.03 201	78.33 181	78.65 188	87.77 209	84.44 191

MIMVSNet		68.66 185	69.43 175	67.76 188	64.92 220	84.68 181	74.16 184	54.10 217	60.85 166	51.27 158	39.47 202	49.48 169	67.48 170	84.86 125	85.57 114	94.63 66	81.10 206

TDRefinement		67.82 186	64.91 197	71.22 176	82.08 144	81.45 196	77.42 161	73.79 122	59.62 171	48.35 178	42.35 194	42.40 199	60.87 189	74.69 201	74.64 207	84.83 217	79.20 209

anonymousdsp		67.61 187	68.94 179	66.04 194	71.44 212	83.97 185	66.45 202	63.53 193	50.54 203	42.42 202	49.39 164	45.63 182	62.84 184	77.99 184	81.34 169	89.59 197	93.75 130

TinyColmap		67.16 188	63.51 204	71.42 173	77.94 169	79.54 210	72.80 189	69.78 152	56.58 181	45.52 190	44.53 179	33.53 223	74.45 132	76.91 193	77.06 198	88.03 208	76.41 212

FC-MVSNet-test		67.04 189	72.47 159	60.70 211	76.92 174	81.41 197	61.52 212	69.45 154	65.58 154	26.74 227	61.79 112	60.40 142	41.17 219	77.60 188	77.78 194	88.41 204	82.70 201

TransMVSNet (Re)		66.87 190	64.30 199	69.88 182	78.32 165	81.35 199	73.88 185	74.34 118	43.19 218	45.20 193	40.12 200	42.37 200	55.97 197	80.85 162	79.15 184	91.56 162	83.06 198

CMPMVS	50.59 17	66.74 191	62.72 208	71.42 173	85.40 122	89.72 132	72.69 190	70.72 144	51.24 199	51.75 155	38.91 204	44.40 184	63.74 182	70.84 212	71.52 212	84.19 218	72.45 220
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
v7n		66.43 192	65.51 192	67.51 189	71.63 211	83.10 189	70.89 196	65.02 187	50.13 206	44.68 196	39.59 201	38.77 212	62.57 185	77.59 189	78.91 185	90.29 186	90.44 162

EG-PatchMatch MVS		66.23 193	65.20 194	67.43 190	77.74 170	86.20 158	72.51 191	63.68 192	43.95 216	43.44 199	36.22 212	45.43 183	54.04 200	81.00 160	80.95 177	93.15 123	82.67 202

WR-MVS		64.98 194	66.59 189	63.09 204	74.34 197	82.68 192	64.98 208	69.17 157	54.42 191	36.18 214	44.32 181	44.35 185	44.65 212	73.60 202	77.83 193	89.21 200	88.96 176

gm-plane-assit		64.86 195	68.15 185	61.02 210	76.44 181	68.29 224	41.60 229	53.37 218	34.68 227	26.19 229	33.22 216	57.09 154	71.97 147	95.12 6	93.97 8	96.54 15	94.66 106

CP-MVSNet		64.84 196	64.97 195	64.69 199	72.09 207	81.04 202	66.66 201	67.53 174	52.45 196	37.40 209	44.00 185	38.37 214	53.54 204	72.26 206	76.93 199	90.94 176	89.75 168

MDTV_nov1_ep13_2view		64.72 197	64.94 196	64.46 200	71.14 213	81.94 195	67.53 199	54.54 214	55.92 183	43.29 200	44.02 184	43.27 192	59.87 190	71.85 208	74.77 206	90.36 184	82.82 200

MVS-HIRNet		64.63 198	64.03 203	65.33 196	75.01 193	82.84 190	58.54 219	52.10 219	55.42 186	49.29 170	29.83 221	43.48 190	66.97 172	78.28 182	78.81 186	90.07 191	79.52 208

pmnet_mix02		64.58 199	64.11 202	65.12 197	74.16 199	80.17 207	63.24 210	67.91 168	57.87 177	41.69 203	45.86 174	40.99 204	53.97 203	69.92 215	71.67 211	89.77 193	82.29 204

LTVRE_ROB	63.07 16	64.49 200	63.16 207	66.04 194	77.47 172	82.64 193	70.98 195	65.02 187	34.01 228	29.61 223	49.12 165	35.58 222	70.57 161	75.10 198	78.45 190	82.60 221	87.24 183
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
PEN-MVS		64.35 201	64.29 200	64.42 201	72.67 203	79.83 208	66.97 200	68.24 163	51.21 200	35.29 216	44.09 182	38.51 213	52.36 207	71.06 209	77.65 195	90.99 172	87.68 182

pmmvs6		64.24 202	61.77 212	67.12 191	72.39 206	81.39 198	71.33 194	65.95 185	36.05 224	48.48 177	30.55 217	43.45 191	58.75 192	77.88 187	76.36 202	85.83 214	86.70 186

pmmvs-eth3d		64.24 202	61.96 210	66.90 192	66.35 217	76.04 216	66.09 204	66.31 182	52.59 195	50.94 163	37.61 207	32.79 225	62.43 186	75.78 196	75.48 204	89.27 199	83.39 196

PS-CasMVS		64.22 204	64.19 201	64.25 202	71.86 209	80.67 203	66.42 203	67.43 176	50.64 202	36.48 211	42.60 191	37.46 217	52.56 206	71.98 207	76.69 201	90.76 178	89.29 174

WR-MVS_H		64.14 205	65.36 193	62.71 206	72.47 205	82.33 194	65.13 205	66.99 178	51.81 198	36.47 212	43.33 188	42.77 196	43.99 214	72.41 205	75.99 203	91.20 170	88.86 178

SixPastTwentyTwo		63.75 206	63.42 205	64.13 203	72.91 202	80.34 205	61.29 213	63.90 190	49.58 207	40.42 205	54.99 144	37.13 219	60.90 188	68.46 216	70.80 213	85.37 216	82.65 203

PM-MVS		63.52 207	62.51 209	64.70 198	64.79 222	76.08 215	65.07 206	62.08 195	58.13 175	46.56 185	44.98 177	31.31 227	62.89 183	72.58 204	69.93 217	86.81 211	84.55 190

DTE-MVSNet		63.26 208	63.41 206	63.08 205	72.59 204	78.56 211	65.03 207	68.28 162	50.53 204	32.38 220	44.03 183	37.79 216	49.48 210	70.83 213	76.73 200	90.73 179	85.42 189

testgi		63.11 209	64.88 198	61.05 209	75.83 189	78.51 212	60.42 214	66.20 183	48.77 208	34.56 217	56.96 128	40.35 207	40.95 220	77.46 190	77.22 197	88.37 206	74.86 218

GG-mvs-BLEND		62.08 210	88.31 43	31.46 226	0.16 238	98.10 12	91.57 41	0.09 234	85.07 67	0.21 239	73.90 60	83.74 40	0.19 236	88.98 76	89.39 67	96.58 13	99.02 16

Anonymous20231206		62.05 211	61.83 211	62.30 208	72.09 207	77.84 213	63.10 211	67.62 173	50.20 205	36.68 210	29.59 222	37.05 220	43.90 215	77.33 191	77.31 196	90.41 183	83.49 195

N_pmnet		60.52 212	58.83 215	62.50 207	68.97 215	75.61 217	59.72 217	66.47 179	51.90 197	41.26 204	35.42 214	35.63 221	52.25 208	67.07 219	70.08 216	86.35 212	76.10 213

EU-MVSNet		58.73 213	60.92 213	56.17 214	66.17 219	72.39 220	58.85 218	61.24 198	48.47 209	27.91 225	46.70 173	40.06 208	39.07 221	68.27 217	70.34 215	83.77 219	80.23 207

test20.03		57.93 214	59.22 214	56.44 213	71.84 210	73.78 219	53.55 223	65.96 184	43.02 219	28.46 224	37.50 208	38.17 215	30.41 225	75.25 197	74.42 209	88.41 204	72.37 221

MDA-MVSNet-bldmvs		54.99 215	52.66 220	57.71 212	52.74 228	74.87 218	55.61 220	68.41 161	43.65 217	32.54 218	37.93 205	22.11 234	54.11 199	48.85 227	67.34 219	82.85 220	73.88 219

FE-MVSNET		54.54 216	55.85 216	53.01 217	44.64 230	70.42 223	54.91 221	64.61 189	39.64 221	23.66 231	26.69 225	32.48 226	41.99 216	71.03 211	74.94 205	88.27 207	75.74 214

new-patchmatchnet		53.91 217	52.69 219	55.33 216	64.83 221	70.90 221	52.24 224	61.75 196	41.09 220	30.82 221	29.90 220	28.22 229	36.69 222	61.52 221	65.08 220	85.64 215	72.14 222

MIMVSNet1		52.76 218	53.95 218	51.38 219	41.96 232	70.79 222	53.56 222	63.03 194	39.36 222	27.83 226	22.73 228	33.07 224	34.47 224	70.49 214	72.69 210	87.41 210	68.51 223

pmmvs3		52.59 219	52.43 221	52.78 218	54.53 227	64.49 226	50.07 225	46.89 226	35.31 226	30.19 222	27.27 224	26.96 231	53.02 205	67.28 218	70.54 214	81.96 222	75.20 216

new_pmnet		50.32 220	51.36 222	49.11 220	49.19 229	64.89 225	48.66 227	47.99 225	47.55 211	26.27 228	29.51 223	28.66 228	44.89 211	61.12 222	62.74 222	77.66 224	65.03 224

FPMVS		50.25 221	45.67 224	55.58 215	70.48 214	60.12 227	59.78 216	59.33 204	46.66 212	37.94 207	30.22 219	27.51 230	35.94 223	50.98 226	47.90 226	70.02 226	56.31 225

test_method		47.92 222	55.39 217	39.21 223	19.90 236	49.24 230	39.29 230	34.65 231	57.37 179	32.54 218	25.11 226	41.02 203	44.31 213	66.58 220	57.57 224	64.59 229	90.82 158

PMVS	36.83 18	40.62 223	36.39 226	45.56 221	58.40 224	33.20 233	32.62 232	56.02 208	28.25 230	37.92 208	22.29 229	26.15 232	25.29 227	48.49 228	43.82 229	63.13 230	52.53 228
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
WB-MVS		39.74 224	42.23 225	36.84 224	66.24 218	50.82 229	26.18 235	66.39 181	31.14 229	4.85 237	37.06 210	24.28 233	7.95 233	54.48 223	54.23 225	49.46 233	43.61 229

Gipuma		35.20 225	33.96 227	36.65 225	43.30 231	32.51 234	26.96 234	48.31 224	38.87 223	20.08 233	8.08 231	7.41 238	26.44 226	53.60 224	58.43 223	54.81 231	38.79 231
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
PMMVS2		32.52 226	33.92 228	30.88 227	34.15 235	44.70 232	27.79 233	39.69 230	22.21 231	4.31 238	15.73 230	14.13 236	12.45 232	40.11 229	47.00 227	66.88 227	53.54 226

E-PMN		21.42 227	17.56 230	25.94 228	36.25 234	19.02 237	11.56 236	43.72 228	15.25 233	6.99 235	8.04 232	4.53 240	21.77 229	16.13 232	26.16 231	35.34 234	33.77 232

MVE	25.07 19	21.25 228	23.51 229	18.62 230	15.07 237	29.77 236	10.67 238	34.60 232	12.51 234	9.46 234	7.84 233	3.82 241	14.38 231	27.45 231	42.42 230	27.56 236	40.74 230
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
EMVS		20.61 229	16.32 231	25.62 229	36.41 233	18.93 238	11.51 237	43.75 227	15.65 232	6.53 236	7.56 234	4.68 239	22.03 228	14.56 233	23.10 232	33.51 235	29.77 233

testmvs		0.76 230	1.23 232	0.21 231	0.05 239	0.21 239	0.38 240	0.09 234	0.94 235	0.05 240	2.13 236	0.08 242	0.60 235	0.82 234	0.77 233	0.11 237	3.62 235

test123		0.67 231	1.11 233	0.16 232	0.01 240	0.14 240	0.20 241	0.04 236	0.77 236	0.02 241	2.15 235	0.02 243	0.61 234	0.23 235	0.72 234	0.07 238	3.76 234

uanet_test		0.00 232	0.00 234	0.00 233	0.00 241	0.00 241	0.00 242	0.00 237	0.00 237	0.00 242	0.00 237	0.00 244	0.00 237	0.00 236	0.00 235	0.00 239	0.00 236

sosnet-low-res		0.00 232	0.00 234	0.00 233	0.00 241	0.00 241	0.00 242	0.00 237	0.00 237	0.00 242	0.00 237	0.00 244	0.00 237	0.00 236	0.00 235	0.00 239	0.00 236

sosnet		0.00 232	0.00 234	0.00 233	0.00 241	0.00 241	0.00 242	0.00 237	0.00 237	0.00 242	0.00 237	0.00 244	0.00 237	0.00 236	0.00 235	0.00 239	0.00 236

TPM-MVS					98.35 1	98.66 4	96.92 2			83.78 25	90.39 24	94.36 1	94.48 4			96.58 13	93.94 120
Ray Leroy Khuboni and Hongjun Xu: Textureless Resilient Propagation Matching in Multiple View Stereosis (TPM-MVS). SATNAC 2025
RE-MVS-def										39.41 206

9.14												91.16 7

SR-MVS					96.04 31			87.51 26				87.60 22

Anonymous202405211			75.59 143		85.13 127	91.06 117	84.62 101	77.96 79	69.47 145		40.79 199	63.84 128	84.57 69	83.55 135	84.69 122	89.69 194	95.75 91

our_test_3					73.80 201	79.57 209	64.47 209

ambc			50.35 223		55.61 226	59.93 228	48.73 226		44.08 215	35.81 215	24.01 227	10.64 237	41.57 218	72.83 203	63.35 221	74.99 225	77.61 210

MTAPA										91.14 7		85.84 28

MTMP										90.95 8		84.13 36

Patchmatch-RL test							8.17 239

tmp_tt				39.78 222	56.31 225	31.71 235	35.84 231	15.08 233	82.57 76	50.83 165	63.07 101	47.51 177	15.28 230	52.23 225	44.24 228	65.35 228

XVS					89.65 71	95.93 47	85.97 87			76.32 59		82.05 45				93.51 102

X-MVStestdata					89.65 71	95.93 47	85.97 87			76.32 59		82.05 45				93.51 102

mPP-MVS					95.90 33							80.22 54

NP-MVS									89.55 47

Patchmtry						87.41 143	78.32 150	54.14 215		51.09 160

DeepMVS_CX						48.96 231	43.77 228	40.58 229	50.93 201	24.67 230	36.95 211	20.18 235	41.60 217	38.92 230		52.37 232	53.31 227

High-resolution multi-view benchmark