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		78.77 2	84.89 1	71.62 5	78.04 4	82.05 1	81.64 11	57.96 7	87.53 1	66.64 2	88.77 1	86.31 1	63.16 10	79.99 7	78.56 7	82.31 23	91.03 1
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
SED-MVS		79.21 1	84.74 2	72.75 1	78.66 3	81.96 2	82.94 5	58.16 4	86.82 2	67.66 1	88.29 4	86.15 3	66.42 2	80.41 4	78.65 6	82.65 17	90.92 2

MSP-MVS		77.82 5	83.46 5	71.24 9	75.26 17	80.22 8	82.95 4	57.85 8	85.90 3	64.79 6	88.54 3	83.43 7	66.24 3	78.21 18	78.56 7	80.34 49	89.39 7
Zhenlong Yuan, Cong Liu, Fei Shen, Zhaoxin Li, Jingguo luo, Tianlu Mao and Zhaoqi Wang: MSP-MVS: Multi-granularity Segmentation Prior Guided Multi-View Stereo. AAAI2025
DVP-MVS++		78.76 3	84.44 3	72.14 2	76.63 8	81.93 3	82.92 6	58.10 5	85.86 4	66.53 3	87.86 5	86.16 2	66.45 1	80.46 3	78.53 9	82.19 28	90.29 4

APDe-MVS		77.58 6	82.93 6	71.35 7	77.86 5	80.55 7	83.38 1	57.61 10	85.57 5	61.11 22	86.10 7	82.98 8	64.76 4	78.29 15	76.78 23	83.40 6	90.20 5

DeepPCF-MVS	66.49 1	74.25 21	80.97 10	66.41 33	67.75 53	78.87 14	75.61 40	54.16 35	84.86 6	58.22 34	77.94 17	81.01 17	62.52 17	78.34 13	77.38 17	80.16 52	88.40 11

SMA-MVS		77.32 7	82.51 7	71.26 8	75.43 15	80.19 9	82.22 8	58.26 3	84.83 7	64.36 9	78.19 16	83.46 6	63.61 8	81.00 1	80.28 1	83.66 4	89.62 6
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
DPE-MVS		78.11 4	83.84 4	71.42 6	77.82 6	81.32 4	82.92 6	57.81 9	84.04 8	63.19 14	88.63 2	86.00 4	64.52 5	78.71 11	77.63 16	82.26 24	90.57 3
Kehua Chen, Zhenlong Yuan, Tianlu Mao, Zhaoqi Wang: Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization. AAAI2025
xxxxxxxxxxxxxcwj		74.63 17	77.07 28	71.79 3	79.32 1	80.76 5	82.96 2	57.49 11	82.82 9	64.79 6	83.69 10	52.03 123	62.83 14	77.13 27	75.21 32	83.35 7	87.85 16

SF-MVS		77.13 8	81.70 8	71.79 3	79.32 1	80.76 5	82.96 2	57.49 11	82.82 9	64.79 6	83.69 10	84.46 5	62.83 14	77.13 27	75.21 32	83.35 7	87.85 16

HPM-MVS++		76.01 10	80.47 12	70.81 10	76.60 9	74.96 37	80.18 18	58.36 2	81.96 11	63.50 13	78.80 15	82.53 11	64.40 6	78.74 10	78.84 5	81.81 34	87.46 19

ACMMP_NAP		76.15 9	81.17 9	70.30 12	74.09 21	79.47 11	81.59 13	57.09 15	81.38 12	63.89 12	79.02 14	80.48 19	62.24 19	80.05 6	79.12 4	82.94 12	88.64 9

SD-MVS		74.43 18	78.87 18	69.26 21	74.39 20	73.70 47	79.06 26	55.24 27	81.04 13	62.71 16	80.18 13	82.61 10	61.70 23	75.43 42	73.92 45	82.44 22	85.22 33
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.		75.22 14	80.06 13	69.56 18	74.61 19	72.74 51	80.59 15	55.70 25	80.80 14	62.65 17	86.25 6	82.92 9	62.07 21	76.89 30	75.66 31	81.77 36	85.19 34
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
zzz-MVS		74.25 21	77.97 24	69.91 16	73.43 25	74.06 45	79.69 20	56.44 19	80.74 15	64.98 5	68.72 31	79.98 21	62.92 13	78.24 17	77.77 15	81.99 32	86.30 24

APD-MVS		75.80 11	80.90 11	69.86 17	75.42 16	78.48 17	81.43 14	57.44 13	80.45 16	59.32 28	85.28 8	80.82 18	63.96 7	76.89 30	76.08 28	81.58 40	88.30 12
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
OMC-MVS		65.16 59	71.35 46	57.94 75	52.95 151	68.82 65	69.00 59	38.28 169	79.89 17	55.20 43	62.76 53	68.31 59	56.14 61	71.30 65	68.70 79	76.06 101	79.67 60

CNVR-MVS		75.62 12	79.91 14	70.61 11	75.76 11	78.82 15	81.66 10	57.12 14	79.77 18	63.04 15	70.69 25	81.15 16	62.99 11	80.23 5	79.54 3	83.11 9	89.16 8

TSAR-MVS + ACMM		72.56 30	79.07 17	64.96 42	73.24 26	73.16 50	78.50 28	48.80 69	79.34 19	55.32 42	85.04 9	81.49 15	58.57 41	75.06 45	73.75 46	75.35 109	85.61 31

CSCG		74.68 16	79.22 16	69.40 19	75.69 13	80.01 10	79.12 25	52.83 43	79.34 19	63.99 11	70.49 26	82.02 12	60.35 34	77.48 25	77.22 20	84.38 1	87.97 15

HFP-MVS		74.87 15	78.86 20	70.21 13	73.99 22	77.91 19	80.36 17	56.63 17	78.41 21	64.27 10	74.54 21	77.75 29	62.96 12	78.70 12	77.82 13	83.02 10	86.91 22

NCCC		74.27 20	77.83 25	70.13 14	75.70 12	77.41 24	80.51 16	57.09 15	78.25 22	62.28 19	65.54 39	78.26 26	62.18 20	79.13 8	78.51 10	83.01 11	87.68 18

DeepC-MVS	66.32 2	73.85 24	78.10 23	68.90 24	67.92 51	79.31 12	78.16 30	59.28 1	78.24 23	61.13 21	67.36 37	76.10 34	63.40 9	79.11 9	78.41 11	83.52 5	88.16 13
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
CNLPA		62.78 68	66.31 67	58.65 69	58.47 105	68.41 68	65.98 83	41.22 146	78.02 24	56.04 38	46.65 128	59.50 92	57.50 47	69.67 82	65.27 130	72.70 141	76.67 81

ACMMPR		73.79 25	78.41 21	68.40 26	72.35 29	77.79 20	79.32 22	56.38 20	77.67 25	58.30 33	74.16 22	76.66 30	61.40 24	78.32 14	77.80 14	82.68 16	86.51 23

TSAR-MVS + COLMAP		62.65 69	69.90 55	54.19 100	46.31 188	66.73 89	65.49 89	41.36 144	76.57 26	46.31 80	76.80 18	56.68 102	53.27 82	69.50 83	66.65 108	72.40 146	76.36 88

DPM-MVS		72.80 28	75.90 31	69.19 22	75.51 14	77.68 21	81.62 12	54.83 28	75.96 27	62.06 20	63.96 47	76.58 31	58.55 42	76.66 34	76.77 24	82.60 19	83.68 43

CP-MVS		72.63 29	76.95 29	67.59 28	70.67 37	75.53 35	77.95 32	56.01 23	75.65 28	58.82 30	69.16 30	76.48 32	60.46 32	77.66 23	77.20 21	81.65 38	86.97 21

MP-MVS		74.31 19	78.87 18	68.99 23	73.49 24	78.56 16	79.25 24	56.51 18	75.33 29	60.69 24	75.30 20	79.12 24	61.81 22	77.78 22	77.93 12	82.18 30	88.06 14
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
MCST-MVS		73.67 26	77.39 26	69.33 20	76.26 10	78.19 18	78.77 27	54.54 32	75.33 29	59.99 26	67.96 33	79.23 23	62.43 18	78.00 19	75.71 30	84.02 2	87.30 20

SteuartSystems-ACMMP		75.23 13	79.60 15	70.13 14	76.81 7	78.92 13	81.74 9	57.99 6	75.30 31	59.83 27	75.69 19	78.45 25	60.48 31	80.58 2	79.77 2	83.94 3	88.52 10
Skip Steuart: Steuart Systems R&D Blog.
TAPA-MVS	54.74 10	60.85 77	66.61 64	54.12 102	47.38 184	65.33 100	65.35 90	36.51 178	75.16 32	48.82 73	54.70 86	63.51 73	53.31 81	68.36 96	64.97 136	73.37 129	74.27 102
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
DeepC-MVS_fast	65.08 3	72.00 31	76.11 30	67.21 30	68.93 47	77.46 22	76.54 36	54.35 33	74.92 33	58.64 32	65.18 41	74.04 44	62.62 16	77.92 20	77.02 22	82.16 31	86.21 25
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
train_agg		73.89 23	78.25 22	68.80 25	75.25 18	72.27 53	79.75 19	56.05 22	74.87 34	58.97 29	81.83 12	79.76 22	61.05 27	77.39 26	76.01 29	81.71 37	85.61 31

canonicalmvs		65.62 56	72.06 42	58.11 71	63.94 75	71.05 56	64.49 97	43.18 129	74.08 35	47.35 75	64.17 45	71.97 48	51.17 96	71.87 60	70.74 58	78.51 67	80.56 57

PGM-MVS		72.89 27	77.13 27	67.94 27	72.47 28	77.25 25	79.27 23	54.63 31	73.71 36	57.95 35	72.38 23	75.33 36	60.75 29	78.25 16	77.36 19	82.57 20	85.62 30

ACMMP		71.57 32	75.84 32	66.59 32	70.30 41	76.85 30	78.46 29	53.95 36	73.52 37	55.56 40	70.13 27	71.36 49	58.55 42	77.00 29	76.23 27	82.71 15	85.81 29
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
HQP-MVS		70.88 35	75.02 35	66.05 36	71.69 32	74.47 42	77.51 33	53.17 40	72.89 38	54.88 46	70.03 28	70.48 51	57.26 50	76.02 37	75.01 37	81.78 35	86.21 25

X-MVS		71.18 34	75.66 34	65.96 37	71.71 31	76.96 27	77.26 34	55.88 24	72.75 39	54.48 50	64.39 44	74.47 39	54.19 68	77.84 21	77.37 18	82.21 27	85.85 28

CPTT-MVS		68.76 43	73.01 39	63.81 49	65.42 64	73.66 48	76.39 38	52.08 45	72.61 40	50.33 66	60.73 61	72.65 47	59.43 38	73.32 54	72.12 51	79.19 62	85.99 27

NP-MVS									72.00 41

CLD-MVS		67.02 51	71.57 44	61.71 54	71.01 36	74.81 39	71.62 54	38.91 161	71.86 42	60.70 23	64.97 42	67.88 64	51.88 93	76.77 33	74.98 38	76.11 97	69.75 125
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
TSAR-MVS + GP.		69.71 36	73.92 38	64.80 44	68.27 49	70.56 58	71.90 52	50.75 53	71.38 43	57.46 37	68.68 32	75.42 35	60.10 35	73.47 53	73.99 44	80.32 50	83.97 39

abl_6				64.36 46	70.08 42	77.45 23	72.88 51	50.15 58	71.31 44	54.77 49	62.79 52	77.99 28	56.80 55			81.50 41	83.91 40

3Dnovator+	62.63 4	69.51 37	72.62 41	65.88 38	68.21 50	76.47 31	73.50 50	52.74 44	70.85 45	58.65 31	55.97 78	69.95 52	61.11 26	76.80 32	75.09 34	81.09 44	83.23 46

PHI-MVS		69.27 40	74.84 36	62.76 53	66.83 56	74.83 38	73.88 48	49.32 63	70.61 46	50.93 64	69.62 29	74.84 37	57.25 51	75.53 41	74.32 42	78.35 69	84.17 38

AdaColmap		67.89 47	68.85 59	66.77 31	73.73 23	74.30 44	75.28 41	53.58 38	70.24 47	57.59 36	51.19 104	59.19 93	60.74 30	75.33 44	73.72 47	79.69 57	77.96 72

MSLP-MVS++		68.17 45	70.72 50	65.19 40	69.41 44	70.64 57	74.99 42	45.76 80	70.20 48	60.17 25	56.42 76	73.01 45	61.14 25	72.80 56	70.54 61	79.70 55	81.42 53

CDPH-MVS		71.47 33	75.82 33	66.41 33	72.97 27	77.15 26	78.14 31	54.71 29	69.88 49	53.07 58	70.98 24	74.83 38	56.95 54	76.22 35	76.57 25	82.62 18	85.09 35

LGP-MVS_train		68.87 41	72.03 43	65.18 41	69.33 45	74.03 46	76.67 35	53.88 37	68.46 50	52.05 62	63.21 49	63.89 71	56.31 58	75.99 38	74.43 41	82.83 14	84.18 37

ACMP	61.42 5	68.72 44	71.37 45	65.64 39	69.06 46	74.45 43	75.88 39	53.30 39	68.10 51	55.74 39	61.53 60	62.29 78	56.97 53	74.70 48	74.23 43	82.88 13	84.31 36
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
PCF-MVS	59.98 8	67.32 50	71.04 48	62.97 52	64.77 66	74.49 41	74.78 44	49.54 60	67.44 52	54.39 53	58.35 70	72.81 46	55.79 64	71.54 62	69.24 72	78.57 64	83.41 44
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
3Dnovator	60.86 6	66.99 53	70.32 52	63.11 51	66.63 57	74.52 40	71.56 55	45.76 80	67.37 53	55.00 45	54.31 89	68.19 60	58.49 44	73.97 51	73.63 48	81.22 43	80.23 58

PLC	52.09 14	59.21 85	62.47 88	55.41 95	53.24 149	64.84 107	64.47 98	40.41 155	65.92 54	44.53 91	46.19 136	55.69 110	55.33 65	68.24 100	65.30 129	74.50 115	71.09 116
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
ACMM	60.30 7	67.58 49	68.82 60	66.13 35	70.59 38	72.01 55	76.54 36	54.26 34	65.64 55	54.78 48	50.35 107	61.72 82	58.74 40	75.79 40	75.03 35	81.88 33	81.17 54
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
CANet		68.77 42	73.01 39	63.83 48	68.30 48	75.19 36	73.73 49	47.90 70	63.86 56	54.84 47	67.51 35	74.36 42	57.62 46	74.22 50	73.57 49	80.56 47	82.36 48

MAR-MVS		68.04 46	70.74 49	64.90 43	71.68 33	76.33 32	74.63 45	50.48 57	63.81 57	55.52 41	54.88 84	69.90 53	57.39 49	75.42 43	74.79 39	79.71 54	80.03 59
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
QAPM		65.27 57	69.49 58	60.35 59	65.43 63	72.20 54	65.69 87	47.23 72	63.46 58	49.14 70	53.56 90	71.04 50	57.01 52	72.60 58	71.41 55	77.62 73	82.14 50

MVS_0304		69.49 38	73.96 37	64.28 47	67.92 51	76.13 33	74.90 43	47.60 71	63.29 59	54.09 54	67.44 36	76.35 33	59.53 37	75.81 39	75.03 35	81.62 39	83.70 42

RPSCF		46.41 183	54.42 155	37.06 198	25.70 221	45.14 205	45.39 194	20.81 215	62.79 60	35.10 135	44.92 150	55.60 111	43.56 132	56.12 192	52.45 200	51.80 209	63.91 170

EPNet		65.14 60	69.54 57	60.00 61	66.61 58	67.67 78	67.53 66	55.32 26	62.67 61	46.22 82	67.74 34	65.93 68	48.07 112	72.17 59	72.12 51	76.28 93	78.47 69
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
casdiffmvs		64.09 61	68.13 62	59.37 66	61.81 81	68.32 69	68.48 62	44.45 98	61.95 62	49.12 71	63.04 50	69.67 55	53.83 72	70.46 74	66.06 118	78.55 65	77.43 74

LS3D		60.20 80	61.70 90	58.45 70	64.18 71	67.77 75	67.19 68	48.84 68	61.67 63	41.27 108	45.89 140	51.81 125	54.18 69	68.78 89	66.50 113	75.03 112	69.48 131

OpenMVS	57.13 9	62.81 67	65.75 71	59.39 65	66.47 59	69.52 63	64.26 99	43.07 131	61.34 64	50.19 67	47.29 125	64.41 70	54.60 67	70.18 79	68.62 81	77.73 71	78.89 65

MVS_111021_HR		67.62 48	70.39 51	64.39 45	69.77 43	70.45 60	71.44 56	51.72 49	60.77 65	55.06 44	62.14 57	66.40 67	58.13 45	76.13 36	74.79 39	80.19 51	82.04 51

MVS_111021_LR		63.05 66	66.43 66	59.10 67	61.33 86	63.77 117	65.87 84	43.58 118	60.20 66	53.70 56	62.09 58	62.38 77	55.84 63	70.24 78	68.08 84	74.30 117	78.28 71

diffmvs		61.64 73	66.55 65	55.90 91	56.63 127	63.71 118	67.13 71	41.27 145	59.49 67	46.70 78	63.93 48	68.01 63	50.46 97	67.30 119	65.51 126	73.24 134	77.87 73

DROMVSNet		67.01 52	70.27 54	63.21 50	67.21 54	70.47 59	69.01 58	46.96 74	59.16 68	53.23 57	64.01 46	69.71 54	60.37 33	74.92 46	71.24 57	82.50 21	82.41 47

CS-MVS		65.88 54	69.71 56	61.41 55	61.76 83	68.14 70	67.65 64	44.00 108	59.14 69	52.69 59	65.19 40	68.13 61	60.90 28	74.74 47	71.58 53	81.46 42	81.04 55

CS-MVS-test		65.18 58	68.70 61	61.07 56	61.92 80	68.06 72	67.09 72	45.18 88	58.47 70	52.02 63	65.76 38	66.44 66	59.24 39	72.71 57	70.05 66	80.98 45	79.40 62

DELS-MVS		65.87 55	70.30 53	60.71 57	64.05 74	72.68 52	70.90 57	45.43 84	57.49 71	49.05 72	64.43 43	68.66 57	55.11 66	74.31 49	73.02 50	79.70 55	81.51 52
Christian Sormann, Emanuele Santellani, Mattia Rossi, Andreas Kuhn, Friedrich Fraundorfer: DELS-MVS: Deep Epipolar Line Search for Multi-View Stereo. Winter Conference on Applications of Computer Vision (WACV), 2023
MVS_Test		62.40 71	66.23 68	57.94 75	59.77 99	64.77 108	66.50 77	41.76 140	57.26 72	49.33 69	62.68 54	67.47 65	53.50 77	68.57 94	66.25 115	76.77 85	76.58 83

FA-MVS(training)		60.00 81	63.14 87	56.33 89	59.50 100	64.30 113	65.15 92	38.75 166	56.20 73	45.77 83	53.08 91	56.45 104	52.10 91	69.04 88	67.67 93	76.69 86	75.27 98

DI_MVS_plusplus_trai		61.88 72	65.17 76	58.06 72	60.05 94	65.26 102	66.03 81	44.22 100	55.75 74	46.73 77	54.64 87	68.12 62	54.13 70	69.13 86	66.66 107	77.18 80	76.61 82

GeoE		62.43 70	64.79 79	59.68 64	64.15 73	67.17 85	68.80 60	44.42 99	55.65 75	47.38 74	51.54 101	62.51 76	54.04 71	69.99 80	68.07 85	79.28 60	78.57 67

OPM-MVS		69.33 39	71.05 47	67.32 29	72.34 30	75.70 34	79.57 21	56.34 21	55.21 76	53.81 55	59.51 65	68.96 56	59.67 36	77.61 24	76.44 26	82.19 28	83.88 41
Ray L. Khuboni, Hongjun Xu: Octagram Propagation Matching for Multi-Scale View Stereopsis (OPM-MVS).
PVSNet_BlendedMVS		61.63 74	64.82 77	57.91 77	57.21 123	67.55 80	63.47 103	46.08 78	54.72 77	52.46 60	58.59 68	60.73 85	51.82 94	70.46 74	65.20 132	76.44 90	76.50 86

PVSNet_Blended		61.63 74	64.82 77	57.91 77	57.21 123	67.55 80	63.47 103	46.08 78	54.72 77	52.46 60	58.59 68	60.73 85	51.82 94	70.46 74	65.20 132	76.44 90	76.50 86

Effi-MVS+		63.28 63	65.96 70	60.17 60	64.26 70	68.06 72	68.78 61	45.71 82	54.08 79	46.64 79	55.92 79	63.13 75	55.94 62	70.38 77	71.43 54	79.68 58	78.70 66

test2506		55.82 120	59.57 118	51.46 120	60.39 92	64.55 110	58.69 125	48.87 66	53.91 80	26.99 171	48.97 113	41.72 188	37.71 160	70.96 69	69.49 69	76.08 98	67.37 145

ECVR-MVS		56.44 115	60.74 97	51.42 121	60.39 92	64.55 110	58.69 125	48.87 66	53.91 80	26.76 173	45.55 145	53.43 117	37.71 160	70.96 69	69.49 69	76.08 98	67.32 147

test_part1		63.06 65	65.27 74	60.47 58	66.24 62	70.17 61	71.86 53	45.36 86	53.75 82	49.61 68	44.85 151	65.53 69	48.93 103	71.39 63	70.65 59	80.82 46	80.59 56

CostFormer		56.57 113	59.13 123	53.60 104	57.52 113	61.12 133	66.94 74	35.95 180	53.44 83	44.68 90	55.87 80	54.44 113	48.21 109	60.37 168	58.33 175	68.27 167	70.33 123

MSDG		58.46 94	58.97 125	57.85 79	66.27 61	66.23 95	67.72 63	42.33 135	53.43 84	43.68 94	43.39 163	45.35 166	49.75 100	68.66 92	67.77 90	77.38 77	67.96 140

ETV-MVS		63.23 64	66.08 69	59.91 62	63.13 78	68.13 71	67.62 65	44.62 95	53.39 85	46.23 81	58.74 67	58.19 96	57.45 48	73.60 52	71.38 56	80.39 48	79.13 63

CANet_DTU		58.88 87	64.68 80	52.12 118	55.77 131	66.75 88	63.92 100	37.04 176	53.32 86	37.45 130	59.81 63	61.81 81	44.43 129	68.25 98	67.47 97	74.12 119	75.33 96

COLMAP_ROB	46.52 15	51.99 149	54.86 153	48.63 145	49.13 178	61.73 127	60.53 115	36.57 177	53.14 87	32.95 142	37.10 185	38.68 200	40.49 147	65.72 142	63.08 151	72.11 150	64.60 167
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
FC-MVSNet-train		58.40 95	63.15 86	52.85 113	64.29 69	61.84 126	55.98 146	46.47 76	53.06 88	34.96 137	61.95 59	56.37 107	39.49 150	68.67 91	68.36 83	75.92 103	71.81 113

EPNet_dtu		52.05 147	58.26 131	44.81 170	54.10 144	50.09 189	52.01 167	40.82 149	53.03 89	27.41 168	54.90 83	57.96 100	26.72 193	62.97 154	62.70 157	67.78 169	66.19 155
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
Fast-Effi-MVS+		60.36 78	63.35 85	56.87 85	58.70 102	65.86 97	65.08 93	37.11 175	53.00 90	45.36 87	52.12 98	56.07 109	56.27 59	71.28 66	69.42 71	78.71 63	75.69 93

DCV-MVSNet		59.49 82	64.00 83	54.23 99	61.81 81	64.33 112	61.42 109	43.77 111	52.85 91	38.94 122	55.62 81	62.15 80	43.24 137	69.39 84	67.66 94	76.22 95	75.97 90

baseline		55.19 129	60.88 95	48.55 146	49.87 175	58.10 162	58.70 124	34.75 184	52.82 92	39.48 121	60.18 62	60.86 84	45.41 124	61.05 164	60.74 167	63.10 184	72.41 111

USDC		51.11 153	53.71 158	48.08 153	44.76 193	55.99 171	53.01 166	40.90 147	52.49 93	36.14 133	44.67 152	33.66 210	43.27 136	63.23 153	61.10 164	70.39 161	64.82 165

ACMH+	53.71 12	59.26 84	60.28 103	58.06 72	64.17 72	68.46 67	67.51 67	50.93 52	52.46 94	35.83 134	40.83 177	45.12 170	52.32 88	69.88 81	69.00 77	77.59 75	76.21 89

test1111		55.24 126	59.98 111	49.71 130	59.80 98	64.10 115	56.48 140	49.34 62	52.27 95	21.56 187	44.49 153	51.96 124	35.93 171	70.59 73	69.07 75	75.13 111	67.40 143

PVSNet_Blended_VisFu		63.65 62	66.92 63	59.83 63	60.03 95	73.44 49	66.33 78	48.95 65	52.20 96	50.81 65	56.07 77	60.25 89	53.56 74	73.23 55	70.01 67	79.30 59	83.24 45

Anonymous20231211		57.71 104	60.79 96	54.13 101	61.68 84	65.81 98	60.81 114	43.70 115	51.97 97	39.67 117	34.82 192	63.59 72	43.31 135	68.55 95	66.63 109	75.59 104	74.13 104

pmmvs4		54.66 133	56.07 144	53.00 111	54.63 138	57.08 168	60.43 116	44.10 102	51.69 98	40.55 112	46.55 132	44.79 175	45.95 122	62.54 157	63.66 146	72.36 147	66.20 154

MS-PatchMatch		58.19 101	60.20 106	55.85 92	65.17 65	64.16 114	64.82 94	41.48 143	50.95 99	42.17 103	45.38 146	56.42 105	48.08 111	68.30 97	66.70 106	73.39 128	69.46 133

IS_MVSNet		57.95 102	64.26 82	50.60 124	61.62 85	65.25 104	57.18 133	45.42 85	50.79 100	26.49 175	57.81 72	60.05 90	34.51 176	71.24 67	70.20 65	78.36 68	74.44 101

tpm cat1		53.30 139	53.41 161	53.17 110	58.16 106	59.15 151	63.73 102	38.27 170	50.73 101	46.98 76	45.57 144	44.00 182	49.20 102	55.90 195	54.02 194	62.65 186	64.50 168

Anonymous202405211			60.60 99		63.44 77	66.71 92	61.00 113	47.23 72	50.62 102		36.85 187	60.63 88	43.03 138	69.17 85	67.72 92	75.41 106	72.54 110

UGNet		57.03 107	65.25 75	47.44 158	46.54 187	66.73 89	56.30 141	43.28 127	50.06 103	32.99 141	62.57 55	63.26 74	33.31 181	68.25 98	67.58 95	72.20 149	78.29 70
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
EIA-MVS		61.53 76	63.79 84	58.89 68	63.82 76	67.61 79	65.35 90	42.15 139	49.98 104	45.66 85	57.47 74	56.62 103	56.59 57	70.91 71	69.15 73	79.78 53	74.80 99

Vis-MVSNet (Re-imp)		50.37 158	57.73 138	41.80 184	57.53 112	54.35 174	45.70 192	45.24 87	49.80 105	13.43 203	58.23 71	56.42 105	20.11 204	62.96 155	63.36 149	68.76 165	58.96 189

EPP-MVSNet		59.39 83	65.45 73	52.32 117	60.96 88	67.70 77	58.42 127	44.75 93	49.71 106	27.23 170	59.03 66	62.20 79	43.34 134	70.71 72	69.13 74	79.25 61	79.63 61

Effi-MVS+-dtu		60.34 79	62.32 89	58.03 74	64.31 68	67.44 82	65.99 82	42.26 136	49.55 107	42.00 104	48.92 115	59.79 91	56.27 59	68.07 105	67.03 99	77.35 78	75.45 95

FMVSNet2		55.04 131	59.95 112	49.31 133	52.42 154	61.44 128	57.03 134	44.08 103	49.55 107	30.40 154	46.89 126	58.84 94	38.22 155	67.07 125	66.21 116	73.69 124	69.65 126

UA-Net		58.50 92	64.68 80	51.30 122	66.97 55	67.13 86	53.68 162	45.65 83	49.51 109	31.58 149	62.91 51	68.47 58	35.85 172	68.20 101	67.28 98	74.03 120	69.24 135

UniMVSNet_NR-MVSNet		56.94 110	61.14 93	52.05 119	60.02 96	65.21 105	57.44 131	52.93 42	49.37 110	24.31 182	54.62 88	50.54 131	39.04 152	68.69 90	68.84 78	78.53 66	70.72 118

Baseline_NR-MVSNet		53.50 137	57.89 134	48.37 149	54.60 139	59.25 150	56.10 142	51.84 46	49.32 111	17.92 197	45.38 146	47.68 143	36.93 167	68.11 103	65.95 120	72.84 136	69.57 129

baseline1		54.48 134	58.69 126	49.57 131	60.63 91	58.29 160	55.70 148	44.95 91	49.20 112	29.62 158	54.77 85	54.75 112	35.29 173	67.15 123	64.08 142	71.21 156	62.58 177

GBi-Net		55.20 127	60.25 104	49.31 133	52.42 154	61.44 128	57.03 134	44.04 104	49.18 113	30.47 151	48.28 117	58.19 96	38.22 155	68.05 106	66.96 100	73.69 124	69.65 126

test1		55.20 127	60.25 104	49.31 133	52.42 154	61.44 128	57.03 134	44.04 104	49.18 113	30.47 151	48.28 117	58.19 96	38.22 155	68.05 106	66.96 100	73.69 124	69.65 126

FMVSNet3		54.78 132	59.58 117	49.17 136	52.37 157	61.31 132	56.72 139	44.04 104	49.18 113	30.47 151	48.28 117	58.19 96	38.09 158	65.48 145	65.20 132	73.31 131	69.45 134

TranMVSNet+NR-MVSNet		55.87 118	60.14 108	50.88 123	59.46 101	63.82 116	57.93 129	52.98 41	48.94 116	20.52 190	52.87 93	47.33 148	36.81 168	69.12 87	69.03 76	77.56 76	69.89 124

ET-MVSNet_ETH3D		58.38 96	61.57 91	54.67 98	42.15 201	65.26 102	65.70 85	43.82 110	48.84 117	42.34 101	59.76 64	47.76 142	56.68 56	67.02 126	68.60 82	77.33 79	73.73 108

Vis-MVSNet		58.48 93	65.70 72	50.06 129	53.40 148	67.20 84	60.24 117	43.32 126	48.83 118	30.23 155	62.38 56	61.61 83	40.35 148	71.03 68	69.77 68	72.82 137	79.11 64
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
MVSTER		57.19 106	61.11 94	52.62 115	50.82 171	58.79 153	61.55 107	37.86 172	48.81 119	41.31 107	57.43 75	52.10 122	48.60 107	68.19 102	66.75 105	75.56 105	75.68 94

TDRefinement		49.31 163	52.44 169	45.67 166	30.44 214	59.42 147	59.24 121	39.78 159	48.76 120	31.20 150	35.73 189	29.90 214	42.81 139	64.24 152	62.59 159	70.55 159	66.43 150

IterMVS-LS		58.30 98	61.39 92	54.71 97	59.92 97	58.40 157	59.42 119	43.64 116	48.71 121	40.25 115	57.53 73	58.55 95	52.15 90	65.42 147	65.34 128	72.85 135	75.77 91
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
tpmrst		48.08 174	49.88 186	45.98 162	52.71 152	48.11 194	53.62 163	33.70 193	48.70 122	39.74 116	48.96 114	46.23 160	40.29 149	50.14 209	49.28 205	55.80 200	57.71 192

UniMVSNet (Re)		55.15 130	60.39 102	49.03 139	55.31 133	64.59 109	55.77 147	50.63 54	48.66 123	20.95 188	51.47 102	50.40 132	34.41 178	67.81 110	67.89 87	77.11 83	71.88 112

DU-MVS		55.41 124	59.59 115	50.54 126	54.60 139	62.97 120	57.44 131	51.80 47	48.62 124	24.31 182	51.99 99	47.00 151	39.04 152	68.11 103	67.75 91	76.03 102	70.72 118

NR-MVSNet		55.35 125	59.46 120	50.56 125	61.33 86	62.97 120	57.91 130	51.80 47	48.62 124	20.59 189	51.99 99	44.73 176	34.10 179	68.58 93	68.64 80	77.66 72	70.67 122

SCA		50.99 155	53.22 165	48.40 148	51.07 167	56.78 169	50.25 171	39.05 160	48.31 126	41.38 106	49.54 109	46.70 156	46.00 121	58.31 178	56.28 178	62.65 186	56.60 194

thisisatest0530		56.68 112	59.68 113	53.19 109	52.97 150	60.96 136	59.41 120	40.51 151	48.26 127	41.06 110	52.67 94	46.30 158	49.78 98	67.66 114	67.83 88	75.39 107	74.07 106

EPMVS		44.66 189	47.86 193	40.92 187	47.97 182	44.70 206	47.58 183	33.27 195	48.11 128	29.58 159	49.65 108	44.38 180	34.65 175	51.71 204	47.90 207	52.49 208	48.57 209

PatchmatchNet		49.92 162	51.29 176	48.32 150	51.83 161	51.86 183	53.38 165	37.63 174	47.90 129	40.83 111	48.54 116	45.30 167	45.19 126	56.86 185	53.99 196	61.08 191	54.57 197
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
tttt0517		56.53 114	59.59 115	52.95 112	52.66 153	60.99 135	59.21 122	40.51 151	47.89 130	40.40 113	52.50 97	46.04 162	49.78 98	67.75 112	67.83 88	75.15 110	74.17 103

v8		58.88 87	60.57 101	56.92 84	57.35 118	65.69 99	66.69 76	42.64 133	47.89 130	45.77 83	49.04 112	52.98 119	52.77 84	67.51 116	65.57 125	76.26 94	75.30 97

V42		56.97 109	60.14 108	53.28 107	48.16 180	62.78 123	66.30 79	37.93 171	47.44 132	42.68 99	48.19 120	52.59 121	51.90 92	67.46 117	65.94 121	72.72 139	76.55 85

baseline2		55.89 117	57.82 135	53.64 103	57.36 117	61.09 134	59.75 118	40.45 153	47.38 133	41.26 109	51.23 103	46.90 153	48.11 110	65.63 144	64.38 141	74.90 113	68.16 139

GG-mvs-BLEND		36.62 206	53.39 162	17.06 215	0.01 227	58.61 154	48.63 177	0.01 224	47.13 134	0.02 228	43.98 156	60.64 87	0.03 223	54.92 199	51.47 202	53.64 206	56.99 193

v10		59.17 86	60.60 99	57.50 80	57.95 108	66.73 89	67.09 72	44.11 101	46.85 135	45.42 86	48.18 121	51.07 127	53.63 73	67.84 109	66.59 111	76.79 84	76.92 79

tpm		48.82 169	51.27 177	45.96 163	54.10 144	47.35 196	56.05 143	30.23 202	46.70 136	43.21 96	52.54 96	47.55 146	37.28 165	54.11 200	50.50 203	54.90 203	60.12 186

CHOSEN 1792x2688		55.85 119	58.01 133	53.33 106	57.26 122	62.82 122	63.29 105	41.55 142	46.65 137	38.34 123	34.55 193	53.50 115	52.43 87	67.10 124	67.56 96	67.13 171	73.92 107

PatchMatch-RL		50.11 161	51.56 175	48.43 147	46.23 189	51.94 182	50.21 172	38.62 168	46.62 138	37.51 128	42.43 174	39.38 197	52.24 89	60.98 165	59.56 171	65.76 175	60.01 187

v2v482		58.69 90	60.12 110	57.03 83	57.16 125	66.05 96	67.17 69	43.52 120	46.33 139	45.19 88	49.46 111	51.02 128	52.51 86	67.30 119	66.03 119	76.61 87	74.62 100

PMMVS		49.20 167	54.28 157	43.28 178	34.13 209	45.70 204	48.98 176	26.09 211	46.31 140	34.92 138	55.22 82	53.47 116	47.48 115	59.43 170	59.04 173	68.05 168	60.77 182

IB-MVS	54.11 11	58.36 97	60.70 98	55.62 93	58.67 103	68.02 74	61.56 106	43.15 130	46.09 141	44.06 93	44.24 155	50.99 130	48.71 106	66.70 129	70.33 62	77.60 74	78.50 68
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
ACMH	52.42 13	58.24 99	59.56 119	56.70 87	66.34 60	69.59 62	66.71 75	49.12 64	46.08 142	28.90 162	42.67 172	41.20 189	52.60 85	71.39 63	70.28 63	76.51 89	75.72 92
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
FMVSNet1		54.08 135	58.68 127	48.71 143	50.90 170	61.35 131	56.73 138	43.94 109	45.91 143	29.32 161	42.72 171	56.26 108	37.70 162	68.05 106	66.96 100	73.69 124	69.50 130

Fast-Effi-MVS+-dtu		56.30 116	59.29 122	52.82 114	58.64 104	64.89 106	65.56 88	32.89 198	45.80 144	35.04 136	45.89 140	54.14 114	49.41 101	67.16 122	66.45 114	75.37 108	70.69 120

HyFIR lowres test		56.87 111	58.60 129	54.84 96	56.62 128	69.27 64	64.77 95	42.21 137	45.66 145	37.50 129	33.08 195	57.47 101	53.33 80	65.46 146	67.94 86	74.60 114	71.35 115

MDTV_nov1_ep13		50.32 159	52.43 170	47.86 156	49.87 175	54.70 172	58.10 128	34.29 188	45.59 146	37.71 127	47.44 124	47.42 147	41.86 142	58.07 181	55.21 187	65.34 178	58.56 190

v1144		58.88 87	60.16 107	57.39 81	58.03 107	67.26 83	67.14 70	44.46 97	45.17 147	44.33 92	47.81 122	49.92 135	53.20 83	67.77 111	66.62 110	77.15 81	76.58 83

IterMVS-SCA-FT		52.18 146	57.75 137	45.68 165	51.01 169	62.06 124	55.10 155	34.75 184	44.85 148	32.86 143	51.13 105	51.22 126	48.74 104	62.47 158	61.51 162	51.61 210	71.02 117

MIMVSNet		43.79 192	48.53 190	38.27 194	41.46 202	48.97 192	50.81 170	32.88 199	44.55 149	22.07 185	32.05 196	47.15 149	24.76 196	58.73 175	56.09 181	57.63 199	52.14 198

IterMVS		53.45 138	57.12 141	49.17 136	49.23 177	60.93 137	59.05 123	34.63 186	44.53 150	33.22 139	51.09 106	51.01 129	48.38 108	62.43 159	60.79 166	70.54 160	69.05 136
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
CR-MVSNet		50.47 156	52.61 167	47.98 154	49.03 179	52.94 178	48.27 178	38.86 163	44.41 151	39.59 118	44.34 154	44.65 178	46.63 118	58.97 173	60.31 168	65.48 176	62.66 174

RPMNet		46.41 183	48.72 189	43.72 174	47.77 183	52.94 178	46.02 191	33.92 190	44.41 151	31.82 148	36.89 186	37.42 205	37.41 163	53.88 201	54.02 194	65.37 177	61.47 180

ADS-MVSNet		40.67 199	43.38 205	37.50 197	44.36 195	39.79 213	42.09 205	32.67 200	44.34 153	28.87 163	40.76 179	40.37 194	30.22 185	48.34 213	45.87 212	46.81 214	44.21 213

v148		55.58 123	57.61 139	53.20 108	54.59 141	61.86 125	61.18 110	38.70 167	44.30 154	42.25 102	47.53 123	50.24 134	48.73 105	65.15 148	62.61 158	73.79 122	71.61 114

v1192		58.51 91	59.66 114	57.17 82	57.82 109	67.72 76	66.21 80	44.83 92	44.15 155	43.49 95	46.68 127	47.94 139	53.55 75	67.39 118	66.51 112	77.13 82	77.20 77

v144192		58.23 100	59.40 121	56.87 85	57.56 110	66.89 87	65.70 85	45.01 90	44.06 156	42.88 97	46.61 129	48.09 138	53.49 78	66.94 127	65.90 122	76.61 87	77.29 75

dps		50.42 157	51.20 178	49.51 132	55.88 130	56.07 170	53.73 160	38.89 162	43.66 157	40.36 114	45.66 142	37.63 204	45.23 125	59.05 171	56.18 179	62.94 185	60.16 185

FC-MVSNet-test		39.65 203	48.35 191	29.49 208	44.43 194	39.28 214	30.23 217	40.44 154	43.59 158	3.12 224	53.00 92	42.03 185	10.02 220	55.09 197	54.77 189	48.66 212	50.71 202

v1921920		57.89 103	59.02 124	56.58 88	57.55 111	66.66 93	64.72 96	44.70 94	43.55 159	42.73 98	46.17 137	46.93 152	53.51 76	66.78 128	65.75 124	76.29 92	77.28 76

pmmvs-eth3d		51.33 152	52.25 171	50.26 128	50.82 171	54.65 173	56.03 144	43.45 125	43.51 160	37.20 131	39.20 181	39.04 199	42.28 140	61.85 162	62.78 155	71.78 152	64.72 166

TinyColmap		47.08 180	47.56 194	46.52 161	42.35 200	53.44 177	51.77 168	40.70 150	43.44 161	31.92 147	29.78 203	23.72 220	45.04 127	61.99 161	59.54 172	67.35 170	61.03 181

PatchT		48.08 174	51.03 179	44.64 171	42.96 198	50.12 188	40.36 207	35.09 182	43.17 162	39.59 118	42.00 175	39.96 196	46.63 118	58.97 173	60.31 168	63.21 183	62.66 174

CMPMVS	37.70 17	49.24 165	52.71 166	45.19 167	45.97 190	51.23 185	47.44 184	29.31 203	43.04 163	44.69 89	34.45 194	48.35 137	43.64 131	62.59 156	59.82 170	60.08 192	69.48 131
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
GA-MVS		55.67 121	58.33 130	52.58 116	55.23 136	63.09 119	61.08 111	40.15 157	42.95 164	37.02 132	52.61 95	47.68 143	47.51 114	65.92 140	65.35 127	74.49 116	70.68 121

thres200		52.39 144	55.37 150	48.90 140	57.39 116	60.18 141	55.60 149	43.73 113	42.93 165	27.41 168	43.35 164	45.09 171	36.61 169	66.36 132	63.92 145	72.66 142	65.78 159

thres400		52.38 145	55.51 147	48.74 142	57.49 114	60.10 143	55.45 151	43.54 119	42.90 166	26.72 174	43.34 165	45.03 174	36.61 169	66.20 137	64.53 139	72.66 142	66.43 150

thres100view900		52.04 148	54.81 154	48.80 141	57.31 119	59.33 148	55.30 153	42.92 132	42.85 167	27.81 166	43.00 169	45.06 172	36.99 166	64.74 150	63.51 147	72.47 145	65.21 163

tfpn200view9		52.53 142	55.51 147	49.06 138	57.31 119	60.24 140	55.42 152	43.77 111	42.85 167	27.81 166	43.00 169	45.06 172	37.32 164	66.38 131	64.54 138	72.71 140	66.54 149

v1240		57.55 105	58.63 128	56.29 90	57.30 121	66.48 94	63.77 101	44.56 96	42.77 169	42.48 100	45.64 143	46.28 159	53.46 79	66.32 134	65.80 123	76.16 96	77.13 78

thres600view7		51.91 151	55.14 151	48.14 151	57.43 115	60.18 141	54.60 157	43.73 113	42.61 170	25.20 178	43.10 168	44.47 179	35.19 174	66.36 132	63.28 150	72.66 142	66.01 157

UniMVSNet_ETH3D		52.62 141	55.98 145	48.70 144	51.04 168	60.71 138	56.87 137	46.74 75	42.52 171	26.96 172	42.50 173	45.95 163	37.87 159	66.22 136	65.15 135	72.74 138	68.78 138

CDS-MVSNet		52.42 143	57.06 142	47.02 160	53.92 146	58.30 159	55.50 150	46.47 76	42.52 171	29.38 160	49.50 110	52.85 120	28.49 191	66.70 129	66.89 103	68.34 166	62.63 176
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
WR-MVS		48.78 170	55.06 152	41.45 185	55.50 132	60.40 139	43.77 200	49.99 59	41.92 173	8.10 216	45.24 149	45.56 164	17.47 205	61.57 163	64.60 137	73.85 121	66.14 156

thisisatest0515		53.85 136	56.84 143	50.37 127	50.25 174	58.17 161	55.99 145	39.90 158	41.88 174	38.16 125	45.91 139	45.30 167	44.58 128	66.15 138	66.89 103	73.36 130	73.57 109

PEN-MVS		49.21 166	54.32 156	43.24 179	54.33 142	59.26 149	47.04 186	51.37 51	41.67 175	9.97 211	46.22 135	41.80 187	22.97 201	60.52 166	64.03 143	73.73 123	66.75 148

CP-MVSNet		48.37 171	53.53 160	42.34 181	51.35 164	58.01 163	46.56 187	50.54 55	41.62 176	10.61 207	46.53 133	40.68 193	23.18 199	58.71 176	61.83 160	71.81 151	67.36 146

PS-CasMVS		48.18 173	53.25 164	42.27 182	51.26 165	57.94 164	46.51 188	50.52 56	41.30 177	10.56 208	45.35 148	40.34 195	23.04 200	58.66 177	61.79 161	71.74 153	67.38 144

FPMVS		38.36 205	40.41 209	35.97 200	38.92 206	39.85 212	45.50 193	25.79 212	41.13 178	18.70 194	30.10 201	24.56 218	31.86 183	49.42 211	46.80 210	55.04 201	51.03 201

DTE-MVSNet		48.03 176	53.28 163	41.91 183	54.64 137	57.50 166	44.63 199	51.66 50	41.02 179	7.97 217	46.26 134	40.90 190	20.24 203	60.45 167	62.89 154	72.33 148	63.97 169

FMVSNet5		40.96 197	45.81 198	35.29 203	34.30 208	44.55 207	47.28 185	28.84 205	40.76 180	21.62 186	29.85 202	42.44 184	24.77 195	57.53 183	55.00 188	54.93 202	50.56 203

WR-MVS_H		47.65 177	53.67 159	40.63 188	51.45 162	59.74 146	44.71 198	49.37 61	40.69 181	7.61 218	46.04 138	44.34 181	17.32 206	57.79 182	61.18 163	73.30 132	65.86 158

PM-MVS		44.55 190	48.13 192	40.37 189	32.85 213	46.82 201	46.11 190	29.28 204	40.48 182	29.99 156	39.98 180	34.39 209	41.80 143	56.08 193	53.88 198	62.19 189	65.31 161

pm-mvs1		51.02 154	55.55 146	45.73 164	54.16 143	58.52 155	50.92 169	42.56 134	40.32 183	25.67 177	43.66 160	50.34 133	30.06 186	65.85 141	63.97 144	70.99 158	66.21 153

pmnet_mix02		40.48 201	43.80 203	36.61 199	45.79 191	40.45 211	42.12 204	33.18 196	40.30 184	24.11 184	38.76 183	37.11 206	24.30 197	52.97 202	46.66 211	50.17 211	50.33 204

v7n		55.67 121	57.46 140	53.59 105	56.06 129	65.29 101	61.06 112	43.26 128	40.17 185	37.99 126	40.79 178	45.27 169	47.09 116	67.67 113	66.21 116	76.08 98	76.82 80

CVMVSNet		46.38 185	52.01 173	39.81 190	42.40 199	50.26 187	46.15 189	37.68 173	40.03 186	15.09 200	46.56 131	47.56 145	33.72 180	56.50 190	55.65 183	63.80 182	67.53 141

MDTV_nov1_ep13_2view		47.62 178	49.72 187	45.18 168	48.05 181	53.70 176	54.90 156	33.80 192	39.90 187	29.79 157	38.85 182	41.89 186	39.17 151	58.99 172	55.55 184	65.34 178	59.17 188

TransMVSNet (Re)		51.92 150	55.38 149	47.88 155	60.95 89	59.90 144	53.95 159	45.14 89	39.47 188	24.85 179	43.87 158	46.51 157	29.15 188	67.55 115	65.23 131	73.26 133	65.16 164

test-LLR		49.28 164	50.29 182	48.10 152	55.26 134	47.16 197	49.52 173	43.48 123	39.22 189	31.98 145	43.65 161	47.93 140	41.29 145	56.80 186	55.36 185	67.08 172	61.94 178

TESTMET0.1,1		46.09 186	50.29 182	41.18 186	36.91 207	47.16 197	49.52 173	20.32 216	39.22 189	31.98 145	43.65 161	47.93 140	41.29 145	56.80 186	55.36 185	67.08 172	61.94 178

TAMVS		44.02 191	49.18 188	37.99 196	47.03 186	45.97 203	45.04 195	28.47 206	39.11 191	20.23 191	43.22 167	48.52 136	28.49 191	58.15 180	57.95 177	58.71 194	51.36 200

test-mter		45.30 187	50.37 181	39.38 191	33.65 211	46.99 199	47.59 182	18.59 217	38.75 192	28.00 165	43.28 166	46.82 155	41.50 144	57.28 184	55.78 182	66.93 174	63.70 171

Anonymous20231206		42.28 194	45.89 197	38.07 195	51.96 159	48.98 191	43.66 201	38.81 165	38.74 193	14.32 202	26.74 208	40.90 190	20.94 202	56.64 189	54.67 191	58.71 194	54.59 196

EG-PatchMatch MVS		56.98 108	58.24 132	55.50 94	64.66 67	68.62 66	61.48 108	43.63 117	38.44 194	41.44 105	38.05 184	46.18 161	43.95 130	71.71 61	70.61 60	77.87 70	74.08 105

ambc			45.54 200		50.66 173	52.63 181	40.99 206		38.36 195	24.67 180	22.62 213	13.94 223	29.14 189	65.71 143	58.06 176	58.60 196	67.43 142

tfpnnormal		50.16 160	52.19 172	47.78 157	56.86 126	58.37 158	54.15 158	44.01 107	38.35 196	25.94 176	36.10 188	37.89 202	34.50 177	65.93 139	63.42 148	71.26 155	65.28 162

pmmvs5		47.07 181	51.02 180	42.46 180	45.18 192	51.47 184	48.23 180	33.09 197	38.17 197	28.62 164	46.60 130	43.48 183	30.74 184	58.28 179	58.63 174	68.92 164	60.48 183

test0.0.03 1		43.15 193	46.95 195	38.72 193	55.26 134	50.56 186	42.48 203	43.48 123	38.16 198	15.11 199	35.07 191	44.69 177	16.47 207	55.95 194	54.34 193	59.54 193	49.87 207

CHOSEN 280x420		40.80 198	45.05 201	35.84 202	32.95 212	29.57 217	44.98 196	23.71 214	37.54 199	18.42 195	31.36 199	47.07 150	46.41 120	56.71 188	54.65 192	48.55 213	58.47 191

N_pmnet		32.67 211	36.85 212	27.79 211	40.55 203	32.13 216	35.80 212	26.79 209	37.24 200	9.10 213	32.02 197	30.94 213	16.30 208	47.22 214	41.21 213	38.21 217	37.21 214

anonymousdsp		52.84 140	57.78 136	47.06 159	40.24 204	58.95 152	53.70 161	33.54 194	36.51 201	32.69 144	43.88 157	45.40 165	47.97 113	67.17 121	70.28 63	74.22 118	82.29 49

MVS-HIRNet		42.24 195	41.15 208	43.51 175	44.06 197	40.74 209	35.77 213	35.35 181	35.38 202	38.34 123	25.63 210	38.55 201	43.48 133	50.77 206	47.03 209	64.07 180	49.98 205

pmmvs6		48.35 172	51.64 174	44.51 172	51.92 160	57.94 164	49.44 175	42.17 138	34.45 203	24.62 181	28.87 206	46.90 153	29.07 190	64.60 151	63.08 151	69.83 162	65.68 160

SixPastTwentyTwo		47.55 179	50.25 184	44.41 173	47.30 185	54.31 175	47.81 181	40.36 156	33.76 204	19.93 192	43.75 159	32.77 212	42.07 141	59.82 169	60.94 165	68.98 163	66.37 152

EU-MVSNet		40.63 200	45.65 199	34.78 204	39.11 205	46.94 200	40.02 208	34.03 189	33.50 205	10.37 209	35.57 190	37.80 203	23.65 198	51.90 203	50.21 204	61.49 190	63.62 172

MDA-MVSNet-bldmvs		41.36 196	43.15 206	39.27 192	28.74 216	52.68 180	44.95 197	40.84 148	32.89 206	18.13 196	31.61 198	22.09 221	38.97 154	50.45 208	56.11 180	64.01 181	56.23 195

test20.03		40.38 202	44.20 202	35.92 201	53.73 147	49.05 190	38.54 209	43.49 122	32.55 207	9.54 212	27.88 207	39.12 198	12.24 212	56.28 191	54.69 190	57.96 198	49.83 208

MIMVSNet1		35.51 207	41.41 207	28.63 209	27.53 218	43.36 208	38.09 210	33.82 191	32.01 208	6.77 219	21.63 214	35.43 207	11.97 214	55.05 198	53.99 196	53.59 207	48.36 210

new-patchmatchnet		33.24 210	37.20 211	28.62 210	44.32 196	38.26 215	29.68 218	36.05 179	31.97 209	6.33 220	26.59 209	27.33 215	11.12 219	50.08 210	41.05 214	44.23 215	45.15 212

gg-mvs-nofinetune		49.07 168	52.56 168	45.00 169	61.99 79	59.78 145	53.55 164	41.63 141	31.62 210	12.08 205	29.56 204	53.28 118	29.57 187	66.27 135	64.49 140	71.19 157	62.92 173

testgi		38.71 204	43.64 204	32.95 205	52.30 158	48.63 193	35.59 214	35.05 183	31.58 211	9.03 215	30.29 200	40.75 192	11.19 218	55.30 196	53.47 199	54.53 205	45.48 211

Gipuma		25.87 212	26.91 215	24.66 212	28.98 215	20.17 220	20.46 219	34.62 187	29.55 212	9.10 213	4.91 223	5.31 227	15.76 209	49.37 212	49.10 206	39.03 216	29.95 216
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
new_pmnet		23.19 213	28.17 214	17.37 213	17.03 222	24.92 218	19.66 220	16.16 220	27.05 213	4.42 221	20.77 215	19.20 222	12.19 213	37.71 215	36.38 215	34.77 218	31.17 215

PMVS	27.84 18	33.81 209	35.28 213	32.09 206	34.13 209	24.81 219	32.51 216	26.48 210	26.41 214	19.37 193	23.76 211	24.02 219	25.18 194	50.78 205	47.24 208	54.89 204	49.95 206
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
pmmvs3		35.10 208	38.47 210	31.17 207	26.37 220	40.47 210	34.51 215	18.09 218	24.75 215	16.88 198	23.05 212	26.69 216	32.69 182	50.73 207	51.60 201	58.46 197	51.98 199

gm-plane-assit		44.74 188	45.95 196	43.33 177	60.88 90	46.79 202	36.97 211	32.24 201	24.15 216	11.79 206	29.26 205	32.97 211	46.64 117	65.09 149	62.95 153	71.45 154	60.42 184

test_method		12.44 218	14.66 218	9.85 219	1.30 226	3.32 226	13.00 222	3.21 221	22.42 217	10.22 210	14.13 216	25.64 217	11.43 217	19.75 218	11.61 221	19.96 221	5.79 222

LTVRE_ROB	44.17 16	47.06 182	50.15 185	43.44 176	51.39 163	58.42 156	42.90 202	43.51 121	22.27 218	14.85 201	41.94 176	34.57 208	45.43 123	62.28 160	62.77 156	62.56 188	68.83 137
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
tmp_tt				5.40 220	3.97 225	2.35 227	3.26 227	0.44 223	17.56 219	12.09 204	11.48 219	7.14 225	1.98 221	15.68 220	15.49 220	10.69 224

PMMVS2		15.84 214	19.68 216	11.35 217	15.74 223	16.95 221	13.31 221	17.64 219	16.08 220	0.36 227	13.12 217	11.47 224	1.69 222	28.82 216	27.24 217	19.38 222	24.09 218

EMVS		14.49 216	12.45 220	16.87 216	27.02 219	12.56 224	8.13 223	27.19 208	15.05 221	3.14 223	6.69 221	2.67 229	15.08 211	14.60 221	18.05 219	20.67 220	17.56 221

E-PMN		15.09 215	13.19 219	17.30 214	27.80 217	12.62 223	7.81 224	27.54 207	14.62 222	3.19 222	6.89 220	2.52 230	15.09 210	15.93 219	20.22 218	22.38 219	19.53 219

DeepMVS_CX						6.95 225	5.98 226	2.25 222	11.73 223	2.07 226	11.85 218	5.43 226	11.75 216	11.40 222		8.10 225	18.38 220

MVE	12.28 19	13.53 217	15.72 217	10.96 218	7.39 224	15.71 222	6.05 225	23.73 213	10.29 224	3.01 225	5.77 222	3.41 228	11.91 215	20.11 217	29.79 216	13.67 223	24.98 217
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
testmvs		0.01 219	0.02 221	0.00 221	0.00 228	0.00 228	0.01 229	0.00 225	0.01 225	0.00 229	0.03 225	0.00 231	0.01 224	0.01 223	0.01 222	0.00 226	0.06 224

test123		0.01 219	0.02 221	0.00 221	0.00 228	0.00 228	0.00 230	0.00 225	0.01 225	0.00 229	0.04 224	0.00 231	0.01 224	0.00 224	0.01 222	0.00 226	0.07 223

uanet_test		0.00 221	0.00 223	0.00 221	0.00 228	0.00 228	0.00 230	0.00 225	0.00 227	0.00 229	0.00 226	0.00 231	0.00 226	0.00 224	0.00 224	0.00 226	0.00 225

sosnet-low-res		0.00 221	0.00 223	0.00 221	0.00 228	0.00 228	0.00 230	0.00 225	0.00 227	0.00 229	0.00 226	0.00 231	0.00 226	0.00 224	0.00 224	0.00 226	0.00 225

sosnet		0.00 221	0.00 223	0.00 221	0.00 228	0.00 228	0.00 230	0.00 225	0.00 227	0.00 229	0.00 226	0.00 231	0.00 226	0.00 224	0.00 224	0.00 226	0.00 225

RE-MVS-def										33.01 140

9.14												81.81 13

SR-MVS					71.46 35			54.67 30				81.54 14

our_test_3					51.15 166	57.31 167	55.12 154

MTAPA										65.14 4		80.20 20

MTMP										62.63 18		78.04 27

Patchmatch-RL test							1.04 228

XVS					70.49 39	76.96 27	74.36 46			54.48 50		74.47 39				82.24 25

X-MVStestdata					70.49 39	76.96 27	74.36 46			54.48 50		74.47 39				82.24 25

mPP-MVS					71.67 34							74.36 42

Patchmtry						47.61 195	48.27 178	38.86 163		39.59 118

High-resolution multi-view benchmark