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.

LTVRE_ROB	95.06 1	97.73 1	98.39 1	96.95 1	96.33 51	96.94 36	98.30 20	94.90 15	98.61 1	97.73 3	97.97 25	98.57 26	95.74 4	99.24 1	98.70 4	98.72 7	98.70 2
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
TDRefinement		97.59 2	98.32 2	96.73 4	95.90 66	98.10 2	99.08 2	93.92 31	98.24 3	96.44 13	98.12 20	97.86 60	96.06 2	99.24 1	98.93 1	99.00 2	97.77 5

WR-MVS		97.53 3	98.20 3	96.76 3	96.93 29	98.17 1	98.60 10	96.67 7	96.39 15	94.46 32	99.14 1	98.92 12	94.57 15	99.06 3	98.80 2	99.32 1	96.92 28

SixPastTwentyTwo		97.36 4	97.73 10	96.92 2	97.36 13	96.15 57	98.29 21	94.43 23	96.50 13	96.96 7	98.74 5	98.74 21	96.04 3	99.03 5	97.74 16	98.44 23	97.22 14

PS-CasMVS		97.22 5	97.84 7	96.50 5	97.08 25	97.92 6	98.17 30	97.02 2	94.71 27	95.32 21	98.52 12	98.97 11	92.91 42	99.04 4	98.47 5	98.49 19	97.24 13

PEN-MVS		97.16 6	97.87 6	96.33 11	97.20 21	97.97 4	98.25 25	96.86 6	95.09 25	94.93 26	98.66 7	99.16 5	92.27 52	98.98 6	98.39 7	98.49 19	96.83 32

DTE-MVSNet		97.16 6	97.75 9	96.47 6	97.40 12	97.95 5	98.20 28	96.89 5	95.30 20	95.15 24	98.66 7	98.80 18	92.77 46	98.97 7	98.27 9	98.44 23	96.28 42

COLMAP_ROB	93.74 2	97.09 8	97.98 4	96.05 17	95.97 63	97.78 9	98.56 11	91.72 86	97.53 7	96.01 15	98.14 19	98.76 20	95.28 5	98.76 11	98.23 10	98.77 5	96.67 36
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
WR-MVS_H		97.06 9	97.78 8	96.23 13	96.74 37	98.04 3	98.25 25	97.32 1	94.40 34	93.71 51	98.55 10	98.89 14	92.97 39	98.91 9	98.45 6	98.38 28	97.19 15

CP-MVSNet		96.97 10	97.42 14	96.44 7	97.06 26	97.82 8	98.12 33	96.98 3	93.50 48	95.21 23	97.98 24	98.44 29	92.83 45	98.93 8	98.37 8	98.46 22	96.91 29

DVP-MVS++		96.63 11	97.92 5	95.12 40	97.77 6	97.52 15	98.29 21	93.83 34	96.72 9	92.52 75	98.10 21	99.07 10	90.87 78	97.83 31	97.44 28	97.44 61	98.76 1

ACMH	90.17 8	96.61 12	97.69 12	95.35 30	95.29 81	96.94 36	98.43 14	92.05 74	98.04 4	95.38 19	98.07 22	99.25 4	93.23 33	98.35 16	97.16 39	97.72 51	96.00 48
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
UA-Net		96.56 13	96.73 24	96.36 9	98.99 1	97.90 7	97.79 43	95.64 10	92.78 65	92.54 74	96.23 78	95.02 137	94.31 18	98.43 15	98.12 11	98.89 3	98.58 3

ACMMPR		96.54 14	96.71 25	96.35 10	97.55 9	97.63 11	98.62 9	94.54 19	94.45 31	94.19 38	95.04 105	97.35 75	94.92 10	97.85 28	97.50 25	98.26 29	97.17 16

v7n		96.49 15	97.20 18	95.65 22	95.57 76	96.04 59	97.93 38	92.49 58	96.40 14	97.13 6	98.99 2	99.41 3	93.79 25	97.84 30	96.15 65	97.00 83	95.60 56

DeepC-MVS	92.47 4	96.44 16	96.75 23	96.08 16	97.57 7	97.19 31	97.96 37	94.28 24	95.29 21	94.92 27	98.31 17	96.92 85	93.69 27	96.81 67	96.50 56	98.06 40	96.27 43
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
ACMM	90.06 9	96.31 17	96.42 32	96.19 14	97.21 20	97.16 33	98.71 5	93.79 37	94.35 35	93.81 45	92.80 140	98.23 40	95.11 6	98.07 20	97.45 27	98.51 18	96.86 31
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
ACMH+	89.90 10	96.27 18	97.52 13	94.81 47	95.19 84	97.18 32	97.97 36	92.52 56	96.72 9	90.50 122	97.31 50	99.11 8	94.10 19	98.67 12	97.90 14	98.56 15	95.79 52

APDe-MVS		96.23 19	97.22 17	95.08 41	96.66 41	97.56 14	98.63 8	93.69 41	94.62 29	89.80 131	97.73 33	98.13 44	93.84 24	97.79 33	97.63 18	97.87 47	97.08 23
Zhaojie Zeng, Yuesong Wang, Tao Guan: Matching Ambiguity-Resilient Multi-View Stereo via Adaptive Patch Deformation. Pattern Recognition
CP-MVS		96.21 20	96.16 43	96.27 12	97.56 8	97.13 34	98.43 14	94.70 18	92.62 69	94.13 40	92.71 141	98.03 50	94.54 16	98.00 24	97.60 20	98.23 31	97.05 24

HFP-MVS		96.18 21	96.53 29	95.77 20	97.34 16	97.26 28	98.16 31	94.54 19	94.45 31	92.52 75	95.05 103	96.95 84	93.89 22	97.28 49	97.46 26	98.19 33	97.25 11

UniMVSNet_ETH3D		96.15 22	97.71 11	94.33 55	97.31 17	96.71 41	95.06 107	96.91 4	97.86 5	90.42 123	98.55 10	99.60 1	88.01 117	98.51 13	97.81 15	98.26 29	94.95 68

MP-MVS		96.13 23	95.93 47	96.37 8	98.19 3	97.31 27	98.49 13	94.53 22	91.39 103	94.38 34	94.32 119	96.43 99	94.59 14	97.75 35	97.44 28	98.04 41	96.88 30
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
ACMMP		96.12 24	96.27 39	95.93 18	97.20 21	97.60 12	98.64 7	93.74 38	92.47 73	93.13 65	93.23 133	98.06 47	94.51 17	97.99 25	97.57 22	98.39 27	96.99 25
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
DVP-MVS		96.10 25	97.23 16	94.79 49	96.28 54	97.49 16	97.90 39	93.60 43	95.47 18	89.57 137	97.32 49	97.72 63	93.89 22	97.74 36	97.53 23	97.51 57	97.34 9
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
LGP-MVS_train		96.10 25	96.29 36	95.87 19	96.72 38	97.35 26	98.43 14	93.83 34	90.81 118	92.67 73	95.05 103	98.86 16	95.01 7	98.11 18	97.37 35	98.52 17	96.50 38

CSCG		96.07 27	97.15 19	94.81 47	96.06 62	97.58 13	96.52 74	90.98 97	96.51 12	93.60 53	97.13 57	98.55 27	93.01 37	97.17 53	95.36 82	98.68 9	97.78 4

DPE-MVS		96.00 28	96.80 22	95.06 42	95.87 69	97.47 21	98.25 25	93.73 39	92.38 76	91.57 103	97.55 43	97.97 53	92.98 38	97.49 47	97.61 19	97.96 45	97.16 17
Kehua Chen, Zhenlong Yuan, Tianlu Mao, Zhaoqi Wang: Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization. AAAI2025
SMA-MVS		95.99 29	96.48 30	95.41 29	97.43 11	97.36 24	97.55 48	93.70 40	94.05 42	93.79 46	97.02 60	94.53 142	92.28 51	97.53 45	97.19 37	97.73 50	97.67 7
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
TSAR-MVS + MP.		95.99 29	96.57 28	95.31 32	96.87 30	96.50 48	98.71 5	91.58 87	93.25 54	92.71 70	96.86 62	96.57 97	93.92 20	98.09 19	97.91 13	98.08 38	96.81 33
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
OPM-MVS		95.96 31	96.59 27	95.23 35	96.67 40	96.52 47	97.86 41	93.28 47	95.27 23	93.46 55	96.26 75	98.85 17	92.89 43	97.09 54	96.37 60	97.22 75	95.78 53
Ray L. Khuboni, Hongjun Xu: Octagram Propagation Matching for Multi-Scale View Stereopsis (OPM-MVS).
SteuartSystems-ACMMP		95.96 31	96.13 44	95.76 21	97.06 26	97.36 24	98.40 18	94.24 26	91.49 97	91.91 93	94.50 115	96.89 86	94.99 8	98.01 23	97.44 28	97.97 44	97.25 11
Skip Steuart: Steuart Systems R&D Blog.
ACMP	89.62 11	95.96 31	96.28 37	95.59 23	96.58 43	97.23 30	98.26 24	93.22 48	92.33 80	92.31 83	94.29 120	98.73 22	94.68 12	98.04 21	97.14 40	98.47 21	96.17 45
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
PGM-MVS		95.90 34	95.72 51	96.10 15	97.53 10	97.45 22	98.55 12	94.12 28	90.25 122	93.71 51	93.20 134	97.18 79	94.63 13	97.68 39	97.34 36	98.08 38	96.97 26

PMVS	87.16 16	95.88 35	96.47 31	95.19 37	97.00 28	96.02 60	96.70 65	91.57 88	94.43 33	95.33 20	97.16 56	95.37 125	92.39 48	98.89 10	98.72 3	98.17 35	94.71 74
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
ACMMP_NAP		95.86 36	96.18 40	95.47 28	97.11 24	97.26 28	98.37 19	93.48 45	93.49 49	93.99 43	95.61 87	94.11 146	92.49 47	97.87 27	97.44 28	97.40 64	97.52 8

Gipuma		95.86 36	96.17 41	95.50 27	95.92 65	94.59 106	94.77 118	92.50 57	97.82 6	97.90 2	95.56 91	97.88 58	94.71 11	98.02 22	94.81 97	97.23 74	94.48 80
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
LS3D		95.83 38	96.35 34	95.22 36	96.47 47	97.49 16	97.99 34	92.35 61	94.92 26	94.58 30	94.88 109	95.11 135	91.52 63	98.48 14	98.05 12	98.42 25	95.49 57

SD-MVS		95.77 39	96.17 41	95.30 33	96.72 38	96.19 56	97.01 57	93.04 49	94.03 43	92.71 70	96.45 73	96.78 93	93.91 21	96.79 68	95.89 71	98.42 25	97.09 22
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
SED-MVS		95.73 40	96.98 20	94.28 56	96.08 60	97.39 23	98.18 29	93.80 36	94.20 37	89.61 136	97.29 51	97.49 71	90.69 82	97.74 36	97.41 32	97.32 69	97.34 9

TranMVSNet+NR-MVSNet		95.72 41	96.42 32	94.91 46	96.21 55	96.77 40	96.90 62	94.99 13	92.62 69	91.92 92	98.51 13	98.63 24	90.82 79	97.27 50	96.83 45	98.63 12	94.31 81

DU-MVS		95.51 42	95.68 52	95.33 31	96.45 48	96.44 50	96.61 71	95.32 11	89.97 128	93.78 47	97.46 45	98.07 46	91.19 70	97.03 57	96.53 53	98.61 13	94.22 82

UniMVSNet (Re)		95.46 43	95.86 49	95.00 45	96.09 58	96.60 42	96.68 69	94.99 13	90.36 121	92.13 86	97.64 39	98.13 44	91.38 64	96.90 62	96.74 47	98.73 6	94.63 76

RPSCF		95.46 43	96.95 21	93.73 79	95.72 73	95.94 64	95.58 98	88.08 147	95.31 19	91.34 106	96.26 75	98.04 49	93.63 28	98.28 17	97.67 17	98.01 42	97.13 18

anonymousdsp		95.45 45	96.70 26	93.99 67	88.43 213	92.05 157	99.18 1	85.42 186	94.29 36	96.10 14	98.63 9	99.08 9	96.11 1	97.77 34	97.41 32	98.70 8	97.69 6

APD-MVS		95.38 46	95.68 52	95.03 43	97.30 18	96.90 38	97.83 42	93.92 31	89.40 135	90.35 124	95.41 95	97.69 65	92.97 39	97.24 52	97.17 38	97.83 48	95.96 49
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
UniMVSNet_NR-MVSNet		95.34 47	95.51 56	95.14 39	95.80 71	96.55 43	96.61 71	94.79 16	90.04 127	93.78 47	97.51 44	97.25 76	91.19 70	96.68 70	96.31 62	98.65 11	94.22 82

X-MVS		95.33 48	95.13 64	95.57 25	97.35 14	97.48 18	98.43 14	94.28 24	92.30 81	93.28 58	86.89 198	96.82 89	91.87 57	97.85 28	97.59 21	98.19 33	96.95 27

MSP-MVS		95.32 49	96.28 37	94.19 59	96.87 30	97.77 10	98.27 23	93.88 33	94.15 41	89.63 135	95.36 96	98.37 32	90.73 80	94.37 118	97.53 23	95.77 122	96.40 39
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
3Dnovator+	92.82 3	95.22 50	95.16 62	95.29 34	96.17 56	96.55 43	97.64 45	94.02 30	94.16 40	94.29 36	92.09 147	93.71 153	91.90 55	96.68 70	96.51 54	97.70 53	96.40 39

HPM-MVS++		95.21 51	94.89 67	95.59 23	97.79 5	95.39 82	97.68 44	94.05 29	91.91 89	94.35 35	93.38 131	95.07 136	92.94 41	96.01 83	95.88 72	96.73 87	96.61 37

TSAR-MVS + ACMM		95.17 52	95.95 45	94.26 57	96.07 61	96.46 49	95.67 96	94.21 27	93.84 45	90.99 114	97.18 54	95.24 133	93.55 29	96.60 73	95.61 79	95.06 141	96.69 35

CPTT-MVS		95.00 53	94.52 78	95.57 25	96.84 34	96.78 39	97.88 40	93.67 42	92.20 82	92.35 82	85.87 205	97.56 70	94.98 9	96.96 60	96.07 68	97.70 53	96.18 44

SF-MVS		94.88 54	95.87 48	93.73 79	95.30 79	95.93 65	94.80 117	91.76 84	93.11 58	91.93 91	95.83 83	97.07 81	91.11 73	96.62 72	96.44 58	97.46 58	96.13 46

Baseline_NR-MVSNet		94.85 55	95.35 60	94.26 57	96.45 48	93.86 122	96.70 65	94.54 19	90.07 126	90.17 128	98.77 4	97.89 55	90.64 85	97.03 57	96.16 64	97.04 82	93.67 95

EG-PatchMatch MVS		94.81 56	95.53 55	93.97 68	95.89 68	94.62 104	95.55 100	88.18 145	92.77 66	94.88 28	97.04 59	98.61 25	93.31 30	96.89 63	95.19 87	95.99 115	93.56 98

CS-MVS		94.76 57	94.41 82	95.18 38	94.95 90	95.99 61	97.28 50	91.99 76	85.51 168	94.55 31	93.07 136	97.69 65	93.77 26	97.08 55	96.79 46	98.53 16	94.72 72

OMC-MVS		94.74 58	95.46 58	93.91 71	94.62 102	96.26 54	96.64 70	89.36 132	94.20 37	94.15 39	94.02 124	97.73 62	91.34 66	96.15 80	95.04 91	97.37 66	94.80 70

DeepC-MVS_fast	91.38 6	94.73 59	94.98 65	94.44 51	96.83 36	96.12 58	96.69 67	92.17 67	92.98 63	93.72 49	94.14 121	95.45 123	90.49 91	95.73 90	95.30 84	96.71 88	95.13 66
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
PHI-MVS		94.65 60	94.84 69	94.44 51	94.95 90	96.55 43	96.46 77	91.10 95	88.96 138	96.00 16	94.55 114	95.32 128	90.67 83	96.97 59	96.69 51	97.44 61	94.84 69

SPE-MVS-test		94.63 61	94.30 88	95.02 44	94.63 100	95.71 72	98.15 32	92.13 69	85.62 167	94.22 37	93.63 129	97.63 69	93.08 36	97.50 46	96.51 54	97.88 46	93.50 99

pmmvs6		94.58 62	97.30 15	91.40 123	94.84 94	94.61 105	93.40 155	92.43 60	98.51 2	85.61 168	98.73 6	99.53 2	84.40 150	97.88 26	97.03 41	97.72 51	94.79 71

DeepPCF-MVS	90.68 7	94.56 63	94.92 66	94.15 60	94.11 115	95.71 72	97.03 56	90.65 102	93.39 52	94.08 41	95.29 100	94.15 145	93.21 34	95.22 103	94.92 95	95.82 121	95.75 54

NR-MVSNet		94.55 64	95.66 54	93.25 91	94.26 111	96.44 50	96.69 67	95.32 11	89.97 128	91.79 98	97.46 45	98.39 31	82.85 162	96.87 65	96.48 57	98.57 14	93.98 88

MVS_0304		94.43 65	94.78 72	94.02 64	96.14 57	97.09 35	97.52 49	92.66 54	90.12 124	93.12 66	95.31 98	93.19 158	87.75 119	96.14 81	95.60 80	96.96 84	96.01 47

Vis-MVSNet		94.39 66	95.85 50	92.68 99	90.91 196	95.88 67	97.62 47	91.41 89	91.95 88	89.20 141	97.29 51	96.26 102	90.60 90	96.95 61	95.91 69	96.32 102	96.71 34
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
TSAR-MVS + GP.		94.25 67	94.81 71	93.60 81	96.52 46	95.80 70	94.37 129	92.47 59	90.89 114	88.92 143	95.34 97	94.38 143	92.85 44	96.36 78	95.62 78	96.47 95	95.28 63

CNVR-MVS		94.24 68	94.47 79	93.96 69	96.56 44	95.67 74	96.43 78	91.95 78	92.08 85	91.28 108	90.51 157	95.35 126	91.20 69	96.34 79	95.50 81	96.34 100	95.88 51

EC-MVSNet		94.23 69	93.81 107	94.71 50	94.85 93	96.23 55	97.14 52	93.40 46	81.79 191	91.58 102	93.29 132	95.21 134	93.13 35	97.73 38	96.95 42	98.20 32	95.45 58

v1192		93.98 70	93.94 100	94.01 65	93.91 124	94.63 103	97.00 58	89.75 122	91.01 112	96.50 10	97.93 26	98.26 38	91.74 59	92.06 155	92.05 142	95.18 136	91.66 140

v10		93.96 71	94.12 95	93.77 78	93.37 138	95.45 78	96.83 64	91.13 94	89.70 132	95.02 25	97.88 29	98.23 40	91.27 67	92.39 150	92.18 137	94.99 144	93.00 108

CDPH-MVS		93.96 71	93.86 102	94.08 62	96.31 52	95.84 68	96.92 60	91.85 81	87.21 155	91.25 110	92.83 138	96.06 110	91.05 75	95.57 93	94.81 97	97.12 77	94.72 72

MSLP-MVS++		93.91 73	94.30 88	93.45 83	95.51 77	95.83 69	93.12 162	91.93 80	91.45 100	91.40 105	87.42 193	96.12 109	93.27 31	96.57 74	96.40 59	95.49 125	96.29 41

v1921920		93.90 74	93.82 105	94.00 66	93.74 130	94.31 110	97.12 53	89.33 133	91.13 109	96.77 9	97.90 27	98.06 47	91.95 54	91.93 160	91.54 151	95.10 139	91.85 132

train_agg		93.89 75	93.46 117	94.40 53	97.35 14	93.78 124	97.63 46	92.19 66	88.12 145	90.52 121	93.57 130	95.78 116	92.31 50	94.78 111	93.46 121	96.36 98	94.70 75

v144192		93.89 75	93.85 103	93.94 70	93.50 135	94.33 109	97.12 53	89.49 127	90.89 114	96.49 11	97.78 31	98.27 37	91.89 56	92.17 154	91.70 148	95.19 135	91.78 135

v1240		93.89 75	93.72 108	94.09 61	93.98 120	94.31 110	97.12 53	89.37 131	90.74 120	96.92 8	98.05 23	97.89 55	92.15 53	91.53 167	91.60 149	94.99 144	91.93 130

NCCC		93.87 78	93.42 118	94.40 53	96.84 34	95.42 79	96.47 76	92.62 55	92.36 78	92.05 88	83.83 213	95.55 119	91.84 58	95.89 85	95.23 86	96.56 92	95.63 55

v1144		93.83 79	93.87 101	93.78 77	93.72 131	94.57 107	96.85 63	89.98 116	91.31 105	95.90 17	97.89 28	98.40 30	91.13 72	92.01 158	92.01 143	95.10 139	90.94 150

MVS_111021_HR		93.82 80	94.26 91	93.31 86	95.01 88	93.97 120	95.73 93	89.75 122	92.06 86	92.49 77	94.01 125	96.05 111	90.61 89	95.95 84	94.78 100	96.28 103	93.04 107

thisisatest0515		93.79 81	94.41 82	93.06 96	94.14 112	92.50 149	95.56 99	88.55 141	91.61 93	92.45 78	96.84 63	95.71 117	90.62 87	94.58 114	95.07 89	97.05 80	94.58 77

TAPA-MVS	88.94 13	93.78 82	94.31 87	93.18 93	94.14 112	95.99 61	95.74 92	86.98 168	93.43 51	93.88 44	90.16 164	96.88 87	91.05 75	94.33 119	93.95 112	97.28 72	95.40 59
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
GeoE		93.72 83	93.62 112	93.84 72	94.75 97	94.90 97	97.24 51	91.81 83	86.97 159	92.74 69	93.83 127	97.24 78	90.46 92	95.10 107	94.09 111	96.08 112	93.18 105

EPP-MVSNet		93.63 84	93.95 99	93.26 89	95.15 85	96.54 46	96.18 85	91.97 77	91.74 90	85.76 166	94.95 107	84.27 200	91.60 62	97.61 43	97.38 34	98.87 4	95.18 65

v8		93.60 85	93.82 105	93.34 84	93.13 147	95.06 90	96.39 79	90.75 100	89.90 130	94.03 42	97.70 35	98.21 42	91.08 74	92.36 151	91.47 152	94.63 156	92.07 126

MCST-MVS		93.60 85	93.40 120	93.83 73	95.30 79	95.40 81	96.49 75	90.87 98	90.08 125	91.72 99	90.28 162	95.99 112	91.69 60	93.94 130	92.99 128	96.93 85	95.13 66

PVSNet_Blended_VisFu		93.60 85	93.41 119	93.83 73	96.31 52	95.65 75	95.71 94	90.58 104	88.08 147	93.17 63	95.29 100	92.20 163	90.72 81	94.69 113	93.41 123	96.51 94	94.54 78

TransMVSNet (Re)		93.55 88	96.32 35	90.32 143	94.38 107	94.05 115	93.30 159	89.53 126	97.15 8	85.12 172	98.83 3	97.89 55	82.21 169	96.75 69	96.14 66	97.35 67	93.46 100

DCV-MVSNet		93.49 89	95.15 63	91.55 117	94.05 116	95.92 66	95.15 105	91.21 91	92.76 67	87.01 162	89.71 168	97.16 80	83.90 156	97.65 40	96.87 44	97.99 43	95.95 50

v2v482		93.42 90	93.49 116	93.32 85	93.44 137	94.05 115	96.36 82	89.76 121	91.41 102	95.24 22	97.63 40	98.34 34	90.44 93	91.65 165	91.76 147	94.69 153	89.62 163

sasdasda		93.38 91	94.36 84	92.24 105	93.94 122	96.41 52	94.18 138	90.47 105	93.07 61	88.47 151	88.66 179	93.78 150	88.80 106	95.74 88	95.75 75	97.57 55	97.13 18

canonicalmvs		93.38 91	94.36 84	92.24 105	93.94 122	96.41 52	94.18 138	90.47 105	93.07 61	88.47 151	88.66 179	93.78 150	88.80 106	95.74 88	95.75 75	97.57 55	97.13 18

3Dnovator	91.81 5	93.36 93	94.27 90	92.29 104	92.99 154	95.03 91	95.76 91	87.79 151	93.82 46	92.38 81	92.19 146	93.37 157	88.14 116	95.26 102	94.85 96	96.69 89	95.40 59

pm-mvs1		93.27 94	95.94 46	90.16 144	94.13 114	93.66 126	92.61 175	89.91 118	95.73 17	84.28 183	98.51 13	98.29 36	82.80 163	96.44 76	95.76 74	97.25 73	93.21 104

casdiffmvs_mvg		93.27 94	94.83 70	91.45 121	93.59 133	94.47 108	94.91 113	89.83 120	92.04 87	87.14 160	97.57 42	98.47 28	86.03 139	94.07 128	94.44 107	97.21 76	92.76 114
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
test1111		93.25 96	94.43 80	91.88 111	95.09 87	94.97 95	94.58 124	92.81 51	93.60 47	83.79 188	97.17 55	89.25 185	87.59 121	97.54 44	96.57 52	97.42 63	91.89 131

Anonymous20231211		93.19 97	95.50 57	90.49 140	93.77 128	95.29 84	94.36 133	90.04 115	91.44 101	84.59 178	96.72 66	97.65 67	82.45 168	97.25 51	96.32 61	97.74 49	93.79 91

TinyColmap		93.17 98	93.33 121	93.00 97	93.84 126	92.76 143	94.75 120	88.90 137	93.97 44	97.48 4	95.28 102	95.29 129	88.37 112	95.31 101	91.58 150	94.65 155	89.10 167

viewmacassd2359aftdt		93.16 99	94.69 75	91.39 124	93.30 140	93.71 125	95.03 109	87.70 152	94.69 28	89.53 138	97.63 40	98.92 12	87.73 120	93.63 134	92.14 139	95.05 142	92.08 125

MVS_111021_LR		93.15 100	93.65 110	92.56 100	93.89 125	92.28 152	95.09 106	86.92 170	91.26 108	92.99 68	94.46 117	96.22 105	90.64 85	95.11 106	93.45 122	95.85 119	92.74 115

CNLPA		93.14 101	93.67 109	92.53 101	94.62 102	94.73 100	95.00 111	86.57 175	92.85 64	92.43 79	90.94 152	94.67 139	90.35 94	95.41 96	93.70 118	96.23 106	93.37 102

PLC	87.27 15	93.08 102	92.92 128	93.26 89	94.67 98	95.03 91	94.38 128	90.10 110	91.69 91	92.14 85	87.24 194	93.91 148	91.61 61	95.05 108	94.73 103	96.67 90	92.80 111
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
CANet		93.07 103	93.05 127	93.10 94	95.90 66	95.41 80	95.88 88	91.94 79	84.77 174	93.36 56	94.05 123	95.25 132	86.25 135	94.33 119	93.94 113	95.30 129	93.58 97

TSAR-MVS + COLMAP		93.06 104	93.65 110	92.36 102	94.62 102	94.28 112	95.36 104	89.46 129	92.18 83	91.64 100	95.55 92	95.27 131	88.60 110	93.24 137	92.50 133	94.46 160	92.55 121

ECVR-MVS		93.05 105	94.25 92	91.65 114	94.76 95	95.23 85	94.26 136	92.80 52	92.49 71	83.90 186	96.75 65	89.99 176	86.84 128	97.62 41	96.72 48	97.32 69	90.92 151

Effi-MVS+		92.93 106	92.16 140	93.83 73	94.29 109	93.53 134	95.04 108	92.98 50	85.27 171	94.46 32	90.24 163	95.34 127	89.99 97	93.72 131	94.23 110	96.22 107	92.79 112

Fast-Effi-MVS+		92.93 106	92.64 133	93.27 88	93.81 127	93.88 121	95.90 87	90.61 103	83.98 180	92.71 70	92.81 139	96.22 105	90.67 83	94.90 110	93.92 114	95.92 117	92.77 113

HQP-MVS		92.87 108	92.49 134	93.31 86	95.75 72	95.01 94	95.64 97	91.06 96	88.54 142	91.62 101	88.16 185	96.25 103	89.47 101	92.26 153	91.81 145	96.34 100	95.40 59

FMVSNet1		92.86 109	95.26 61	90.06 146	92.40 170	95.16 87	94.37 129	92.22 63	93.18 57	82.16 198	96.76 64	97.48 73	81.85 173	95.32 98	94.98 92	97.34 68	93.93 89

CLD-MVS		92.81 110	94.32 86	91.05 130	95.39 78	95.31 83	95.82 90	81.44 213	89.40 135	91.94 90	95.86 81	97.36 74	85.83 140	95.35 97	94.59 105	95.85 119	92.34 122
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
IS_MVSNet		92.76 111	93.25 123	92.19 107	94.91 92	95.56 76	95.86 89	92.12 70	88.10 146	82.71 193	93.15 135	88.30 188	88.86 105	97.29 48	96.95 42	98.66 10	93.38 101

FC-MVSNet-train		92.75 112	95.40 59	89.66 155	95.21 83	94.82 98	97.00 58	89.40 130	91.13 109	81.71 199	97.72 34	96.43 99	77.57 199	96.89 63	96.72 48	97.05 80	94.09 85

V42		92.67 113	93.50 115	91.71 113	91.41 186	92.96 141	95.71 94	85.00 189	89.67 133	93.22 61	97.67 38	98.01 51	91.02 77	92.65 146	92.12 140	93.86 168	91.42 141

PM-MVS		92.65 114	93.20 125	92.00 109	92.11 178	90.16 183	95.99 86	84.81 193	91.31 105	92.41 80	95.87 80	96.64 95	92.35 49	93.65 133	92.91 129	94.34 163	91.85 132

QAPM		92.57 115	93.51 114	91.47 120	92.91 156	94.82 98	93.01 164	87.51 157	91.49 97	91.21 111	92.24 144	91.70 166	88.74 108	94.54 116	94.39 109	95.41 126	95.37 62

MIMVSNet1		92.52 116	94.88 68	89.77 151	96.09 58	91.99 158	96.92 60	89.68 124	95.92 16	84.55 179	96.64 69	98.21 42	78.44 192	96.08 82	95.10 88	92.91 185	90.22 160

viewmanbaseed2359cas		92.46 117	93.85 103	90.83 133	93.07 149	93.47 136	94.55 126	87.10 166	92.76 67	88.70 149	96.72 66	98.35 33	86.85 127	92.70 144	91.22 155	94.71 152	91.76 137

tfpnnormal		92.45 118	94.77 73	89.74 152	93.95 121	93.44 138	93.25 160	88.49 143	95.27 23	83.20 191	96.51 71	96.23 104	83.17 161	95.47 95	94.52 106	96.38 97	91.97 129

PCF-MVS	87.46 14	92.44 119	91.80 142	93.19 92	94.66 99	95.80 70	96.37 80	90.19 109	87.57 152	92.23 84	89.26 173	93.97 147	89.24 102	91.32 170	90.82 164	96.46 96	93.86 90
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
casdiffmvs		92.42 120	93.99 98	90.60 138	93.25 142	93.82 123	94.28 135	88.73 139	91.53 95	84.53 181	97.74 32	98.64 23	86.60 131	93.21 139	91.20 156	96.21 108	91.76 137
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		92.41 121	93.33 121	91.34 126	93.24 143	93.43 139	94.96 112	88.94 136	92.44 75	90.07 129	96.53 70	98.31 35	86.27 134	91.34 169	90.17 170	94.57 158	91.11 147

AdaColmap		92.41 121	91.49 146	93.48 82	95.96 64	95.02 93	95.37 103	91.73 85	87.97 149	91.28 108	82.82 217	91.04 170	90.62 87	95.82 87	95.07 89	95.95 116	92.67 116

v148		92.38 123	92.78 131	91.91 110	92.86 157	92.13 155	94.84 115	87.03 167	91.47 99	93.07 67	96.92 61	98.89 14	90.10 96	92.05 156	89.69 174	93.56 172	88.27 177

pmmvs-eth3d		92.34 124	92.33 135	92.34 103	92.67 161	90.67 177	96.37 80	89.06 134	90.98 113	93.60 53	97.13 57	97.02 83	88.29 113	90.20 178	91.42 153	94.07 166	88.89 171

DELS-MVS		92.33 125	93.61 113	90.83 133	92.84 158	95.13 89	94.76 119	87.22 165	87.78 151	88.42 154	95.78 84	95.28 130	85.71 143	94.44 117	93.91 115	96.01 114	92.97 109
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
Effi-MVS+-dtu		92.32 126	91.66 144	93.09 95	95.13 86	94.73 100	94.57 125	92.14 68	81.74 192	90.33 125	88.13 186	95.91 113	89.24 102	94.23 124	93.65 120	97.12 77	93.23 103

MGCFI-Net		92.31 127	94.25 92	90.04 149	93.75 129	95.96 63	93.32 157	90.28 108	93.28 53	80.57 203	88.79 177	93.78 150	84.89 145	95.55 94	95.31 83	97.45 60	97.10 21

UGNet		92.31 127	94.70 74	89.53 157	90.99 194	95.53 77	96.19 84	92.10 72	91.35 104	85.76 166	95.31 98	95.48 122	76.84 204	95.22 103	94.79 99	95.32 128	95.19 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
viewdifsd2359ckpt13		92.24 129	93.22 124	91.10 129	93.01 153	93.63 128	94.65 123	87.69 153	90.81 118	88.80 147	95.59 90	97.98 52	87.51 122	91.98 159	90.83 163	94.94 146	91.74 139

USDC		92.17 130	92.17 139	92.18 108	92.93 155	92.22 153	93.66 148	87.41 160	93.49 49	97.99 1	94.10 122	96.68 94	86.46 132	92.04 157	89.18 180	94.61 157	87.47 180

ETV-MVS		92.12 131	90.44 155	94.08 62	96.36 50	93.63 128	96.27 83	92.00 75	78.90 211	92.13 86	85.29 207	89.85 179	90.26 95	97.07 56	96.29 63	97.46 58	92.04 127

IterMVS-LS		92.10 132	92.33 135	91.82 112	93.18 144	93.66 126	92.80 171	92.27 62	90.82 116	90.59 120	97.19 53	90.97 171	87.76 118	89.60 185	90.94 160	94.34 163	93.16 106
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
MSDG		92.09 133	92.84 130	91.22 128	92.55 163	92.97 140	93.42 154	85.43 185	90.24 123	91.83 95	94.70 111	94.59 140	88.48 111	94.91 109	93.31 125	95.59 124	89.15 166

EIA-MVS		91.95 134	90.36 157	93.81 76	96.54 45	94.65 102	95.38 102	90.40 107	78.01 216	93.72 49	86.70 201	91.95 165	89.93 98	95.67 92	94.72 104	96.89 86	90.79 153

MAR-MVS		91.86 135	91.14 151	92.71 98	94.29 109	94.24 113	94.91 113	91.82 82	81.66 193	93.32 57	84.51 210	93.42 156	86.86 126	95.16 105	94.44 107	95.05 142	94.53 79
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
viewdifsd2359ckpt11		91.80 136	94.01 96	89.22 161	92.52 164	91.95 159	93.78 144	84.14 197	93.11 58	83.97 184	97.68 36	99.12 6	86.05 137	94.17 125	90.89 161	94.88 148	91.18 145

viewmsd2359difaftdt		91.80 136	94.01 96	89.22 161	92.52 164	91.95 159	93.78 144	84.14 197	93.11 58	83.97 184	97.68 36	99.12 6	86.05 137	94.16 126	90.89 161	94.88 148	91.18 145

EU-MVSNet		91.63 138	92.73 132	90.35 142	88.36 214	87.89 194	96.53 73	81.51 212	92.45 74	91.82 96	96.44 74	97.05 82	93.26 32	94.10 127	88.94 185	90.61 192	92.24 123

FC-MVSNet-test		91.49 139	94.43 80	88.07 178	94.97 89	90.53 180	95.42 101	91.18 93	93.24 55	72.94 223	98.37 15	93.86 149	78.78 186	97.82 32	96.13 67	95.13 137	91.05 148

FA-MVS(training)		91.38 140	91.18 150	91.62 116	93.49 136	92.38 150	95.03 109	90.81 99	87.20 156	91.46 104	93.00 137	89.47 182	84.19 152	93.20 141	92.08 141	94.74 151	90.90 152

FE-MVSNET		91.21 141	92.90 129	89.24 160	90.93 195	91.69 163	93.46 152	87.85 150	92.35 79	85.06 174	94.84 110	96.63 96	82.80 163	92.98 143	93.22 126	95.36 127	88.58 173

OpenMVS	89.22 12	91.09 142	91.42 147	90.71 136	92.79 160	93.61 131	92.74 173	85.47 184	86.10 165	90.73 115	85.71 206	93.07 161	86.69 130	94.07 128	93.34 124	95.86 118	94.02 87

diffmvs_AUTHOR		91.06 143	93.06 126	88.71 170	91.67 183	91.66 164	92.77 172	85.36 187	91.29 107	85.38 170	97.45 47	98.26 38	83.74 157	91.81 162	89.70 173	93.37 177	91.27 143

FPMVS		90.81 144	91.60 145	89.88 150	92.52 164	88.18 190	93.31 158	83.62 201	91.59 94	88.45 153	88.96 176	89.73 181	86.96 124	96.42 77	95.69 77	94.43 161	90.65 154

DI_MVS_pp		90.68 145	90.40 156	91.00 131	92.43 169	92.61 147	94.17 140	88.98 135	88.32 144	88.76 148	93.67 128	87.58 190	86.44 133	89.74 183	90.33 167	95.24 132	90.56 157

Vis-MVSNet (Re-imp)		90.68 145	92.18 138	88.92 165	94.63 100	92.75 144	92.91 167	91.20 92	89.21 137	75.01 219	93.96 126	89.07 186	82.72 166	95.88 86	95.30 84	97.08 79	89.08 168

DPM-MVS		90.67 147	89.86 161	91.63 115	95.29 81	94.16 114	94.52 127	89.63 125	89.59 134	89.67 134	81.95 219	88.64 187	85.75 142	90.46 175	90.43 166	94.91 147	93.77 92

diffmvs		90.44 148	92.23 137	88.35 174	91.36 188	91.38 168	92.45 179	84.84 192	89.88 131	85.09 173	96.69 68	97.71 64	83.33 160	90.01 182	88.96 184	93.03 183	91.00 149
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
FMVSNet2		90.28 149	92.04 141	88.23 176	91.22 190	94.05 115	92.88 168	90.69 101	86.53 162	79.89 207	94.38 118	92.73 162	78.54 189	91.64 166	92.26 136	96.17 109	92.67 116

IterMVS-SCA-FT		90.24 150	89.37 167	91.26 127	92.50 167	92.11 156	91.69 190	87.48 158	87.05 158	91.82 96	95.76 85	87.25 191	91.36 65	89.02 190	85.53 201	92.68 186	88.90 170

MVS_Test		90.19 151	90.58 152	89.74 152	92.12 177	91.74 162	92.51 176	88.54 142	82.80 186	87.50 158	94.62 112	95.02 137	83.97 154	88.69 193	89.32 178	93.79 169	91.85 132

EPNet		90.17 152	89.07 169	91.45 121	97.25 19	90.62 179	94.84 115	93.54 44	80.96 195	91.85 94	86.98 197	85.88 196	77.79 196	92.30 152	92.58 132	93.41 174	94.20 84
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
viewmambaseed2359dif		90.16 153	91.38 148	88.72 168	91.64 184	90.75 175	92.73 174	85.32 188	87.92 150	84.90 175	95.63 86	97.49 71	84.05 153	90.27 177	87.28 190	93.71 171	90.35 159

PVSNet_BlendedMVS		90.09 154	90.12 159	90.05 147	92.40 170	92.74 145	91.74 186	85.89 180	80.54 198	90.30 126	88.54 181	95.51 120	84.69 148	92.64 147	90.25 168	95.28 130	90.61 155

PVSNet_Blended		90.09 154	90.12 159	90.05 147	92.40 170	92.74 145	91.74 186	85.89 180	80.54 198	90.30 126	88.54 181	95.51 120	84.69 148	92.64 147	90.25 168	95.28 130	90.61 155

pmmvs4		89.95 156	89.32 168	90.69 137	91.60 185	89.17 187	94.37 129	87.63 154	88.07 148	91.02 113	94.50 115	90.50 174	86.13 136	86.33 207	89.40 177	93.39 175	87.29 183

MDA-MVSNet-bldmvs		89.75 157	91.67 143	87.50 183	74.25 233	90.88 173	94.68 121	85.89 180	91.64 92	91.03 112	95.86 81	94.35 144	89.10 104	96.87 65	86.37 197	90.04 193	85.72 188

WB-MVS		89.70 158	94.13 94	84.54 203	88.16 216	92.57 148	88.90 210	88.32 144	96.67 11	73.61 222	98.29 18	98.80 18	80.60 179	95.73 90	92.18 137	87.66 205	84.64 191

tttt0517		89.64 159	88.05 180	91.49 119	93.52 134	91.65 165	93.67 147	87.53 155	82.77 187	89.39 140	90.37 161	70.05 225	88.21 114	93.71 132	93.79 116	96.63 91	94.04 86

PatchMatch-RL		89.59 160	88.80 173	90.51 139	92.20 176	88.00 193	91.72 188	86.64 172	84.75 175	88.25 155	87.10 196	90.66 173	89.85 100	93.23 138	92.28 135	94.41 162	85.60 189

Fast-Effi-MVS+-dtu		89.57 161	88.42 177	90.92 132	93.35 139	91.57 166	93.01 164	95.71 9	78.94 210	87.65 157	84.68 209	93.14 160	82.00 171	90.84 173	91.01 159	93.78 170	88.77 172

thisisatest0530		89.54 162	87.99 182	91.35 125	93.17 145	91.31 169	93.45 153	87.53 155	82.96 185	89.17 142	90.45 158	70.32 224	88.21 114	93.37 136	93.79 116	96.54 93	93.71 94

test2506		89.51 163	87.77 185	91.55 117	94.76 95	95.23 85	94.26 136	92.80 52	92.49 71	83.31 190	89.97 166	50.93 239	86.84 128	97.62 41	96.72 48	97.32 69	91.42 141

GBi-Net		89.35 164	90.58 152	87.91 179	91.22 190	94.05 115	92.88 168	90.05 112	79.40 202	78.60 210	90.58 154	87.05 192	78.54 189	95.32 98	94.98 92	96.17 109	92.67 116

test1		89.35 164	90.58 152	87.91 179	91.22 190	94.05 115	92.88 168	90.05 112	79.40 202	78.60 210	90.58 154	87.05 192	78.54 189	95.32 98	94.98 92	96.17 109	92.67 116

thres600view7		89.14 166	88.83 171	89.51 158	93.71 132	93.55 132	93.93 143	88.02 148	87.30 154	82.40 194	81.18 220	80.63 211	82.69 167	94.27 121	95.90 70	96.27 104	88.94 169

CVMVSNet		88.97 167	89.73 163	88.10 177	87.33 221	85.22 204	94.68 121	78.68 214	88.94 139	86.98 163	95.55 92	85.71 197	89.87 99	91.19 171	89.69 174	91.05 190	91.78 135

CANet_DTU		88.95 168	89.51 166	88.29 175	93.12 148	91.22 171	93.61 149	83.47 204	80.07 201	90.71 119	89.19 174	93.68 154	76.27 208	91.44 168	91.17 158	92.59 187	89.83 162

GA-MVS		88.76 169	88.04 181	89.59 156	92.32 173	91.46 167	92.28 181	86.62 173	83.82 182	89.84 130	92.51 143	81.94 205	83.53 159	89.41 187	89.27 179	92.95 184	87.90 178

pmmvs5		88.63 170	89.70 164	87.39 184	89.24 207	90.64 178	91.87 185	82.13 208	83.34 183	87.86 156	94.58 113	96.15 108	79.87 183	87.33 202	89.07 183	93.39 175	86.76 184

thres400		88.54 171	88.15 179	88.98 163	93.17 145	92.84 142	93.56 150	86.93 169	86.45 163	82.37 195	79.96 222	81.46 208	81.83 174	93.21 139	94.76 101	96.04 113	88.39 175

CDS-MVSNet		88.41 172	89.79 162	86.79 189	94.55 105	90.82 174	92.50 177	89.85 119	83.26 184	80.52 204	91.05 150	89.93 178	69.11 219	93.17 142	92.71 131	94.21 165	87.63 179
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
gg-mvs-nofinetune		88.32 173	88.81 172	87.75 181	93.07 149	89.37 186	89.06 209	95.94 8	95.29 21	87.15 159	97.38 48	76.38 214	68.05 222	91.04 172	89.10 182	93.24 179	83.10 198

IterMVS		88.32 173	88.25 178	88.41 173	90.83 197	91.24 170	93.07 163	81.69 210	86.77 160	88.55 150	95.61 87	86.91 195	87.01 123	87.38 201	83.77 203	89.29 195	86.06 187
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
thres200		88.29 175	87.88 183	88.76 167	92.50 167	93.55 132	92.47 178	88.02 148	84.80 173	81.44 200	79.28 224	82.20 204	81.83 174	94.27 121	93.67 119	96.27 104	87.40 181

IB-MVS	86.01 17	88.24 176	87.63 186	88.94 164	92.03 179	91.77 161	92.40 180	85.58 183	78.24 213	84.85 176	71.99 229	93.45 155	83.96 155	93.48 135	92.33 134	94.84 150	92.15 124
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
MDTV_nov1_ep13_2view		88.22 177	87.85 184	88.65 171	91.40 187	86.75 198	94.07 141	84.97 190	88.86 141	93.20 62	96.11 79	96.21 107	83.70 158	87.29 203	80.29 210	84.56 214	79.46 211

test20.03		88.20 178	91.26 149	84.63 201	96.64 42	89.39 185	90.73 197	89.97 117	91.07 111	72.02 225	94.98 106	95.45 123	69.35 218	92.70 144	91.19 157	89.06 197	84.02 192

HyFIR lowres test		88.19 179	86.56 193	90.09 145	91.24 189	92.17 154	94.30 134	88.79 138	84.06 177	85.45 169	89.52 171	85.64 198	88.64 109	85.40 210	87.28 190	92.14 189	81.87 201

ET-MVSNet_ETH3D		88.06 180	85.75 198	90.74 135	92.82 159	90.68 176	93.77 146	88.59 140	81.22 194	89.78 132	89.15 175	66.79 232	84.29 151	91.72 164	91.34 154	95.22 133	89.36 165

tfpn200view9		87.94 181	87.51 188	88.44 172	92.28 174	93.63 128	93.35 156	88.11 146	80.90 196	80.89 201	78.25 225	82.25 202	79.65 185	94.27 121	94.76 101	96.36 98	88.48 174

FMVSNet3		87.90 182	88.63 175	87.04 185	89.78 205	93.46 137	91.62 191	90.05 112	79.40 202	78.60 210	90.58 154	87.05 192	77.07 203	88.03 198	89.86 172	95.12 138	92.04 127

MS-PatchMatch		87.72 183	88.62 176	86.66 190	90.81 198	88.18 190	90.92 194	82.25 207	85.86 166	80.40 205	90.14 165	89.29 184	84.93 144	89.39 188	89.12 181	90.67 191	88.34 176

Anonymous20231206		87.45 184	89.66 165	84.87 198	94.00 117	87.73 196	91.36 192	86.41 177	88.89 140	75.03 218	92.59 142	96.82 89	72.48 216	89.72 184	88.06 187	89.93 194	83.81 194

EPNet_dtu		87.40 185	86.27 194	88.72 168	95.68 74	83.37 210	92.09 183	90.08 111	78.11 215	91.29 107	86.33 202	89.74 180	75.39 211	89.07 189	87.89 188	87.81 202	89.38 164
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
baseline1		86.96 186	87.58 187	86.24 192	93.07 149	90.44 181	89.24 208	86.85 171	85.14 172	77.26 216	90.45 158	76.09 216	75.79 209	91.80 163	91.81 145	95.20 134	87.35 182

baseline		86.71 187	88.89 170	84.16 204	87.85 217	85.23 203	89.82 202	77.69 217	84.03 179	84.75 177	94.91 108	94.59 140	77.19 202	86.57 206	86.51 196	87.66 205	90.36 158

CHOSEN 1792x2688		86.64 188	86.62 191	86.65 191	90.33 201	87.86 195	93.19 161	83.30 205	83.95 181	82.32 196	87.93 188	89.34 183	86.92 125	85.64 209	84.95 202	83.85 218	86.68 185

dmvs_re		86.51 189	86.14 196	86.95 187	93.07 149	86.11 200	92.01 184	86.04 179	72.70 226	79.10 208	75.37 228	89.99 176	78.10 195	94.56 115	93.01 127	93.35 178	91.26 144

testgi		86.49 190	90.31 158	82.03 208	95.63 75	88.18 190	93.47 151	84.89 191	93.23 56	69.54 229	87.16 195	97.96 54	60.66 226	91.90 161	89.90 171	87.99 200	83.84 193

thres100view900		86.46 191	86.00 197	86.99 186	92.28 174	91.03 172	91.09 193	84.49 195	80.90 196	80.89 201	78.25 225	82.25 202	77.57 199	90.17 179	92.84 130	95.63 123	86.57 186

gm-plane-assit		86.15 192	82.51 206	90.40 141	95.81 70	92.29 151	97.99 34	84.66 194	92.15 84	93.15 64	97.84 30	44.65 240	78.60 188	88.02 199	85.95 198	92.20 188	76.69 219

CMPMVS	66.55 18	85.55 193	87.46 189	83.32 205	84.99 223	81.97 215	79.19 230	75.93 219	79.32 205	88.82 145	85.09 208	91.07 169	82.12 170	92.56 149	89.63 176	88.84 198	92.56 120
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
CR-MVSNet		85.32 194	81.58 208	89.69 154	90.36 200	84.79 206	86.72 221	92.22 63	75.38 221	90.73 115	90.41 160	67.88 229	84.86 146	83.76 213	85.74 199	93.24 179	83.14 196

baseline2		84.95 195	82.68 205	87.59 182	92.64 162	88.41 189	90.09 199	84.25 196	75.88 219	85.23 171	82.49 218	71.15 222	80.14 182	88.21 197	87.21 194	93.21 182	85.39 190

pmnet_mix02		84.85 196	86.58 192	82.83 206	90.19 202	81.10 218	88.52 213	78.58 215	91.50 96	80.32 206	96.48 72	95.86 114	75.42 210	85.17 211	76.44 219	83.91 217	79.51 210

MVSTER		84.79 197	83.79 201	85.96 194	89.14 208	89.80 184	89.39 206	82.99 206	74.16 225	82.78 192	85.97 204	66.81 231	76.84 204	90.77 174	88.83 186	94.66 154	90.19 161

MIMVSNet		84.76 198	86.75 190	82.44 207	91.71 182	85.95 201	89.74 204	89.49 127	85.28 170	69.69 228	87.93 188	90.88 172	64.85 224	88.26 196	87.74 189	89.18 196	81.24 202

SCA		84.69 199	81.10 209	88.87 166	89.02 209	90.31 182	92.21 182	92.09 73	82.72 188	89.68 133	86.83 199	73.08 218	85.80 141	80.50 221	77.51 216	84.45 216	76.80 218

new-patchmatchnet		84.45 200	88.75 174	79.43 214	93.28 141	81.87 216	81.68 227	83.48 203	94.47 30	71.53 226	98.33 16	97.88 58	58.61 229	90.35 176	77.33 217	87.99 200	81.05 204

PatchT		83.44 201	81.10 209	86.18 193	77.92 231	82.58 214	89.87 201	87.39 161	75.88 219	90.73 115	89.86 167	66.71 233	84.86 146	83.76 213	85.74 199	86.33 211	83.14 196

RPMNet		83.42 202	78.40 218	89.28 159	89.79 204	84.79 206	90.64 198	92.11 71	75.38 221	87.10 161	79.80 223	61.99 238	82.79 165	81.88 219	82.07 207	93.23 181	82.87 199

TAMVS		82.96 203	86.15 195	79.24 217	90.57 199	83.12 213	87.29 217	75.12 221	84.06 177	65.81 230	92.22 145	88.27 189	69.11 219	88.72 191	87.26 193	87.56 207	79.38 212

PatchmatchNet		82.44 204	78.69 217	86.83 188	89.81 203	81.55 217	90.78 196	87.27 164	82.39 190	88.85 144	88.31 184	70.96 223	81.90 172	78.58 225	74.33 225	82.35 222	74.69 222
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
MDTV_nov1_ep13		82.33 205	79.66 212	85.45 196	88.83 211	83.88 208	90.09 199	81.98 209	79.07 209	88.82 145	88.70 178	73.77 217	78.41 193	80.29 223	76.08 220	84.56 214	75.83 220

CostFormer		82.15 206	79.54 213	85.20 197	88.92 210	85.70 202	90.87 195	86.26 178	79.19 208	83.87 187	87.89 190	69.20 227	76.62 206	77.50 228	75.28 222	84.69 213	82.02 200

PMMVS		81.93 207	83.48 203	80.12 213	72.35 234	75.05 227	88.54 212	64.01 226	77.02 218	82.22 197	87.51 192	91.12 168	79.70 184	86.59 204	86.64 195	93.88 167	80.41 205

pmmvs3		81.69 208	83.83 200	79.19 218	78.33 230	78.57 221	89.53 205	58.71 229	78.88 212	84.34 182	88.36 183	91.96 164	77.69 198	87.48 200	82.42 206	86.54 210	79.18 213

tpm		81.58 209	78.84 215	84.79 200	91.11 193	79.50 219	89.79 203	83.75 199	79.30 206	92.05 88	90.98 151	64.78 235	74.54 212	80.50 221	76.67 218	77.49 227	80.15 208

test0.0.03 1		81.51 210	83.30 204	79.42 215	93.99 118	86.50 199	85.93 225	87.32 162	78.16 214	61.62 231	80.78 221	81.78 206	59.87 227	88.40 195	87.27 192	87.78 204	80.19 207

dps		81.42 211	77.88 223	85.56 195	87.67 219	85.17 205	88.37 215	87.46 159	74.37 224	84.55 179	86.80 200	62.18 237	80.20 181	81.13 220	77.52 215	85.10 212	77.98 216

test-LLR		80.62 212	77.20 226	84.62 202	93.99 118	75.11 225	87.04 218	87.32 162	70.11 229	78.59 213	83.17 215	71.60 220	73.88 214	82.32 217	79.20 212	86.91 208	78.87 214

tpm cat1		80.03 213	75.93 229	84.81 199	89.31 206	83.26 212	88.86 211	86.55 176	79.24 207	86.10 165	84.22 211	63.62 236	77.37 201	73.43 229	70.88 228	80.67 223	76.87 217

N_pmnet		79.33 214	84.22 199	73.62 224	91.72 181	73.72 228	86.11 223	76.36 218	92.38 76	53.38 232	95.54 94	95.62 118	59.14 228	84.23 212	74.84 224	75.03 230	73.25 226

EPMVS		79.26 215	78.20 221	80.49 211	87.04 222	78.86 220	86.08 224	83.51 202	82.63 189	73.94 221	89.59 169	68.67 228	72.03 217	78.17 226	75.08 223	80.37 224	74.37 223

CHOSEN 280x420		79.24 216	78.26 220	80.38 212	79.60 229	68.80 233	89.32 207	75.38 220	77.25 217	78.02 215	75.57 227	76.17 215	81.19 177	88.61 194	81.39 208	78.79 225	80.03 209

ADS-MVSNet		79.11 217	79.38 214	78.80 220	81.90 227	75.59 224	84.36 226	83.69 200	87.31 153	76.76 217	87.58 191	76.90 213	68.55 221	78.70 224	75.56 221	77.53 226	74.07 224

FMVSNet5		79.08 218	78.83 216	79.38 216	87.52 220	86.78 197	87.64 216	78.15 216	69.54 231	70.64 227	65.97 232	65.44 234	63.87 225	90.17 179	90.46 165	88.48 199	83.45 195

tpmrst		78.81 219	76.18 228	81.87 209	88.56 212	77.45 222	86.74 220	81.52 211	80.08 200	83.48 189	90.84 153	66.88 230	74.54 212	73.04 230	71.02 227	76.38 228	73.95 225

test-mter		78.71 220	78.35 219	79.12 219	84.03 224	76.58 223	88.51 214	59.06 228	71.06 227	78.87 209	83.73 214	71.83 219	76.44 207	83.41 216	80.61 209	87.79 203	81.24 202

MVS-HIRNet		78.28 221	75.28 230	81.79 210	80.33 228	69.38 232	76.83 231	86.59 174	70.76 228	86.66 164	89.57 170	81.04 209	77.74 197	77.81 227	71.65 226	82.62 220	66.73 230

E-PMN		77.81 222	77.88 223	77.73 223	88.26 215	70.48 231	80.19 229	71.20 223	86.66 161	72.89 224	88.09 187	81.74 207	78.75 187	90.02 181	68.30 229	75.10 229	59.85 231

EMVS		77.65 223	77.49 225	77.83 221	87.75 218	71.02 230	81.13 228	70.54 224	86.38 164	74.52 220	89.38 172	80.19 212	78.22 194	89.48 186	67.13 230	74.83 231	58.84 232

TESTMET0.1,1		77.47 224	77.20 226	77.78 222	81.94 226	75.11 225	87.04 218	58.33 230	70.11 229	78.59 213	83.17 215	71.60 220	73.88 214	82.32 217	79.20 212	86.91 208	78.87 214

new_pmnet		76.65 225	83.52 202	68.63 225	82.60 225	72.08 229	76.76 232	64.17 225	84.41 176	49.73 234	91.77 148	91.53 167	56.16 230	86.59 204	83.26 205	82.37 221	75.02 221

MVE	60.41 19	73.21 226	80.84 211	64.30 226	56.34 235	57.24 235	75.28 234	72.76 222	87.14 157	41.39 236	86.31 203	85.30 199	80.66 178	86.17 208	83.36 204	59.35 233	80.38 206
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
PMMVS2		69.86 227	82.14 207	55.52 227	75.19 232	63.08 234	75.52 233	60.97 227	88.50 143	25.11 238	91.77 148	96.44 98	25.43 232	88.70 192	79.34 211	70.93 232	67.17 229

GG-mvs-BLEND		54.28 228	77.89 222	26.72 230	0.37 240	83.31 211	70.04 235	0.39 236	74.71 223	5.36 239	68.78 230	83.06 201	0.62 236	83.73 215	78.99 214	83.55 219	72.68 228

test_method		43.16 229	51.13 231	33.85 228	7.35 237	12.38 238	51.70 237	11.91 232	62.51 233	47.64 235	62.49 233	80.78 210	28.84 231	59.55 233	34.48 232	55.68 234	45.72 233

testmvs		2.38 230	3.35 232	1.26 232	0.83 238	0.96 240	1.53 240	0.83 234	3.59 235	1.63 241	6.03 235	2.93 242	1.55 235	3.49 234	2.51 234	1.21 238	3.92 235

test123		2.16 231	2.82 233	1.41 231	0.62 239	1.18 239	1.53 240	0.82 235	2.78 236	2.27 240	4.18 236	1.98 243	1.64 234	2.58 235	3.01 233	1.56 237	4.00 234

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					94.35 108	93.52 135	92.94 166			89.43 139	84.20 212	90.07 175	80.21 180			94.56 159	93.77 92
Ray Leroy Khuboni and Hongjun Xu: Textureless Resilient Propagation Matching in Multiple View Stereosis (TPM-MVS). SATNAC 2025
RE-MVS-def										97.21 5

9.14												93.19 158

SR-MVS					97.13 23			94.77 17				97.77 61

Anonymous202405211			94.63 76		94.51 106	94.96 96	93.94 142	91.35 90	90.82 116		95.60 89	95.85 115	81.74 176	96.47 75	95.84 73	97.39 65	92.85 110

our_test_3					91.78 180	88.87 188	94.37 129

ambc			94.61 77		98.09 4	95.14 88	91.71 189		94.18 39	96.46 12	96.26 75	96.30 101	91.26 68	94.70 112	92.00 144	93.45 173	93.67 95

MTAPA										94.88 28		96.88 87

MTMP										95.43 18		97.25 76

Patchmatch-RL test							8.96 239

tmp_tt				28.44 229	36.05 236	15.86 237	21.29 238	6.40 233	54.52 234	51.96 233	50.37 234	38.68 241	9.55 233	61.75 232	59.66 231	45.36 236

XVS					96.86 32	97.48 18	98.73 3			93.28 58		96.82 89				98.17 35

X-MVStestdata					96.86 32	97.48 18	98.73 3			93.28 58		96.82 89				98.17 35

mPP-MVS					98.24 2							97.65 67

NP-MVS									85.48 169

Patchmtry						83.74 209	86.72 221	92.22 63		90.73 115

DeepMVS_CX						47.68 236	53.20 236	19.21 231	63.24 232	26.96 237	66.50 231	69.82 226	66.91 223	64.27 231		54.91 235	72.72 227

High-resolution multi-view benchmark