High-res multi-view results

This table lists the benchmark results for the high-res multi-view scenario. The following metrics are evaluated:

Accuracy [%]: The fraction of the reconstruction which is closer to the ground truth than the evaluation threshold distance (*). Larger is better.
Completeness [%]: The fraction of the ground truth which is closer to the reconstruction than the evaluation threshold distance (*). Larger is better.
F₁ score [%]: The harmonic mean of accuracy and completeness, used to rank methods based on both metrics. Larger is better.
Time [s]: The runtime of the method. Smaller is better.

(*) For exact definitions, detailing how potentially incomplete ground truth is taken into account, see our paper.

The datasets are grouped into different categories, and result averages are computed for a category and method if results of the method are available for all datasets within the category. Note that the category "all" includes both the high-res multi-view and the low-res many-view scenarios.

Methods with suffix _ROB may participate in the Robust Vision Challenge.

Click a dataset result cell to show a visualization of the reconstruction. For training datasets, ground truth and accuracy / completeness visualizations are also available. The visualizations may not work with mobile browsers.

Method	all	high-res multi-view	indoor	outdoor	courty.	delive.	electro	facade	kicker	meadow	office	pipes	playgr.	relief	relief.	terrace	terrai.

DVP-MVS++		98.07 1	98.46 1	97.62 1	99.08 3	99.29 2	98.84 3	96.63 4	97.89 1	95.35 3	97.83 4	99.48 3	96.98 9	97.99 2	97.14 11	98.82 11	99.60 1

SED-MVS		97.98 2	98.36 2	97.54 4	98.94 16	99.29 2	98.81 4	96.64 3	97.14 3	95.16 4	97.96 2	99.61 2	96.92 12	98.00 1	97.24 8	98.75 17	99.25 3

DVP-MVS		97.93 3	98.23 3	97.58 3	99.05 6	99.31 1	98.64 6	96.62 5	97.56 2	95.08 5	96.61 13	99.64 1	97.32 1	97.91 4	97.31 6	98.77 15	99.26 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
DPE-MVS		97.83 4	98.13 4	97.48 5	98.83 22	99.19 4	98.99 1	96.70 1	96.05 18	94.39 9	98.30 1	99.47 4	97.02 6	97.75 7	97.02 14	98.98 3	99.10 9
Kehua Chen, Zhenlong Yuan, Tianlu Mao, Zhaoqi Wang: Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization. AAAI2025
MSP-MVS		97.70 6	98.09 5	97.24 6	99.00 11	99.17 5	98.76 5	96.41 9	96.91 5	93.88 14	97.72 5	99.04 7	96.93 11	97.29 17	97.31 6	98.45 37	99.23 4
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
APDe-MVS		97.79 5	97.96 6	97.60 2	99.20 2	99.10 6	98.88 2	96.68 2	96.81 7	94.64 6	97.84 3	98.02 11	97.24 3	97.74 8	97.02 14	98.97 5	99.16 6

SMA-MVS		97.53 7	97.93 7	97.07 10	99.21 1	99.02 8	98.08 19	96.25 11	96.36 12	93.57 15	96.56 14	99.27 5	96.78 16	97.91 4	97.43 3	98.51 26	98.94 12
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
SD-MVS		97.35 8	97.73 8	96.90 14	97.35 43	98.66 14	97.85 25	96.25 11	96.86 6	94.54 8	96.75 11	99.13 6	96.99 7	96.94 26	96.58 23	98.39 44	99.20 5
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.		97.31 9	97.64 9	96.92 13	97.28 45	98.56 23	98.61 7	95.48 28	96.72 8	94.03 13	96.73 12	98.29 9	97.15 4	97.61 12	96.42 26	98.96 6	99.13 7
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
SteuartSystems-ACMMP		97.10 15	97.49 10	96.65 18	98.97 13	98.95 9	98.43 9	95.96 17	95.12 28	91.46 28	96.85 9	97.60 18	96.37 24	97.76 6	97.16 10	98.68 18	98.97 11
Skip Steuart: Steuart Systems R&D Blog.
CNVR-MVS		97.30 10	97.41 11	97.18 8	99.02 10	98.60 21	98.15 16	96.24 13	96.12 17	94.10 11	95.54 25	97.99 12	96.99 7	97.97 3	97.17 9	98.57 24	98.50 29

HPM-MVS++		97.22 11	97.40 12	97.01 11	99.08 3	98.55 24	98.19 14	96.48 7	96.02 19	93.28 20	96.26 17	98.71 8	96.76 17	97.30 16	96.25 37	98.30 54	98.68 15

TSAR-MVS + ACMM		96.19 23	97.39 13	94.78 37	97.70 39	98.41 35	97.72 27	95.49 27	96.47 11	86.66 67	96.35 15	97.85 13	93.99 50	97.19 20	96.37 30	97.12 130	99.13 7

SF-MVS		97.20 12	97.29 14	97.10 9	98.95 15	98.51 29	97.51 29	96.48 7	96.17 16	94.64 6	97.32 6	97.57 19	96.23 26	96.78 29	96.15 41	98.79 14	98.55 27

ACMMP_NAP		96.93 16	97.27 15	96.53 23	99.06 5	98.95 9	98.24 13	96.06 15	95.66 21	90.96 32	95.63 24	97.71 16	96.53 20	97.66 10	96.68 20	98.30 54	98.61 20

HFP-MVS		97.11 14	97.19 16	97.00 12	98.97 13	98.73 12	98.37 11	95.69 21	96.60 9	93.28 20	96.87 8	96.64 28	97.27 2	96.64 35	96.33 35	98.44 38	98.56 22

MCST-MVS		96.83 18	97.06 17	96.57 19	98.88 20	98.47 32	98.02 21	96.16 14	95.58 23	90.96 32	95.78 23	97.84 14	96.46 22	97.00 25	96.17 39	98.94 7	98.55 27

APD-MVS		97.12 13	97.05 18	97.19 7	99.04 7	98.63 19	98.45 8	96.54 6	94.81 36	93.50 16	96.10 19	97.40 22	96.81 13	97.05 22	96.82 19	98.80 12	98.56 22
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
ACMMPR		96.92 17	96.96 19	96.87 15	98.99 12	98.78 11	98.38 10	95.52 24	96.57 10	92.81 24	96.06 20	95.90 35	97.07 5	96.60 37	96.34 34	98.46 34	98.42 33

DeepPCF-MVS	92.65 2	95.50 33	96.96 19	93.79 51	96.44 56	98.21 42	93.51 95	94.08 36	96.94 4	89.29 43	93.08 31	96.77 27	93.82 54	97.68 9	97.40 4	95.59 176	98.65 16

TSAR-MVS + GP.		95.86 28	96.95 21	94.60 41	94.07 85	98.11 46	96.30 43	91.76 49	95.67 20	91.07 30	96.82 10	97.69 17	95.71 32	95.96 52	95.75 50	98.68 18	98.63 17

PHI-MVS		95.86 28	96.93 22	94.61 40	97.60 41	98.65 18	96.49 40	93.13 40	94.07 43	87.91 57	97.12 7	97.17 24	93.90 53	96.46 40	96.93 17	98.64 20	98.10 49

MP-MVS		96.56 21	96.72 23	96.37 24	98.93 18	98.48 30	98.04 20	95.55 23	94.32 40	90.95 34	95.88 22	97.02 25	96.29 25	96.77 30	96.01 47	98.47 32	98.56 22
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
NCCC		96.75 19	96.67 24	96.85 16	99.03 9	98.44 34	98.15 16	96.28 10	96.32 13	92.39 25	92.16 34	97.55 20	96.68 19	97.32 14	96.65 22	98.55 25	98.26 38

train_agg		96.15 25	96.64 25	95.58 33	98.44 27	98.03 48	98.14 18	95.40 31	93.90 45	87.72 58	96.26 17	98.10 10	95.75 31	96.25 47	95.45 55	98.01 83	98.47 31

CP-MVS		96.68 20	96.59 26	96.77 17	98.85 21	98.58 22	98.18 15	95.51 26	95.34 25	92.94 23	95.21 28	96.25 30	96.79 15	96.44 42	95.77 49	98.35 46	98.56 22

DeepC-MVS_fast	93.32 1	96.48 22	96.42 27	96.56 20	98.70 25	98.31 38	97.97 22	95.76 20	96.31 14	92.01 27	91.43 39	95.42 39	96.46 22	97.65 11	97.69 1	98.49 31	98.12 47
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
X-MVS		96.07 26	96.33 28	95.77 29	98.94 16	98.66 14	97.94 23	95.41 30	95.12 28	88.03 53	93.00 32	96.06 31	95.85 29	96.65 34	96.35 31	98.47 32	98.48 30

PGM-MVS		96.16 24	96.33 28	95.95 26	99.04 7	98.63 19	98.32 12	92.76 42	93.42 47	90.49 37	96.30 16	95.31 40	96.71 18	96.46 40	96.02 46	98.38 45	98.19 42

MVS_111021_HR		94.84 39	95.91 30	93.60 52	97.35 43	98.46 33	95.08 59	91.19 54	94.18 42	85.97 72	95.38 26	92.56 51	93.61 57	96.61 36	96.25 37	98.40 42	97.92 56

MSLP-MVS++		96.05 27	95.63 31	96.55 21	98.33 29	98.17 44	96.94 36	94.61 34	94.70 38	94.37 10	89.20 51	95.96 34	96.81 13	95.57 58	97.33 5	98.24 62	98.47 31

MVS_111021_LR		94.84 39	95.57 32	94.00 44	97.11 48	97.72 60	94.88 63	91.16 55	95.24 27	88.74 48	96.03 21	91.52 58	94.33 47	95.96 52	95.01 62	97.79 95	97.49 72

CDPH-MVS		94.80 41	95.50 33	93.98 46	98.34 28	98.06 47	97.41 30	93.23 39	92.81 52	82.98 97	92.51 33	94.82 41	93.53 58	96.08 50	96.30 36	98.42 40	97.94 54

ACMMP		95.54 31	95.49 34	95.61 32	98.27 30	98.53 26	97.16 34	94.86 32	94.88 34	89.34 42	95.36 27	91.74 54	95.50 35	95.51 59	94.16 74	98.50 29	98.22 40
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
CSCG		95.68 30	95.46 35	95.93 27	98.71 24	99.07 7	97.13 35	93.55 37	95.48 24	93.35 19	90.61 44	93.82 45	95.16 37	94.60 82	95.57 53	97.70 104	99.08 10

CS-MVS-test		94.63 43	95.28 36	93.88 49	96.56 55	98.67 13	93.41 97	89.31 79	94.27 41	89.64 41	90.84 42	91.64 56	95.58 33	97.04 23	96.17 39	98.77 15	98.32 36

CPTT-MVS		95.54 31	95.07 37	96.10 25	97.88 35	97.98 50	97.92 24	94.86 32	94.56 39	92.16 26	91.01 40	95.71 36	96.97 10	94.56 83	93.50 90	96.81 153	98.14 45

DROMVSNet		94.19 48	95.05 38	93.18 58	93.56 101	97.65 61	95.34 57	86.37 115	92.05 59	88.71 49	89.91 47	93.32 47	96.14 27	97.29 17	96.42 26	98.98 3	98.70 14

CANet		94.85 38	94.92 39	94.78 37	97.25 46	98.52 28	97.20 32	91.81 48	93.25 49	91.06 31	86.29 65	94.46 43	92.99 64	97.02 24	96.68 20	98.34 48	98.20 41

DPM-MVS		95.07 35	94.84 40	95.34 34	97.44 42	97.49 65	97.76 26	95.52 24	94.88 34	88.92 46	87.25 58	96.44 29	94.41 43	95.78 55	96.11 43	97.99 85	95.95 123

DeepC-MVS	92.10 3	95.22 34	94.77 41	95.75 30	97.77 37	98.54 25	97.63 28	95.96 17	95.07 31	88.85 47	85.35 73	91.85 53	95.82 30	96.88 28	97.10 12	98.44 38	98.63 17
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
CS-MVS		94.53 44	94.73 42	94.31 42	96.30 59	98.53 26	94.98 60	89.24 81	93.37 48	90.24 39	88.96 53	89.76 70	96.09 28	97.48 13	96.42 26	98.99 2	98.59 21

MVS_0304		94.30 47	94.68 43	93.86 50	96.33 58	98.48 30	97.41 30	91.20 53	92.75 53	86.96 64	86.03 68	93.81 46	92.64 69	96.89 27	96.54 25	98.61 22	98.24 39

ETV-MVS		93.80 51	94.57 44	92.91 65	93.98 87	97.50 64	93.62 92	88.70 86	91.95 60	87.57 59	90.21 46	90.79 61	94.56 42	97.20 19	96.35 31	99.02 1	97.98 51

3Dnovator+	90.56 5	95.06 36	94.56 45	95.65 31	98.11 31	98.15 45	97.19 33	91.59 51	95.11 30	93.23 22	81.99 100	94.71 42	95.43 36	96.48 39	96.88 18	98.35 46	98.63 17

EPNet		93.92 50	94.40 46	93.36 54	97.89 34	96.55 89	96.08 46	92.14 45	91.65 64	89.16 44	94.07 30	90.17 69	87.78 123	95.24 64	94.97 63	97.09 132	98.15 44
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
OMC-MVS		94.49 45	94.36 47	94.64 39	97.17 47	97.73 58	95.49 54	92.25 44	96.18 15	90.34 38	88.51 54	92.88 50	94.90 40	94.92 70	94.17 73	97.69 106	96.15 116

3Dnovator	90.28 7	94.70 42	94.34 48	95.11 35	98.06 32	98.21 42	96.89 37	91.03 57	94.72 37	91.45 29	82.87 91	93.10 49	94.61 41	96.24 48	97.08 13	98.63 21	98.16 43

QAPM		94.13 49	94.33 49	93.90 47	97.82 36	98.37 37	96.47 41	90.89 58	92.73 55	85.63 80	85.35 73	93.87 44	94.17 48	95.71 57	95.90 48	98.40 42	98.42 33

AdaColmap		95.02 37	93.71 50	96.54 22	98.51 26	97.76 56	96.69 39	95.94 19	93.72 46	93.50 16	89.01 52	90.53 65	96.49 21	94.51 85	93.76 83	98.07 77	96.69 96

TAPA-MVS	90.35 6	93.69 53	93.52 51	93.90 47	96.89 51	97.62 62	96.15 44	91.67 50	94.94 32	85.97 72	87.72 57	91.96 52	94.40 44	93.76 99	93.06 106	98.30 54	95.58 132
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
DELS-MVS		93.71 52	93.47 52	94.00 44	96.82 52	98.39 36	96.80 38	91.07 56	89.51 98	89.94 40	83.80 83	89.29 71	90.95 87	97.32 14	97.65 2	98.42 40	98.32 36
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
UGNet		91.52 76	93.41 53	89.32 107	94.13 82	97.15 75	91.83 123	89.01 82	90.62 73	85.86 76	86.83 59	91.73 55	77.40 188	94.68 79	94.43 69	97.71 102	98.40 35
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
IS_MVSNet		91.87 70	93.35 54	90.14 101	94.09 84	97.73 58	93.09 102	88.12 94	88.71 105	79.98 114	84.49 78	90.63 64	87.49 127	97.07 21	96.96 16	98.07 77	97.88 60

canonicalmvs		93.08 55	93.09 55	93.07 62	94.24 81	97.86 52	95.45 56	87.86 102	94.00 44	87.47 60	88.32 55	82.37 104	95.13 38	93.96 98	96.41 29	98.27 58	98.73 13

EPP-MVSNet		92.13 65	93.06 56	91.05 90	93.66 100	97.30 68	92.18 114	87.90 98	90.24 81	83.63 94	86.14 67	90.52 67	90.76 89	94.82 75	94.38 70	98.18 68	97.98 51

PLC	90.69 4	94.32 46	92.99 57	95.87 28	97.91 33	96.49 91	95.95 50	94.12 35	94.94 32	94.09 12	85.90 69	90.77 62	95.58 33	94.52 84	93.32 97	97.55 113	95.00 144
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
EIA-MVS		92.72 59	92.96 58	92.44 70	93.86 94	97.76 56	93.13 101	88.65 88	89.78 94	86.68 66	86.69 62	87.57 72	93.74 55	96.07 51	95.32 56	98.58 23	97.53 70

CANet_DTU		90.74 90	92.93 59	88.19 118	94.36 80	96.61 87	94.34 71	84.66 130	90.66 71	68.75 167	90.41 45	86.89 76	89.78 99	95.46 60	94.87 64	97.25 122	95.62 130

Vis-MVSNet (Re-imp)		90.54 92	92.76 60	87.94 122	93.73 98	96.94 83	92.17 116	87.91 97	88.77 104	76.12 128	83.68 84	90.80 60	79.49 184	96.34 45	96.35 31	98.21 65	96.46 103

MAR-MVS		92.71 60	92.63 61	92.79 66	97.70 39	97.15 75	93.75 88	87.98 96	90.71 70	85.76 78	86.28 66	86.38 78	94.35 46	94.95 68	95.49 54	97.22 123	97.44 73
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
UA-Net		90.81 85	92.58 62	88.74 113	94.87 78	97.44 66	92.61 107	88.22 92	82.35 158	78.93 118	85.20 75	95.61 37	79.56 183	96.52 38	96.57 24	98.23 63	94.37 151

CNLPA		93.69 53	92.50 63	95.06 36	97.11 48	97.36 67	93.88 85	93.30 38	95.64 22	93.44 18	80.32 109	90.73 63	94.99 39	93.58 101	93.33 95	97.67 108	96.57 101

HQP-MVS		92.39 63	92.49 64	92.29 73	95.65 65	95.94 103	95.64 53	92.12 46	92.46 57	79.65 115	91.97 36	82.68 100	92.92 67	93.47 106	92.77 111	97.74 100	98.12 47

PVSNet_BlendedMVS		92.80 57	92.44 65	93.23 55	96.02 61	97.83 54	93.74 89	90.58 59	91.86 61	90.69 35	85.87 71	82.04 107	90.01 97	96.39 43	95.26 58	98.34 48	97.81 61

PVSNet_Blended		92.80 57	92.44 65	93.23 55	96.02 61	97.83 54	93.74 89	90.58 59	91.86 61	90.69 35	85.87 71	82.04 107	90.01 97	96.39 43	95.26 58	98.34 48	97.81 61

PVSNet_Blended_VisFu		91.92 69	92.39 67	91.36 88	95.45 71	97.85 53	92.25 113	89.54 76	88.53 108	87.47 60	79.82 111	90.53 65	85.47 148	96.31 46	95.16 61	97.99 85	98.56 22

TSAR-MVS + COLMAP		92.39 63	92.31 68	92.47 69	95.35 73	96.46 93	96.13 45	92.04 47	95.33 26	80.11 113	94.95 29	77.35 136	94.05 49	94.49 86	93.08 104	97.15 127	94.53 148

MVS_Test		91.81 72	92.19 69	91.37 87	93.24 104	96.95 82	94.43 67	86.25 116	91.45 67	83.45 95	86.31 64	85.15 86	92.93 65	93.99 94	94.71 67	97.92 89	96.77 94

CHOSEN 280x420		90.77 88	92.14 70	89.17 109	93.86 94	92.81 161	93.16 100	80.22 177	90.21 82	84.67 92	89.89 48	91.38 59	90.57 94	94.94 69	92.11 123	92.52 197	93.65 161

PCF-MVS	90.19 8	92.98 56	92.07 71	94.04 43	96.39 57	97.87 51	96.03 47	95.47 29	87.16 116	85.09 90	84.81 77	93.21 48	93.46 60	91.98 132	91.98 128	97.78 96	97.51 71
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
LGP-MVS_train		91.83 71	92.04 72	91.58 80	95.46 69	96.18 99	95.97 49	89.85 68	90.45 77	77.76 120	91.92 37	80.07 118	92.34 73	94.27 88	93.47 91	98.11 74	97.90 59

CLD-MVS		92.50 62	91.96 73	93.13 59	93.93 91	96.24 97	95.69 51	88.77 85	92.92 50	89.01 45	88.19 56	81.74 110	93.13 63	93.63 100	93.08 104	98.23 63	97.91 58
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
baseline		91.19 79	91.89 74	90.38 92	92.76 114	95.04 110	93.55 94	84.54 133	92.92 50	85.71 79	86.68 63	86.96 75	89.28 107	92.00 131	92.62 114	96.46 158	96.99 88

thisisatest0530		91.04 82	91.74 75	90.21 96	92.93 112	97.00 80	92.06 119	87.63 107	90.74 69	81.51 101	86.81 60	82.48 101	89.23 108	94.81 76	93.03 108	97.90 90	97.33 78

tttt0517		91.01 83	91.71 76	90.19 98	92.98 108	97.07 79	91.96 122	87.63 107	90.61 74	81.42 102	86.76 61	82.26 105	89.23 108	94.86 74	93.03 108	97.90 90	97.36 76

ACMP	89.13 9	92.03 66	91.70 77	92.41 71	94.92 76	96.44 95	93.95 81	89.96 67	91.81 63	85.48 85	90.97 41	79.12 121	92.42 71	93.28 112	92.55 115	97.76 98	97.74 64
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
MVSTER		91.73 73	91.61 78	91.86 77	93.18 105	94.56 112	94.37 69	87.90 98	90.16 85	88.69 50	89.23 50	81.28 112	88.92 116	95.75 56	93.95 80	98.12 72	96.37 107

OpenMVS	88.18 11	92.51 61	91.61 78	93.55 53	97.74 38	98.02 49	95.66 52	90.46 61	89.14 101	86.50 68	75.80 134	90.38 68	92.69 68	94.99 67	95.30 57	98.27 58	97.63 65

Vis-MVSNet		89.36 108	91.49 80	86.88 133	92.10 122	97.60 63	92.16 117	85.89 118	84.21 144	75.20 130	82.58 95	87.13 74	77.40 188	95.90 54	95.63 51	98.51 26	97.36 76
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
FA-MVS(training)		90.79 87	91.33 81	90.17 99	93.76 97	97.22 72	92.74 106	77.79 187	90.60 75	88.03 53	78.80 115	87.41 73	91.00 86	95.40 62	93.43 93	97.70 104	96.46 103

DCV-MVSNet		91.24 78	91.26 82	91.22 89	92.84 113	93.44 139	93.82 86	86.75 112	91.33 68	85.61 81	84.00 82	85.46 85	91.27 81	92.91 114	93.62 85	97.02 136	98.05 50

casdiffmvs_mvg		91.94 68	91.25 83	92.75 67	93.41 103	97.19 74	95.48 55	89.77 70	89.86 92	86.41 69	81.02 107	82.23 106	92.93 65	95.44 61	95.61 52	98.51 26	97.40 75
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
PMMVS		89.88 100	91.19 84	88.35 116	89.73 148	91.97 181	90.62 129	81.92 164	90.57 76	80.58 112	92.16 34	86.85 77	91.17 83	92.31 124	91.35 139	96.11 164	93.11 168

casdiffmvs		91.72 74	91.16 85	92.38 72	93.16 106	97.15 75	93.95 81	89.49 77	91.58 66	86.03 71	80.75 108	80.95 113	93.16 62	95.25 63	95.22 60	98.50 29	97.23 81
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
diffmvs		91.37 77	91.09 86	91.70 79	92.71 117	96.47 92	94.03 79	88.78 84	92.74 54	85.43 87	83.63 85	80.37 115	91.76 78	93.39 108	93.78 82	97.50 115	97.23 81
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
LS3D		91.97 67	90.98 87	93.12 60	97.03 50	97.09 78	95.33 58	95.59 22	92.47 56	79.26 117	81.60 103	82.77 99	94.39 45	94.28 87	94.23 72	97.14 129	94.45 150

ET-MVSNet_ETH3D		89.93 99	90.84 88	88.87 111	79.60 210	96.19 98	94.43 67	86.56 113	90.63 72	80.75 110	90.71 43	77.78 132	93.73 56	91.36 140	93.45 92	98.15 69	95.77 127

EPNet_dtu		88.32 117	90.61 89	85.64 145	96.79 53	92.27 173	92.03 120	90.31 62	89.05 102	65.44 188	89.43 49	85.90 83	74.22 197	92.76 115	92.09 124	95.02 186	92.76 170
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
ACMM	88.76 10	91.70 75	90.43 90	93.19 57	95.56 66	95.14 109	93.35 99	91.48 52	92.26 58	87.12 62	84.02 81	79.34 120	93.99 50	94.07 93	92.68 112	97.62 112	95.50 133
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
baseline1		90.81 85	90.29 91	91.42 84	93.67 99	95.86 104	93.94 83	89.69 74	89.29 100	82.85 98	82.91 90	80.30 116	89.60 100	95.05 66	94.79 66	98.80 12	93.82 159

GG-mvs-BLEND		62.84 209	90.21 92	30.91 217	0.57 225	94.45 116	86.99 179	0.34 223	88.71 105	0.98 225	81.55 105	91.58 57	0.86 222	92.66 118	91.43 138	95.73 170	91.11 180

FC-MVSNet-train		90.55 91	90.19 93	90.97 91	93.78 96	95.16 108	92.11 118	88.85 83	87.64 113	83.38 96	84.36 80	78.41 127	89.53 101	94.69 78	93.15 103	98.15 69	97.92 56

DI_MVS_plusplus_trai		91.05 81	90.15 94	92.11 74	92.67 118	96.61 87	96.03 47	88.44 90	90.25 80	85.92 74	73.73 142	84.89 88	91.92 75	94.17 91	94.07 78	97.68 107	97.31 79

GBi-Net		90.21 96	90.11 95	90.32 94	88.66 158	93.65 135	94.25 74	85.78 120	90.03 86	85.56 82	77.38 120	86.13 79	89.38 104	93.97 95	94.16 74	98.31 51	95.47 134

test1		90.21 96	90.11 95	90.32 94	88.66 158	93.65 135	94.25 74	85.78 120	90.03 86	85.56 82	77.38 120	86.13 79	89.38 104	93.97 95	94.16 74	98.31 51	95.47 134

FMVSNet3		90.19 98	90.06 97	90.34 93	88.69 157	93.85 127	94.58 64	85.78 120	90.03 86	85.56 82	77.38 120	86.13 79	89.22 110	93.29 111	94.36 71	98.20 66	95.40 138

Effi-MVS+		89.79 102	89.83 98	89.74 103	92.98 108	96.45 94	93.48 96	84.24 135	87.62 114	76.45 126	81.76 101	77.56 135	93.48 59	94.61 81	93.59 86	97.82 94	97.22 83

baseline2		88.97 112	89.50 99	88.36 115	91.14 134	95.30 106	90.13 140	85.17 127	87.24 115	80.80 109	84.46 79	78.44 126	85.60 145	93.54 104	91.87 129	97.31 120	95.66 129

OPM-MVS		91.08 80	89.34 100	93.11 61	96.18 60	96.13 100	96.39 42	92.39 43	82.97 155	81.74 100	82.55 97	80.20 117	93.97 52	94.62 80	93.23 98	98.00 84	95.73 128
Ray L. Khuboni, Hongjun Xu: Octagram Propagation Matching for Multi-Scale View Stereopsis (OPM-MVS).
ECVR-MVS		90.77 88	89.27 101	92.52 68	94.97 74	98.30 39	94.53 65	90.25 64	89.91 90	85.80 77	73.64 143	74.31 145	90.69 90	96.75 32	96.10 44	98.87 8	95.91 125

RPSCF		89.68 103	89.24 102	90.20 97	92.97 110	92.93 157	92.30 111	87.69 104	90.44 78	85.12 89	91.68 38	85.84 84	90.69 90	87.34 187	86.07 189	92.46 198	90.37 186

test2506		90.93 84	89.20 103	92.95 63	94.97 74	98.30 39	94.53 65	90.25 64	89.91 90	88.39 52	83.23 87	64.17 190	90.69 90	96.75 32	96.10 44	98.87 8	95.97 122

FMVSNet2		89.61 104	89.14 104	90.16 100	88.66 158	93.65 135	94.25 74	85.44 124	88.57 107	84.96 91	73.53 145	83.82 91	89.38 104	94.23 89	94.68 68	98.31 51	95.47 134

test1111		90.47 93	89.10 105	92.07 75	94.92 76	98.30 39	94.17 78	90.30 63	89.56 97	83.92 93	73.25 150	73.66 146	90.26 96	96.77 30	96.14 42	98.87 8	96.04 120

FC-MVSNet-test		86.15 134	89.10 105	82.71 183	89.83 146	93.18 149	87.88 172	84.69 129	86.54 123	62.18 198	82.39 98	83.31 94	74.18 198	92.52 122	91.86 130	97.50 115	93.88 158

PatchMatch-RL		90.30 95	88.93 107	91.89 76	95.41 72	95.68 105	90.94 126	88.67 87	89.80 93	86.95 65	85.90 69	72.51 148	92.46 70	93.56 103	92.18 121	96.93 145	92.89 169

CDS-MVSNet		88.34 116	88.71 108	87.90 123	90.70 142	94.54 113	92.38 109	86.02 117	80.37 167	79.42 116	79.30 112	83.43 93	82.04 171	93.39 108	94.01 79	96.86 151	95.93 124
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
GeoE		89.29 110	88.68 109	89.99 102	92.75 116	96.03 102	93.07 104	83.79 142	86.98 118	81.34 103	74.72 139	78.92 122	91.22 82	93.31 110	93.21 100	97.78 96	97.60 69

IterMVS-LS		88.60 113	88.45 110	88.78 112	92.02 123	92.44 171	92.00 121	83.57 146	86.52 124	78.90 119	78.61 117	81.34 111	89.12 111	90.68 153	93.18 101	97.10 131	96.35 108
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
test-mter		86.09 137	88.38 111	83.43 173	87.89 168	92.61 165	86.89 180	77.11 190	84.30 142	68.62 169	82.57 96	82.45 102	84.34 155	92.40 123	90.11 165	95.74 169	94.21 154

test-LLR		86.88 126	88.28 112	85.24 149	91.22 132	92.07 177	87.41 175	83.62 144	84.58 137	69.33 163	83.00 88	82.79 97	84.24 156	92.26 125	89.81 171	95.64 174	93.44 162

TESTMET0.1,1		86.11 136	88.28 112	83.59 170	87.80 169	92.07 177	87.41 175	77.12 189	84.58 137	69.33 163	83.00 88	82.79 97	84.24 156	92.26 125	89.81 171	95.64 174	93.44 162

MSDG		90.42 94	88.25 114	92.94 64	96.67 54	94.41 118	93.96 80	92.91 41	89.59 96	86.26 70	76.74 127	80.92 114	90.43 95	92.60 120	92.08 125	97.44 118	91.41 176

Anonymous20231211		89.82 101	88.18 115	91.74 78	92.52 119	96.09 101	93.38 98	89.30 80	88.95 103	85.90 75	64.55 191	84.39 89	92.41 72	92.24 127	93.06 106	96.93 145	97.95 53

Effi-MVS+-dtu		87.51 122	88.13 116	86.77 135	91.10 135	94.90 111	90.91 127	82.67 154	83.47 151	71.55 146	81.11 106	77.04 137	89.41 103	92.65 119	91.68 135	95.00 187	96.09 118

Anonymous202405211			88.00 117		93.16 106	96.38 96	93.58 93	89.34 78	87.92 112		65.04 187	83.03 96	92.07 74	92.67 117	93.33 95	96.96 140	97.63 65

Fast-Effi-MVS+		88.56 115	87.99 118	89.22 108	91.56 129	95.21 107	92.29 112	82.69 153	86.82 119	77.73 121	76.24 132	73.39 147	93.36 61	94.22 90	93.64 84	97.65 109	96.43 105

IB-MVS	85.10 14	87.98 118	87.97 119	87.99 121	94.55 79	96.86 85	84.52 192	88.21 93	86.48 126	88.54 51	74.41 141	77.74 133	74.10 199	89.65 170	92.85 110	98.06 79	97.80 63
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
tfpn200view9		89.55 105	87.86 120	91.53 82	93.90 92	97.26 69	94.31 73	89.74 71	85.87 128	81.15 105	76.46 129	70.38 157	91.76 78	94.92 70	93.51 87	98.28 57	96.61 98

CHOSEN 1792x2688		88.57 114	87.82 121	89.44 106	95.46 69	96.89 84	93.74 89	85.87 119	89.63 95	77.42 123	61.38 197	83.31 94	88.80 118	93.44 107	93.16 102	95.37 181	96.95 90

thres200		89.49 106	87.72 122	91.55 81	93.95 89	97.25 70	94.34 71	89.74 71	85.66 131	81.18 104	76.12 133	70.19 160	91.80 76	94.92 70	93.51 87	98.27 58	96.40 106

Fast-Effi-MVS+-dtu		86.25 131	87.70 123	84.56 158	90.37 145	93.70 132	90.54 130	78.14 184	83.50 150	65.37 189	81.59 104	75.83 143	86.09 144	91.70 135	91.70 133	96.88 149	95.84 126

test0.0.03 1		85.58 142	87.69 124	83.11 176	91.22 132	92.54 168	85.60 191	83.62 144	85.66 131	67.84 174	82.79 93	79.70 119	73.51 201	91.15 145	90.79 144	96.88 149	91.23 179

thres100view900		89.36 108	87.61 125	91.39 85	93.90 92	96.86 85	94.35 70	89.66 75	85.87 128	81.15 105	76.46 129	70.38 157	91.17 83	94.09 92	93.43 93	98.13 71	96.16 115

MS-PatchMatch		87.63 120	87.61 125	87.65 126	93.95 89	94.09 123	92.60 108	81.52 169	86.64 121	76.41 127	73.46 147	85.94 82	85.01 152	92.23 128	90.00 168	96.43 160	90.93 182

thres400		89.40 107	87.58 127	91.53 82	94.06 86	97.21 73	94.19 77	89.83 69	85.69 130	81.08 107	75.50 136	69.76 161	91.80 76	94.79 77	93.51 87	98.20 66	96.60 99

SCA		86.25 131	87.52 128	84.77 154	91.59 127	93.90 126	89.11 159	73.25 204	90.38 79	72.84 140	83.26 86	83.79 92	88.49 120	86.07 194	85.56 192	93.33 190	89.67 191

MDTV_nov1_ep13		86.64 130	87.50 129	85.65 144	90.73 140	93.69 133	89.96 144	78.03 186	89.48 99	76.85 125	84.92 76	82.42 103	86.14 142	86.85 191	86.15 188	92.17 199	88.97 194

thres600view7		89.28 111	87.47 130	91.39 85	94.12 83	97.25 70	93.94 83	89.74 71	85.62 133	80.63 111	75.24 138	69.33 162	91.66 80	94.92 70	93.23 98	98.27 58	96.72 95

thisisatest0515		85.70 140	87.00 131	84.19 163	88.16 165	93.67 134	84.20 194	84.14 138	83.39 153	72.91 139	76.79 126	74.75 144	78.82 186	92.57 121	91.26 140	96.94 142	96.56 102

IterMVS-SCA-FT		85.44 146	86.71 132	83.97 167	90.59 143	90.84 194	89.73 150	78.34 183	84.07 148	66.40 183	77.27 125	78.66 124	83.06 164	91.20 142	90.10 166	95.72 171	94.78 145

PatchmatchNet		85.70 140	86.65 133	84.60 157	91.79 124	93.40 140	89.27 155	73.62 199	90.19 83	72.63 142	82.74 94	81.93 109	87.64 124	84.99 197	84.29 199	92.64 196	89.00 193
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
ACMH	85.51 13	87.31 124	86.59 134	88.14 119	93.96 88	94.51 114	89.00 162	87.99 95	81.58 161	70.15 157	78.41 118	71.78 153	90.60 93	91.30 141	91.99 127	97.17 126	96.58 100
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
IterMVS		85.25 148	86.49 135	83.80 168	90.42 144	90.77 197	90.02 142	78.04 185	84.10 146	66.27 184	77.28 124	78.41 127	83.01 165	90.88 147	89.72 175	95.04 185	94.24 152
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
HyFIR lowres test		87.87 119	86.42 136	89.57 104	95.56 66	96.99 81	92.37 110	84.15 137	86.64 121	77.17 124	57.65 203	83.97 90	91.08 85	92.09 130	92.44 116	97.09 132	95.16 141

CR-MVSNet		85.48 144	86.29 137	84.53 159	91.08 137	92.10 175	89.18 157	73.30 202	84.75 135	71.08 151	73.12 152	77.91 131	86.27 140	91.48 137	90.75 147	96.27 162	93.94 156

EPMVS		85.77 139	86.24 138	85.23 150	92.76 114	93.78 129	89.91 146	73.60 200	90.19 83	74.22 132	82.18 99	78.06 129	87.55 126	85.61 196	85.38 194	93.32 191	88.48 198

CostFormer		86.78 128	86.05 139	87.62 128	92.15 121	93.20 148	91.55 125	75.83 192	88.11 111	85.29 88	81.76 101	76.22 141	87.80 122	84.45 199	85.21 195	93.12 192	93.42 164

COLMAP_ROB	84.39 15	87.61 121	86.03 140	89.46 105	95.54 68	94.48 115	91.77 124	90.14 66	87.16 116	75.50 129	73.41 148	76.86 139	87.33 129	90.05 164	89.76 174	96.48 157	90.46 185
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
ACMH+	85.75 12	87.19 125	86.02 141	88.56 114	93.42 102	94.41 118	89.91 146	87.66 106	83.45 152	72.25 144	76.42 131	71.99 152	90.78 88	89.86 165	90.94 142	97.32 119	95.11 143

FMVSNet1		87.33 123	86.00 142	88.89 110	87.13 184	92.83 160	93.08 103	84.46 134	81.35 163	82.20 99	66.33 178	77.96 130	88.96 113	93.97 95	94.16 74	97.54 114	95.38 139

USDC		86.73 129	85.96 143	87.63 127	91.64 126	93.97 125	92.76 105	84.58 132	88.19 109	70.67 154	80.10 110	67.86 169	89.43 102	91.81 133	89.77 173	96.69 155	90.05 189

RPMNet		84.82 153	85.90 144	83.56 171	91.10 135	92.10 175	88.73 166	71.11 207	84.75 135	68.79 166	73.56 144	77.62 134	85.33 149	90.08 163	89.43 177	96.32 161	93.77 160

UniMVSNet_NR-MVSNet		86.80 127	85.86 145	87.89 124	88.17 164	94.07 124	90.15 138	88.51 89	84.20 145	73.45 137	72.38 154	70.30 159	88.95 114	90.25 158	92.21 120	98.12 72	97.62 67

anonymousdsp		84.51 156	85.85 146	82.95 180	86.30 195	93.51 138	85.77 189	80.38 176	78.25 181	63.42 195	73.51 146	72.20 150	84.64 154	93.21 113	92.16 122	97.19 125	98.14 45

GA-MVS		85.08 149	85.65 147	84.42 160	89.77 147	94.25 121	89.26 156	84.62 131	81.19 164	62.25 197	75.72 135	68.44 166	84.14 159	93.57 102	91.68 135	96.49 156	94.71 147

CVMVSNet		83.83 167	85.53 148	81.85 190	89.60 149	90.92 192	87.81 173	83.21 150	80.11 170	60.16 202	76.47 128	78.57 125	76.79 190	89.76 166	90.13 161	93.51 189	92.75 171

PatchT		83.86 166	85.51 149	81.94 189	88.41 161	91.56 187	78.79 206	71.57 206	84.08 147	71.08 151	70.62 158	76.13 142	86.27 140	91.48 137	90.75 147	95.52 179	93.94 156

UniMVSNet (Re)		86.22 133	85.46 150	87.11 130	88.34 162	94.42 117	89.65 152	87.10 111	84.39 141	74.61 131	70.41 162	68.10 167	85.10 151	91.17 144	91.79 131	97.84 93	97.94 54

ADS-MVSNet		84.08 163	84.95 151	83.05 179	91.53 131	91.75 184	88.16 169	70.70 208	89.96 89	69.51 162	78.83 114	76.97 138	86.29 139	84.08 201	84.60 197	92.13 201	88.48 198

TAMVS		84.94 152	84.95 151	84.93 153	88.82 154	93.18 149	88.44 168	81.28 171	77.16 186	73.76 136	75.43 137	76.57 140	82.04 171	90.59 154	90.79 144	95.22 183	90.94 181

DU-MVS		86.12 135	84.81 153	87.66 125	87.77 171	93.78 129	90.15 138	87.87 100	84.40 139	73.45 137	70.59 159	64.82 187	88.95 114	90.14 159	92.33 117	97.76 98	97.62 67

FMVSNet5		84.47 159	84.72 154	84.18 164	83.30 205	88.43 202	88.09 170	79.42 180	84.25 143	74.14 134	73.15 151	78.74 123	83.65 162	91.19 143	91.19 141	96.46 158	86.07 203

NR-MVSNet		85.46 145	84.54 155	86.52 138	88.33 163	93.78 129	90.45 131	87.87 100	84.40 139	71.61 145	70.59 159	62.09 197	82.79 167	91.75 134	91.75 132	98.10 75	97.44 73

TranMVSNet+NR-MVSNet		85.57 143	84.41 156	86.92 132	87.67 174	93.34 142	90.31 134	88.43 91	83.07 154	70.11 158	69.99 165	65.28 182	86.96 132	89.73 167	92.27 118	98.06 79	97.17 85

pmmvs4		86.00 138	84.28 157	88.00 120	87.80 169	92.01 180	89.94 145	84.91 128	86.79 120	80.98 108	73.41 148	66.34 178	88.12 121	89.31 173	88.90 182	96.24 163	93.20 167

testgi		81.94 187	84.09 158	79.43 196	89.53 151	90.83 195	82.49 198	81.75 167	80.59 165	59.46 204	82.82 92	65.75 179	67.97 203	90.10 162	89.52 176	95.39 180	89.03 192

MIMVSNet		82.97 179	84.00 159	81.77 191	82.23 206	92.25 174	87.40 177	72.73 205	81.48 162	69.55 161	68.79 167	72.42 149	81.82 174	92.23 128	92.25 119	96.89 148	88.61 196

tpm		83.16 175	83.64 160	82.60 185	90.75 139	91.05 191	88.49 167	73.99 197	82.36 157	67.08 180	78.10 119	68.79 163	84.17 158	85.95 195	85.96 190	91.09 204	93.23 166

gg-mvs-nofinetune		81.83 188	83.58 161	79.80 195	91.57 128	96.54 90	93.79 87	68.80 211	62.71 215	43.01 220	55.28 206	85.06 87	83.65 162	96.13 49	94.86 65	97.98 88	94.46 149

pm-mvs1		84.55 155	83.46 162	85.82 141	88.16 165	93.39 141	89.05 161	85.36 126	74.03 200	72.43 143	65.08 186	71.11 154	82.30 170	93.48 105	91.70 133	97.64 110	95.43 137

tpmrst		83.72 169	83.45 163	84.03 166	92.21 120	91.66 185	88.74 165	73.58 201	88.14 110	72.67 141	77.37 123	72.11 151	86.34 138	82.94 204	82.05 203	90.63 207	89.86 190

SixPastTwentyTwo		83.12 177	83.44 164	82.74 182	87.71 173	93.11 153	82.30 199	82.33 159	79.24 175	64.33 192	78.77 116	62.75 193	84.11 160	88.11 182	87.89 184	95.70 172	94.21 154

Baseline_NR-MVSNet		85.28 147	83.42 165	87.46 129	87.77 171	90.80 196	89.90 148	87.69 104	83.93 149	74.16 133	64.72 189	66.43 177	87.48 128	90.14 159	90.83 143	97.73 101	97.11 86

TDRefinement		84.97 151	83.39 166	86.81 134	92.97 110	94.12 122	92.18 114	87.77 103	82.78 156	71.31 149	68.43 168	68.07 168	81.10 179	89.70 169	89.03 181	95.55 178	91.62 174

WR-MVS		83.14 176	83.38 167	82.87 181	87.55 175	93.29 144	86.36 185	84.21 136	80.05 171	66.41 182	66.91 174	66.92 174	75.66 195	88.96 178	90.56 152	97.05 134	96.96 89

V42		84.48 158	83.36 168	85.79 143	87.14 183	93.28 145	90.03 141	83.98 140	80.30 168	71.20 150	66.90 175	67.17 171	85.55 146	89.35 171	90.27 158	96.82 152	96.27 113

v8		84.45 160	83.30 169	85.80 142	87.53 176	92.95 155	90.31 134	82.46 158	80.46 166	71.43 147	66.99 173	67.16 172	86.14 142	89.26 174	90.22 160	96.94 142	96.06 119

dps		85.00 150	83.21 170	87.08 131	90.73 140	92.55 167	89.34 154	75.29 194	84.94 134	87.01 63	79.27 113	67.69 170	87.27 130	84.22 200	83.56 200	92.83 195	90.25 187

v10		84.18 161	83.17 171	85.37 146	87.34 178	92.68 163	90.32 133	81.33 170	79.93 174	69.23 165	66.33 178	65.74 180	87.03 131	90.84 148	90.38 155	96.97 138	96.29 112

v2v482		84.51 156	83.05 172	86.20 140	87.25 180	93.28 145	90.22 136	85.40 125	79.94 173	69.78 160	67.74 170	65.15 184	87.57 125	89.12 176	90.55 153	96.97 138	95.60 131

v1144		84.03 165	82.88 173	85.37 146	87.17 182	93.15 152	90.18 137	83.31 149	78.83 177	67.85 173	65.99 180	64.99 185	86.79 134	90.75 150	90.33 157	96.90 147	96.15 116

pmmvs5		83.37 173	82.68 174	84.18 164	87.13 184	93.18 149	86.74 181	82.08 163	76.48 190	67.28 178	71.26 156	62.70 194	84.71 153	90.77 149	90.12 164	97.15 127	94.24 152

WR-MVS_H		82.86 181	82.66 175	83.10 177	87.44 177	93.33 143	85.71 190	83.20 151	77.36 185	68.20 172	66.37 177	65.23 183	76.05 194	89.35 171	90.13 161	97.99 85	96.89 92

v1192		83.56 171	82.35 176	84.98 151	86.84 189	92.84 158	90.01 143	82.70 152	78.54 178	66.48 181	64.88 188	62.91 192	86.91 133	90.72 151	90.25 159	96.94 142	96.32 110

v144192		83.48 172	82.23 177	84.94 152	86.65 190	92.84 158	89.63 153	82.48 157	77.87 182	67.36 177	65.33 185	63.50 191	86.51 136	89.72 168	89.99 169	97.03 135	96.35 108

CP-MVSNet		83.11 178	82.15 178	84.23 162	87.20 181	92.70 162	86.42 184	83.53 147	77.83 183	67.67 175	66.89 176	60.53 205	82.47 168	89.23 175	90.65 151	98.08 76	97.20 84

v148		83.61 170	82.10 179	85.37 146	87.34 178	92.94 156	87.48 174	85.72 123	78.92 176	73.87 135	65.71 183	64.69 188	81.78 175	87.82 183	89.35 178	96.01 165	95.26 140

v1921920		83.30 174	82.09 180	84.70 155	86.59 193	92.67 164	89.82 149	82.23 161	78.32 179	65.76 186	64.64 190	62.35 195	86.78 135	90.34 157	90.02 167	97.02 136	96.31 111

TinyColmap		84.04 164	82.01 181	86.42 139	90.87 138	91.84 182	88.89 164	84.07 139	82.11 160	69.89 159	71.08 157	60.81 203	89.04 112	90.52 155	89.19 179	95.76 168	88.50 197

tpm cat1		84.13 162	81.99 182	86.63 137	91.74 125	91.50 188	90.68 128	75.69 193	86.12 127	85.44 86	72.39 153	70.72 155	85.16 150	80.89 208	81.56 204	91.07 205	90.71 183

LTVRE_ROB	81.71 16	82.44 185	81.84 183	83.13 175	89.01 153	92.99 154	88.90 163	82.32 160	66.26 212	54.02 212	74.68 140	59.62 209	88.87 117	90.71 152	92.02 126	95.68 173	96.62 97
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
v1240		82.88 180	81.66 184	84.29 161	86.46 194	92.52 170	89.06 160	81.82 166	77.16 186	65.09 190	64.17 192	61.50 200	86.36 137	90.12 161	90.13 161	96.95 141	96.04 120

PEN-MVS		82.49 184	81.58 185	83.56 171	86.93 187	92.05 179	86.71 182	83.84 141	76.94 188	64.68 191	67.24 171	60.11 206	81.17 178	87.78 184	90.70 150	98.02 82	96.21 114

v7n		82.25 186	81.54 186	83.07 178	85.55 199	92.58 166	86.68 183	81.10 174	76.54 189	65.97 185	62.91 194	60.56 204	82.36 169	91.07 146	90.35 156	96.77 154	96.80 93

PS-CasMVS		82.53 183	81.54 186	83.68 169	87.08 186	92.54 168	86.20 186	83.46 148	76.46 191	65.73 187	65.71 183	59.41 210	81.61 176	89.06 177	90.55 153	98.03 81	97.07 87

UniMVSNet_ETH3D		84.57 154	81.40 188	88.28 117	89.34 152	94.38 120	90.33 132	86.50 114	74.74 199	77.52 122	59.90 201	62.04 198	88.78 119	88.82 180	92.65 113	97.22 123	97.24 80

EG-PatchMatch MVS		81.70 190	81.31 189	82.15 188	88.75 155	93.81 128	87.14 178	78.89 182	71.57 204	64.12 194	61.20 199	68.46 165	76.73 192	91.48 137	90.77 146	97.28 121	91.90 173

tfpnnormal		83.80 168	81.26 190	86.77 135	89.60 149	93.26 147	89.72 151	87.60 109	72.78 201	70.44 155	60.53 200	61.15 202	85.55 146	92.72 116	91.44 137	97.71 102	96.92 91

DTE-MVSNet		81.76 189	81.04 191	82.60 185	86.63 191	91.48 190	85.97 188	83.70 143	76.45 192	62.44 196	67.16 172	59.98 207	78.98 185	87.15 188	89.93 170	97.88 92	95.12 142

MDTV_nov1_ep13_2view		80.43 192	80.94 192	79.84 194	84.82 202	90.87 193	84.23 193	73.80 198	80.28 169	64.33 192	70.05 164	68.77 164	79.67 181	84.83 198	83.50 201	92.17 199	88.25 200

TransMVSNet (Re)		82.67 182	80.93 193	84.69 156	88.71 156	91.50 188	87.90 171	87.15 110	71.54 206	68.24 171	63.69 193	64.67 189	78.51 187	91.65 136	90.73 149	97.64 110	92.73 172

EU-MVSNet		78.43 197	80.25 194	76.30 201	83.81 204	87.27 208	80.99 201	79.52 179	76.01 193	54.12 211	70.44 161	64.87 186	67.40 205	86.23 193	85.54 193	91.95 202	91.41 176

pmnet_mix02		80.14 194	80.21 195	80.06 193	86.61 192	89.66 199	80.40 203	82.20 162	82.29 159	61.35 199	71.52 155	66.67 176	76.75 191	82.55 205	80.18 208	93.05 193	88.62 195

PM-MVS		80.29 193	79.30 196	81.45 192	81.91 207	88.23 203	82.61 197	79.01 181	79.99 172	67.15 179	69.07 166	51.39 215	82.92 166	87.55 186	85.59 191	95.08 184	93.28 165

pmmvs6		80.90 191	78.77 197	83.38 174	85.84 196	91.61 186	86.01 187	82.54 156	64.17 213	70.43 156	54.14 210	67.06 173	80.73 180	90.50 156	89.17 180	94.74 188	94.75 146

gm-plane-assit		77.65 200	78.50 198	76.66 200	87.96 167	85.43 210	64.70 216	74.50 195	64.15 214	51.26 215	61.32 198	58.17 211	84.11 160	95.16 65	93.83 81	97.45 117	91.41 176

test20.03		76.41 202	78.49 199	73.98 203	85.64 198	87.50 205	75.89 208	80.71 175	70.84 207	51.07 216	68.06 169	61.40 201	54.99 212	88.28 181	87.20 186	95.58 177	86.15 202

Anonymous20231206		78.09 199	78.11 200	78.07 199	85.19 201	89.17 200	80.99 201	81.24 173	75.46 197	58.25 206	54.78 209	59.90 208	66.73 206	88.94 179	88.26 183	96.01 165	90.25 187

pmmvs-eth3d		79.78 196	77.58 201	82.34 187	81.57 208	87.46 206	82.92 196	81.28 171	75.33 198	71.34 148	61.88 195	52.41 214	81.59 177	87.56 185	86.90 187	95.36 182	91.48 175

MVS-HIRNet		78.16 198	77.57 202	78.83 197	85.83 197	87.76 204	76.67 207	70.22 209	75.82 196	67.39 176	55.61 205	70.52 156	81.96 173	86.67 192	85.06 196	90.93 206	81.58 209

CMPMVS	61.19 17	79.86 195	77.46 203	82.66 184	91.54 130	91.82 183	83.25 195	81.57 168	70.51 208	68.64 168	59.89 202	66.77 175	79.63 182	84.00 202	84.30 198	91.34 203	84.89 206
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
N_pmnet		77.55 201	76.68 204	78.56 198	85.43 200	87.30 207	78.84 205	81.88 165	78.30 180	60.61 200	61.46 196	62.15 196	74.03 200	82.04 206	80.69 207	90.59 208	84.81 207

MIMVSNet1		73.19 204	73.70 205	72.60 206	65.42 218	86.69 209	75.56 209	79.65 178	67.87 211	55.30 208	45.24 214	56.41 212	63.79 208	86.98 189	87.66 185	95.85 167	85.04 205

new_pmnet		72.29 206	73.25 206	71.16 208	75.35 212	81.38 212	73.72 212	69.27 210	75.97 194	49.84 217	56.27 204	56.12 213	69.08 202	81.73 207	80.86 206	89.72 211	80.44 211

MDA-MVSNet-bldmvs		73.81 203	72.56 207	75.28 202	72.52 215	88.87 201	74.95 210	82.67 154	71.57 204	55.02 209	65.96 181	42.84 221	76.11 193	70.61 214	81.47 205	90.38 209	86.59 201

new-patchmatchnet		72.32 205	71.09 208	73.74 204	81.17 209	84.86 211	72.21 213	77.48 188	68.32 210	54.89 210	55.10 207	49.31 218	63.68 209	79.30 210	76.46 211	93.03 194	84.32 208

pmmvs3		71.13 207	71.06 209	71.21 207	73.54 214	80.19 213	71.69 214	64.86 213	62.04 216	52.10 213	54.92 208	48.00 219	75.03 196	83.75 203	83.24 202	90.04 210	85.27 204

ambc			67.96 210		73.69 213	79.79 214	73.82 211		71.61 203	59.80 203	46.00 213	20.79 223	66.15 207	86.92 190	80.11 209	89.13 212	90.50 184

FPMVS		69.87 208	67.10 211	73.10 205	84.09 203	78.35 215	79.40 204	76.41 191	71.92 202	57.71 207	54.06 211	50.04 216	56.72 210	71.19 213	68.70 213	84.25 213	75.43 213

test_method		58.10 212	64.61 212	50.51 212	28.26 223	41.71 222	61.28 217	32.07 219	75.92 195	52.04 214	47.94 212	61.83 199	51.80 213	79.83 209	63.95 217	77.60 217	81.05 210

PMVS	56.77 18	61.27 210	58.64 213	64.35 209	75.66 211	54.60 219	53.62 219	74.23 196	53.69 217	58.37 205	44.27 215	49.38 217	44.16 216	69.51 215	65.35 215	80.07 215	73.66 214
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
Gipuma		58.52 211	56.17 214	61.27 210	67.14 217	58.06 218	52.16 220	68.40 212	69.00 209	45.02 219	22.79 217	20.57 224	55.11 211	76.27 211	79.33 210	79.80 216	67.16 216
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
PMMVS2		53.68 213	55.72 215	51.30 211	58.84 219	67.02 217	54.23 218	60.97 216	47.50 218	19.42 222	34.81 216	31.97 222	30.88 218	65.84 216	69.99 212	83.47 214	72.92 215

MVE	39.81 19	39.52 215	41.58 216	37.11 216	33.93 222	49.06 220	26.45 224	54.22 217	29.46 221	24.15 221	20.77 219	10.60 227	34.42 217	51.12 218	65.27 216	49.49 222	64.81 217
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
E-PMN		40.00 214	35.74 217	44.98 214	57.69 221	39.15 224	28.05 222	62.70 214	35.52 220	17.78 223	20.90 218	14.36 226	44.47 215	35.89 219	47.86 218	59.15 220	56.47 218

EMVS		39.04 216	34.32 218	44.54 215	58.25 220	39.35 223	27.61 223	62.55 215	35.99 219	16.40 224	20.04 220	14.77 225	44.80 214	33.12 220	44.10 219	57.61 221	52.89 219

testmvs		4.35 217	6.54 219	1.79 218	0.60 224	1.82 225	3.06 226	0.95 221	7.22 222	0.88 226	12.38 221	1.25 228	3.87 221	6.09 221	5.58 220	1.40 223	11.42 221

test123		3.48 218	5.31 220	1.34 219	0.20 226	1.52 226	2.17 227	0.58 222	6.13 223	0.31 227	9.85 222	0.31 229	3.90 220	2.65 222	5.28 221	0.87 224	11.46 220

uanet_test		0.00 219	0.00 221	0.00 220	0.00 227	0.00 227	0.00 228	0.00 224	0.00 224	0.00 228	0.00 223	0.00 230	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	0.00 227	0.00 227	0.00 228	0.00 224	0.00 224	0.00 228	0.00 223	0.00 230	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	0.00 227	0.00 227	0.00 228	0.00 224	0.00 224	0.00 228	0.00 223	0.00 230	0.00 223	0.00 223	0.00 222	0.00 225	0.00 222

RE-MVS-def										60.19 201

9.14												97.28 23

SR-MVS					98.93 18			96.00 16				97.75 15

our_test_3					86.93 187	89.77 198	81.61 200

MTAPA										95.36 2		97.46 21

MTMP										95.70 1		96.90 26

Patchmatch-RL test							18.47 225

tmp_tt				50.24 213	68.55 216	46.86 221	48.90 221	18.28 220	86.51 125	68.32 170	70.19 163	65.33 181	26.69 219	74.37 212	66.80 214	70.72 219

XVS					95.68 63	98.66 14	94.96 61			88.03 53		96.06 31				98.46 34

X-MVStestdata					95.68 63	98.66 14	94.96 61			88.03 53		96.06 31				98.46 34

mPP-MVS					98.76 23							95.49 38

NP-MVS									91.63 65

Patchmtry						92.39 172	89.18 157	73.30 202		71.08 151

DeepMVS_CX						71.82 216	68.37 215	48.05 218	77.38 184	46.88 218	65.77 182	47.03 220	67.48 204	64.27 217		76.89 218	76.72 212

High-resolution multi-view benchmark