High-res multi-view results

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

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

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

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

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

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

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

SED-MVS		88.94 1	90.98 1	86.56 1	92.53 8	95.09 1	88.55 7	76.83 9	94.16 1	86.57 2	90.85 7	87.07 1	86.18 1	86.36 8	85.08 15	88.67 38	98.21 3

DVP-MVS		88.07 3	90.73 3	84.97 6	91.98 12	95.01 2	87.86 14	76.88 8	93.90 2	85.15 4	90.11 9	86.90 2	79.46 16	86.26 11	84.67 20	88.50 46	98.25 2
Zhenlong Yuan, Jinguo Luo, Fei Shen, Zhaoxin Li, Cong Liu, Tianlu Mao, Zhaoqi Wang: DVP-MVS: Synergize Depth-Edge and Visibility Prior for Multi-View Stereo. AAAI2025
DVP-MVS++		87.98 4	89.76 7	85.89 2	92.57 7	94.57 3	88.34 8	76.61 10	92.40 8	83.40 6	89.26 12	85.57 7	86.04 2	86.24 12	84.89 17	88.39 49	95.42 23

SF-MVS		87.30 8	88.71 8	85.64 5	94.57 1	94.55 4	91.01 1	79.94 1	89.15 14	79.85 10	92.37 5	83.29 13	79.75 13	83.52 29	82.72 36	88.75 36	95.37 26

TPM-MVS					94.34 2	93.91 5	89.34 4			75.49 21	82.52 22	83.34 12	83.53 4			89.62 12	90.78 100
Ray Leroy Khuboni and Hongjun Xu: Textureless Resilient Propagation Matching in Multiple View Stereosis (TPM-MVS). SATNAC 2025
MED-MVS		88.53 2	90.83 2	85.84 3	92.32 10	93.45 6	89.69 3	77.14 7	93.69 3	86.32 3	94.60 2	86.09 4	81.66 6	86.22 13	85.36 11	87.93 63	96.41 14

DPE-MVS		87.60 7	90.44 5	84.29 9	92.09 11	93.44 7	88.69 6	75.11 12	93.06 6	80.80 9	94.23 4	86.70 3	81.44 9	84.84 20	83.52 30	87.64 75	97.28 5
Kehua Chen, Zhenlong Yuan, Tianlu Mao, Zhaoqi Wang: Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization. AAAI2025
CSCG		82.90 23	84.52 26	81.02 20	91.85 13	93.43 8	87.14 16	74.01 17	81.96 36	76.14 17	70.84 41	82.49 16	69.71 90	82.32 44	85.18 14	87.26 91	95.40 25

MGCNet		83.82 19	86.88 18	80.26 23	88.48 35	93.17 9	82.93 36	67.66 49	88.28 18	74.90 22	77.08 36	80.93 24	78.09 21	85.83 16	85.88 7	89.53 17	96.96 10

ACM-MVS					93.98 4	92.96 10	89.10 5		88.78 16	69.60 39	79.43 29	82.20 19	80.91 11			88.69 37	94.58 32

MCST-MVS		85.75 11	86.99 15	84.31 8	94.07 3	92.80 11	88.15 13	79.10 2	85.66 25	70.72 33	76.50 37	80.45 26	82.17 5	88.35 2	87.49 3	91.63 2	97.65 4

DELS-MVS		79.49 34	79.84 43	79.08 30	88.26 41	92.49 12	84.12 30	70.63 30	65.27 94	69.60 39	61.29 69	66.50 64	72.75 50	88.07 4	88.03 2	89.13 28	97.22 6
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
CHOSEN 1792x2688		72.55 97	71.98 108	73.22 79	86.57 49	92.41 13	75.63 102	66.77 55	62.08 111	52.32 128	30.27 239	50.74 172	66.14 119	86.22 13	85.41 9	91.90 1	96.75 13

CNVR-MVS		85.96 10	87.58 13	84.06 10	92.58 6	92.40 14	87.62 15	77.77 6	88.44 17	75.93 19	79.49 28	81.97 21	81.65 7	87.04 7	86.58 4	88.79 34	97.18 7

CANet		80.90 30	82.93 31	78.53 32	86.83 48	92.26 15	81.19 48	66.95 53	81.60 39	69.90 36	66.93 52	74.80 35	76.79 26	84.68 21	84.77 19	89.50 19	95.50 21

aaEdge-Enhanced		87.94 5	89.84 6	85.72 4	91.74 14	92.20 16	88.32 10	77.84 4	92.47 7	85.03 5	94.60 2	85.70 6	81.31 10	83.94 27	83.57 29	90.10 7	96.41 14

QAPM		77.50 48	77.43 56	77.59 38	91.52 17	92.00 17	81.41 44	70.63 30	66.22 85	58.05 104	54.70 102	71.79 47	74.49 37	82.46 40	82.04 40	89.46 21	92.79 67

DPM-MVS		85.41 13	86.72 19	83.89 12	91.66 16	91.92 18	90.49 2	78.09 3	86.90 21	73.95 24	74.52 39	82.01 20	79.29 17	90.24 1	90.65 1	89.86 9	90.78 100

APDe-MVS		86.37 9	88.41 10	84.00 11	91.43 18	91.83 19	88.34 8	74.67 13	91.19 9	81.76 8	91.13 6	81.94 22	80.07 12	83.38 30	82.58 38	87.69 73	96.78 11
Zhaojie Zeng, Yuesong Wang, Tao Guan: Matching Ambiguity-Resilient Multi-View Stereo via Adaptive Patch Deformation. Pattern Recognition
MSP-MVS		87.87 6	90.57 4	84.73 7	89.38 30	91.60 20	88.24 12	74.15 15	93.55 4	82.28 7	94.99 1	83.21 14	85.96 3	87.67 5	84.67 20	88.32 50	98.29 1
Zhenlong Yuan, Cong Liu, Fei Shen, Zhaoxin Li, Jingguo luo, Tianlu Mao and Zhaoqi Wang: MSP-MVS: Multi-granularity Segmentation Prior Guided Multi-View Stereo. AAAI2025
PHI-MVS		79.43 36	84.06 28	74.04 71	86.15 51	91.57 21	80.85 52	68.90 41	82.22 35	51.81 131	78.10 31	74.28 36	70.39 84	84.01 26	84.00 24	86.14 121	94.24 36

MVSMamba_PlusPlus		80.76 31	82.78 32	78.41 33	81.93 65	91.55 22	81.27 47	68.39 45	83.28 30	66.70 47	69.11 45	68.52 57	81.56 8	88.17 3	86.51 6	90.62 5	92.28 76

DeepPCF-MVS	76.94 1	83.08 22	87.77 12	77.60 37	90.11 23	90.96 23	78.48 66	72.63 25	93.10 5	65.84 50	80.67 26	81.55 23	74.80 33	85.94 15	85.39 10	83.75 190	96.77 12

SMA-MVS		85.24 14	88.27 11	81.72 17	91.74 14	90.71 24	86.71 17	73.16 22	90.56 12	74.33 23	83.07 20	85.88 5	77.16 25	86.28 10	85.58 8	87.23 92	95.77 16
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
MVS_111021_HR		77.42 49	78.40 52	76.28 43	86.95 46	90.68 25	77.41 85	70.56 33	66.21 87	62.48 74	66.17 56	63.98 75	72.08 59	82.87 36	83.15 31	88.24 53	95.71 18

MAR-MVS		77.19 51	78.37 53	75.81 47	89.87 25	90.58 26	79.33 60	65.56 64	77.62 54	58.33 103	59.24 79	67.98 59	74.83 32	82.37 43	83.12 32	86.95 99	87.67 143
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
NCCC		84.16 18	85.46 24	82.64 13	92.34 9	90.57 27	86.57 18	76.51 11	86.85 22	72.91 27	77.20 35	78.69 30	79.09 19	84.64 22	84.88 18	88.44 47	95.41 24

OpenMVS	67.62 8	74.92 67	73.91 87	76.09 45	90.10 24	90.38 28	78.01 77	66.35 58	66.09 88	62.80 69	46.33 162	64.55 73	71.77 64	79.92 74	80.88 71	87.52 79	89.20 124

3Dnovator	70.49 5	78.42 41	76.77 62	80.35 22	91.43 18	90.27 29	81.84 41	70.79 29	72.10 64	71.95 28	50.02 137	67.86 61	77.46 24	82.89 35	84.24 22	88.61 41	89.99 115

HPM-MVS++		85.64 12	88.43 9	82.39 14	92.65 5	90.24 30	85.83 21	74.21 14	90.68 11	75.63 20	86.77 15	84.15 10	78.68 20	86.33 9	85.26 12	87.32 87	95.60 20

EPNet		79.28 39	82.25 33	75.83 46	88.31 40	90.14 31	79.43 59	68.07 46	81.76 38	61.26 88	77.26 34	70.08 53	70.06 88	82.43 42	82.00 42	87.82 67	92.09 84
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
ACMMP_NAP		83.54 20	86.37 21	80.25 24	89.57 29	90.10 32	85.27 25	71.66 26	87.38 19	73.08 26	84.23 19	80.16 27	75.31 29	84.85 19	83.64 26	86.57 109	94.21 38

GG-mvs-BLEND		54.54 227	77.58 55	27.67 260	0.03 281	90.09 33	77.20 89	0.02 275	66.83 84	0.05 281	59.90 74	73.33 39	0.04 277	78.40 99	79.30 95	88.65 39	95.20 28

PVSNet_BlendedMVS		76.84 53	78.47 50	74.95 57	82.37 60	89.90 34	75.45 106	65.45 65	74.99 60	70.66 34	63.07 62	58.27 120	67.60 108	84.24 24	81.70 49	88.18 54	97.10 8

PVSNet_Blended		76.84 53	78.47 50	74.95 57	82.37 60	89.90 34	75.45 106	65.45 65	74.99 60	70.66 34	63.07 62	58.27 120	67.60 108	84.24 24	81.70 49	88.18 54	97.10 8

sasdasda		77.65 45	79.59 44	75.39 48	81.52 67	89.83 36	81.32 45	60.74 140	80.05 44	66.72 44	68.43 46	65.09 67	74.72 35	78.87 89	82.73 34	87.32 87	92.16 79

canonicalmvs		77.65 45	79.59 44	75.39 48	81.52 67	89.83 36	81.32 45	60.74 140	80.05 44	66.72 44	68.43 46	65.09 67	74.72 35	78.87 89	82.73 34	87.32 87	92.16 79

SteuartSystems-ACMMP		82.51 24	85.35 25	79.20 28	90.25 21	89.39 38	84.79 26	70.95 28	82.86 32	68.32 42	86.44 16	77.19 31	73.07 46	83.63 28	83.64 26	87.82 67	94.34 35
Skip Steuart: Steuart Systems R&D Blog.
casdiffmvs_mvg		75.57 60	76.04 67	75.02 56	80.48 78	89.31 39	80.79 53	64.04 77	66.95 83	63.87 60	57.52 85	61.33 88	72.90 48	82.01 50	81.99 43	88.03 59	93.16 58
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
APD-MVS		84.83 15	87.00 14	82.30 15	89.61 28	89.21 40	86.51 19	73.64 19	90.98 10	77.99 15	89.89 10	80.04 28	79.18 18	82.00 51	81.37 57	86.88 101	95.49 22
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
hybridcas		74.86 68	74.70 79	75.04 55	79.57 83	89.12 41	78.97 61	64.02 78	65.29 93	65.36 52	54.81 101	60.39 100	73.16 44	80.41 66	80.49 83	89.18 27	92.39 74

E2		75.18 65	75.21 73	75.15 53	79.77 81	89.10 42	78.62 63	64.19 73	65.19 95	65.90 49	58.15 82	58.36 118	72.56 52	80.74 63	81.78 46	89.84 10	93.19 56

viewdifsd2359ckpt13		74.11 76	74.06 86	74.18 69	79.34 90	89.07 43	78.31 72	64.25 72	62.52 107	62.06 77	55.80 94	56.70 136	72.29 54	80.35 69	81.47 55	88.80 33	92.47 72

casdiffmvs		75.20 63	75.69 70	74.63 62	79.26 93	89.07 43	78.47 67	63.59 89	67.05 82	63.79 61	55.72 96	60.32 101	73.58 40	82.16 46	81.78 46	89.08 30	93.72 48
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
IB-MVS	64.48 11	69.02 125	68.97 134	69.09 117	81.75 66	89.01 45	64.50 192	64.91 68	56.65 142	62.59 73	47.89 146	45.23 185	51.99 195	69.18 207	81.88 45	88.77 35	92.93 61
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
viewcassd2359sk11		74.75 69	74.61 83	74.90 59	79.62 82	88.96 46	78.47 67	64.08 75	63.51 101	65.27 53	57.02 88	57.89 124	72.25 55	80.30 70	81.57 53	89.72 11	93.04 60

viewmanbaseed2359cas		74.53 70	74.69 81	74.35 65	79.37 89	88.90 47	78.96 62	64.07 76	63.67 98	62.19 76	56.95 89	58.42 117	72.04 60	80.08 71	81.92 44	89.47 20	92.91 62

MVS_Test		75.22 62	76.69 63	73.51 72	79.30 91	88.82 48	80.06 56	58.74 151	69.77 72	57.50 108	59.78 76	61.35 86	75.31 29	82.07 48	83.60 28	90.13 6	91.41 92

E3new		74.17 74	73.83 89	74.57 63	79.40 87	88.76 49	78.30 73	63.89 83	61.21 114	64.38 59	55.65 97	57.34 129	71.87 61	79.73 78	81.28 60	89.55 15	92.86 63

E3		74.17 74	73.83 89	74.57 63	79.40 87	88.76 49	78.30 73	63.89 83	61.22 113	64.40 58	55.64 98	57.35 128	71.86 62	79.73 78	81.27 61	89.55 15	92.86 63

SD-MVS		84.31 17	86.96 16	81.22 18	88.98 34	88.68 51	85.65 22	73.85 18	89.09 15	79.63 11	87.34 14	84.84 8	73.71 39	82.66 38	81.60 52	85.48 144	94.51 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
Casviewmamba		75.20 63	75.26 72	75.13 54	80.13 80	88.67 52	78.61 64	64.02 78	67.43 81	66.72 44	56.60 91	60.53 93	73.45 43	80.41 66	81.03 66	87.84 65	92.13 83

gg-mvs-nofinetune		62.34 178	66.19 159	57.86 203	76.15 131	88.61 53	71.18 150	41.24 259	25.74 260	13.16 265	22.91 257	63.97 76	54.52 187	85.06 18	85.25 13	90.92 3	91.78 88

E5new		73.48 85	72.84 98	74.23 67	79.06 95	88.52 54	78.32 70	63.99 80	58.33 129	63.34 65	54.07 112	56.89 132	71.29 71	78.99 86	80.82 74	89.35 22	92.26 77

E5		73.48 85	72.84 98	74.23 67	79.06 95	88.52 54	78.32 70	63.99 80	58.33 129	63.34 65	54.07 112	56.89 132	71.29 71	78.99 86	80.82 74	89.35 22	92.26 77

DeepC-MVS	74.46 3	80.30 33	81.05 38	79.42 26	87.42 44	88.50 56	83.23 32	73.27 21	82.78 33	71.01 32	62.86 64	69.93 54	74.80 33	84.30 23	84.20 23	86.79 104	94.77 29
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
E4		73.32 88	72.68 100	74.06 70	79.06 95	88.47 57	77.98 78	63.57 90	57.73 138	63.18 67	53.48 115	56.74 135	71.26 73	78.95 88	80.84 72	89.30 24	92.55 68

viewmacassd2359aftdt		73.00 90	72.63 101	73.44 75	78.70 104	88.45 58	78.52 65	63.49 91	57.74 137	60.15 99	52.57 121	57.01 131	70.69 80	78.85 92	81.29 59	89.10 29	92.48 70

train_agg		83.35 21	86.93 17	79.17 29	89.70 27	88.41 59	85.60 24	72.89 24	86.31 23	66.58 48	90.48 8	82.24 18	73.06 47	83.10 34	82.64 37	87.21 96	95.30 27

viewdifsd2359ckpt09		73.89 80	73.57 91	74.26 66	78.54 108	88.37 60	78.34 69	63.79 85	63.31 102	64.90 55	57.29 87	56.53 138	72.15 58	79.12 83	77.91 116	87.83 66	92.48 70

CDPH-MVS		79.39 38	82.13 34	76.19 44	89.22 33	88.34 61	84.20 29	71.00 27	79.67 48	56.97 109	77.77 32	72.24 45	68.50 104	81.33 55	82.74 33	87.23 92	92.84 65

3Dnovator+	70.16 6	77.87 44	77.29 58	78.55 31	89.25 32	88.32 62	80.09 55	67.95 47	74.89 62	71.83 29	52.05 127	70.68 51	76.27 28	82.27 45	82.04 40	85.92 126	90.77 102

E6new		72.71 95	72.05 105	73.49 73	79.01 99	88.31 63	77.06 90	62.71 109	56.63 143	62.00 78	52.31 122	55.75 145	70.93 76	78.51 96	80.72 77	89.20 25	92.14 81

E6		72.71 95	72.05 105	73.49 73	79.01 99	88.31 63	77.06 90	62.71 109	56.63 143	62.00 78	52.31 122	55.75 145	70.93 76	78.51 96	80.72 77	89.20 25	92.14 81

TSAR-MVS + GP.		82.27 26	85.98 22	77.94 35	80.72 75	88.25 65	81.12 49	67.71 48	87.10 20	73.31 25	85.23 17	83.68 11	76.64 27	80.43 65	81.47 55	88.15 56	95.66 19

MP-MVS		80.94 29	83.49 29	77.96 34	88.48 35	88.16 66	82.82 37	69.34 37	80.79 42	69.67 37	82.35 23	77.13 32	71.60 66	80.97 61	80.96 69	85.87 129	94.06 41
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
viewdifsd2359ckpt07		72.78 93	72.24 103	73.41 78	78.58 107	88.14 67	76.95 92	63.73 87	57.28 139	63.47 63	54.45 107	56.62 137	69.16 99	78.86 91	79.98 85	88.58 44	90.33 109

HyFIR lowres test		68.39 130	68.28 142	68.52 123	80.85 72	88.11 68	71.08 152	58.09 156	54.87 160	47.80 150	27.55 247	55.80 144	64.97 123	79.11 84	79.14 98	88.31 51	93.35 53

CLD-MVS		77.36 50	77.29 58	77.45 39	82.21 62	88.11 68	81.92 40	68.96 40	77.97 52	69.62 38	62.08 65	59.44 111	73.57 41	81.75 53	81.27 61	88.41 48	90.39 108
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
ET-MVSNet_ETH3D		71.38 107	74.70 79	67.51 133	51.61 249	88.06 70	77.29 86	60.95 139	63.61 99	48.36 147	66.60 54	60.67 91	79.55 14	73.56 160	80.58 80	87.30 90	89.80 117

PCF-MVS	70.85 4	75.73 59	76.55 65	74.78 61	83.67 56	88.04 71	81.47 42	70.62 32	69.24 78	57.52 107	60.59 73	69.18 56	70.65 81	77.11 114	77.65 118	84.75 171	94.01 42
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
ETV-MVS		76.25 55	80.22 41	71.63 99	78.23 110	87.95 72	72.75 130	60.27 146	77.50 55	57.73 105	71.53 40	66.60 63	73.16 44	80.99 60	81.23 63	87.63 76	95.73 17

DeepC-MVS_fast	75.41 2	81.69 27	82.10 35	81.20 19	91.04 20	87.81 73	83.42 31	74.04 16	83.77 29	71.09 31	66.88 53	72.44 41	79.48 15	85.08 17	84.97 16	88.12 57	93.78 45
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
HFP-MVS		82.48 25	84.12 27	80.56 21	90.15 22	87.55 74	84.28 28	69.67 35	85.22 26	77.95 16	84.69 18	75.94 34	75.04 31	81.85 52	81.17 64	86.30 117	92.40 73

baseline		72.89 91	74.46 85	71.07 100	75.99 133	87.50 75	74.57 112	60.49 143	70.72 68	57.60 106	60.63 72	60.97 89	70.79 79	75.27 138	76.33 131	86.94 100	89.79 118

onestephybrid01		73.58 83	74.69 81	72.29 90	76.11 132	87.32 76	76.53 98	62.91 101	68.13 80	63.40 64	58.47 80	60.61 92	68.74 103	76.69 121	78.09 111	86.05 124	93.54 50

MGCFI-Net		74.26 72	78.69 48	69.10 115	80.64 76	87.32 76	73.21 129	59.20 149	79.76 47	50.18 141	68.10 48	64.86 72	64.65 127	78.28 102	80.83 73	86.69 106	91.69 89

diffmvs		74.32 71	75.42 71	73.04 81	75.60 138	87.27 78	78.20 75	62.96 97	68.66 79	61.89 80	59.79 75	59.84 107	71.80 63	78.30 101	79.87 86	87.80 69	94.23 37
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
casdiffseed414692147		71.49 104	70.06 128	73.15 80	79.11 94	87.26 79	77.82 81	62.34 116	58.44 128	60.33 98	46.19 163	51.26 168	71.53 67	77.07 115	79.56 92	87.80 69	90.61 105

hybridnocas07		74.06 77	75.21 73	72.71 84	75.43 140	87.22 80	76.90 94	62.70 111	69.87 70	62.72 71	59.53 77	59.98 106	71.03 75	77.21 113	79.23 96	87.49 80	93.44 52

EIA-MVS		73.48 85	76.05 66	70.47 105	78.12 111	87.21 81	71.78 140	60.63 142	69.66 73	55.56 114	64.86 58	60.69 90	69.53 93	77.35 112	78.59 102	87.22 94	94.01 42

PVSNet_Blended_VisFu		71.76 103	73.54 93	69.69 110	79.01 99	87.16 82	72.05 137	61.80 123	56.46 146	59.66 100	53.88 114	62.48 78	59.08 164	81.17 57	78.90 99	86.53 111	94.74 30

diffmvs_AUTHOR		73.73 82	74.73 78	72.56 88	75.05 142	87.15 83	77.82 81	62.29 117	66.22 85	61.10 91	57.92 83	59.72 109	71.43 68	78.25 103	79.68 89	87.71 72	94.17 39

viewmamba		73.51 84	74.57 84	72.28 91	75.68 137	87.10 84	76.82 95	62.81 103	69.38 75	61.26 88	58.32 81	59.73 108	70.35 85	76.34 124	78.81 101	86.77 105	92.32 75

ACMMPR		80.62 32	82.98 30	77.87 36	88.41 37	87.05 85	83.02 33	69.18 38	83.91 28	68.35 41	82.89 21	73.64 38	72.16 57	80.78 62	81.13 65	86.10 122	91.43 90

hybrid		73.86 81	75.13 75	72.38 89	75.05 142	87.04 86	76.72 96	62.53 113	69.51 74	62.37 75	59.27 78	60.40 99	70.21 87	77.07 115	79.17 97	87.39 83	93.46 51

SPE-MVS-test		75.09 66	77.84 54	71.87 98	79.27 92	86.92 87	70.53 159	60.36 144	75.13 59	63.13 68	67.92 49	65.08 69	71.43 68	78.15 104	78.51 105	86.53 111	93.16 58

DI_MVS_pp		73.94 79	74.85 77	72.88 82	76.57 128	86.80 88	80.41 54	61.47 129	62.35 109	59.44 101	47.91 145	68.12 58	72.24 56	82.84 37	81.50 54	87.15 98	94.42 34

CS-MVS		75.84 58	78.61 49	72.61 87	79.03 98	86.74 89	74.43 120	60.27 146	74.15 63	62.78 70	66.26 55	64.25 74	72.81 49	83.36 31	81.69 51	86.32 115	93.85 44

PGM-MVS		79.42 37	81.84 36	76.60 42	88.38 39	86.69 90	82.97 35	65.75 62	80.39 43	64.94 54	81.95 25	72.11 46	71.41 70	80.45 64	80.55 81	86.18 119	90.76 103

test1111		66.72 145	67.80 145	65.45 144	77.42 121	86.63 91	69.69 163	62.98 96	55.29 154	39.47 189	40.12 192	47.11 180	55.70 182	79.96 73	80.00 84	87.47 81	85.49 164

EC-MVSNet		76.05 57	78.87 47	72.77 83	78.87 103	86.63 91	77.50 84	57.04 180	75.34 58	61.68 84	64.20 59	69.56 55	73.96 38	82.12 47	80.65 79	87.57 77	93.57 49

CANet_DTU		72.84 92	76.63 64	68.43 126	76.81 125	86.62 93	75.54 105	54.71 207	72.06 65	43.54 167	67.11 51	58.46 115	72.40 53	81.13 59	80.82 74	87.57 77	90.21 111

dtuplus		72.12 101	72.21 104	72.01 95	74.74 147	86.54 94	77.22 88	61.74 127	60.26 120	61.52 86	54.43 108	57.46 127	70.32 86	75.64 134	77.35 121	86.51 113	93.75 46

TSAR-MVS + ACMM		81.59 28	85.84 23	76.63 41	89.82 26	86.53 95	86.32 20	66.72 56	85.96 24	65.43 51	88.98 13	82.29 17	67.57 111	82.06 49	81.33 58	83.93 188	93.75 46

TSAR-MVS + MP.		84.39 16	86.58 20	81.83 16	88.09 42	86.47 96	85.63 23	73.62 20	90.13 13	79.24 12	89.67 11	82.99 15	77.72 23	81.22 56	80.92 70	86.68 107	94.66 31
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
viewmambaseed2359dif		72.54 98	72.88 97	72.13 93	74.78 146	86.45 97	77.24 87	61.65 128	62.61 106	61.83 81	55.85 92	57.51 126	70.64 82	75.71 132	77.90 117	86.65 108	94.16 40

test2506		69.26 119	70.79 121	67.48 134	78.64 105	86.40 98	72.22 135	62.75 107	58.05 133	45.24 157	50.76 132	54.93 152	58.05 170	79.82 75	79.70 87	87.96 61	85.90 159

ECVR-MVS		67.93 135	68.49 137	67.28 137	78.64 105	86.40 98	72.22 135	62.75 107	58.05 133	44.06 165	40.92 187	48.20 177	58.05 170	79.82 75	79.70 87	87.96 61	86.32 153

baseline2		71.22 109	73.01 96	69.13 114	75.76 135	86.34 100	71.23 148	62.78 105	62.62 105	52.85 127	57.32 86	54.31 155	63.27 135	79.74 77	79.31 94	88.89 32	91.43 90

XVS					82.43 58	86.27 101	75.70 100			61.07 92		72.27 42				85.67 138

X-MVStestdata					82.43 58	86.27 101	75.70 100			61.07 92		72.27 42				85.67 138

X-MVS		78.16 43	80.55 40	75.38 50	87.99 43	86.27 101	81.05 50	68.98 39	78.33 50	61.07 92	75.25 38	72.27 42	67.52 113	80.03 72	80.52 82	85.66 141	91.20 94

CostFormer		72.18 99	73.90 88	70.18 107	79.47 85	86.19 104	76.94 93	48.62 230	66.07 89	60.40 97	54.14 110	65.82 65	67.98 105	75.84 131	76.41 130	87.67 74	92.83 66

FA-MVS(training)		70.24 115	71.77 111	68.45 125	77.52 119	86.03 105	73.33 127	49.12 229	63.55 100	55.77 111	48.91 142	56.26 140	67.78 107	77.60 107	79.62 90	87.19 97	90.40 107

CP-MVS		79.44 35	81.51 37	77.02 40	86.95 46	85.96 106	82.00 39	68.44 44	81.82 37	67.39 43	77.43 33	73.68 37	71.62 65	79.56 81	79.58 91	85.73 134	92.51 69

ACMMP		77.61 47	79.59 44	75.30 51	85.87 52	85.58 107	81.42 43	67.38 52	79.38 49	62.61 72	78.53 30	65.79 66	68.80 102	78.56 95	78.50 106	85.75 131	90.80 99
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
MS-PatchMatch		70.34 114	69.00 133	71.91 97	85.20 55	85.35 108	77.84 80	61.77 124	58.01 135	55.40 115	41.26 183	58.34 119	61.69 144	81.70 54	78.29 107	89.56 14	80.02 207

Effi-MVS+		70.42 110	71.23 115	69.47 111	78.04 112	85.24 109	75.57 104	58.88 150	59.56 123	48.47 146	52.73 120	54.94 151	69.69 91	78.34 100	77.06 123	86.18 119	90.73 104

MVS_111021_LR		74.26 72	75.95 68	72.27 92	79.43 86	85.04 110	72.71 131	65.27 67	70.92 67	63.58 62	69.32 43	60.31 103	69.43 95	77.01 117	77.15 122	83.22 198	91.93 87

AdaColmap		76.23 56	73.55 92	79.35 27	89.38 30	85.00 111	79.99 57	73.04 23	76.60 56	71.17 30	55.18 100	57.99 122	77.87 22	76.82 119	76.82 125	84.67 173	86.45 150

0.4-1-1-0.2		70.06 116	70.92 120	69.06 118	67.65 190	84.98 112	74.41 122	62.76 106	63.03 103	53.95 120	51.07 131	60.32 101	67.52 113	73.73 158	74.85 148	88.04 58	88.45 136

0.3-1-1-0.015		70.01 117	70.93 118	68.93 119	67.63 192	84.94 113	74.17 123	62.69 112	62.88 104	53.78 122	51.37 130	60.47 94	67.27 115	73.70 159	74.70 150	88.00 60	88.47 135

MSLP-MVS++		78.57 40	77.33 57	80.02 25	88.39 38	84.79 114	84.62 27	66.17 60	75.96 57	78.40 13	61.59 67	71.47 48	73.54 42	78.43 98	78.88 100	88.97 31	90.18 112

0.4-1-1-0.1		69.62 118	70.57 123	68.51 124	67.55 194	84.77 115	73.54 125	62.45 115	62.23 110	53.25 126	50.57 135	60.25 104	66.36 117	73.49 162	74.34 158	87.90 64	88.30 138

Vis-MVSNet		65.53 153	69.83 129	60.52 184	70.80 172	84.59 116	66.37 188	55.47 197	48.40 184	40.62 186	57.67 84	58.43 116	45.37 226	77.49 108	76.24 133	84.47 179	85.99 158
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
baseline1		71.47 105	72.02 107	70.82 102	80.56 77	84.51 117	76.61 97	66.93 54	56.22 148	48.66 145	55.40 99	60.43 98	62.55 140	83.35 32	80.99 67	89.60 13	83.28 184

thisisatest0530		68.38 131	70.98 117	65.35 145	72.61 159	84.42 118	68.21 172	57.98 158	59.77 122	50.80 136	54.63 103	58.48 114	57.92 172	76.99 118	77.47 119	84.60 176	85.07 166

Anonymous202405211			66.35 158		78.00 113	84.41 119	74.85 110	63.18 94	51.00 173		31.37 236	53.73 159	69.67 92	76.28 125	76.84 124	83.21 200	90.85 98

OPM-MVS		72.74 94	70.93 118	74.85 60	85.30 54	84.34 120	82.82 37	69.79 34	49.96 177	55.39 116	54.09 111	60.14 105	70.04 89	80.38 68	79.43 93	85.74 133	88.20 139
Ray L. Khuboni, Hongjun Xu: Octagram Propagation Matching for Multi-Scale View Stereopsis (OPM-MVS).
HQP-MVS		78.26 42	80.91 39	75.17 52	85.67 53	84.33 121	83.01 34	69.38 36	79.88 46	55.83 110	79.85 27	64.90 71	70.81 78	82.46 40	81.78 46	86.30 117	93.18 57

IS_MVSNet		67.29 142	71.98 108	61.82 177	76.92 124	84.32 122	65.90 190	58.22 154	55.75 152	39.22 192	54.51 105	62.47 79	45.99 223	78.83 93	78.52 104	84.70 172	89.47 121

EPMVS		66.21 147	67.49 148	64.73 150	75.81 134	84.20 123	68.94 168	44.37 245	61.55 112	48.07 149	49.21 141	54.87 153	62.88 136	71.82 180	71.40 197	88.28 52	79.37 210

tttt0517		67.99 134	70.61 122	64.94 148	71.94 164	83.96 124	67.62 176	57.98 158	59.30 124	49.90 142	54.50 106	57.98 123	57.92 172	76.48 123	77.47 119	84.24 183	84.58 170

thres100view900		67.14 144	66.09 160	68.38 127	77.70 114	83.84 125	74.52 116	66.33 59	49.16 181	43.40 169	43.24 167	41.34 195	62.59 139	79.31 82	75.92 136	85.73 134	89.81 116

Anonymous20231211		68.44 129	66.37 157	70.86 101	77.58 117	83.49 126	75.15 109	61.89 121	52.54 170	58.50 102	28.89 241	56.78 134	69.29 98	74.96 142	76.61 126	82.73 204	91.36 93

EPP-MVSNet		67.58 138	71.10 116	63.48 162	75.71 136	83.35 127	66.85 182	57.83 163	53.02 168	41.15 182	55.82 93	67.89 60	56.01 181	74.40 147	72.92 179	83.33 196	90.30 110

GeoE		68.96 126	69.32 130	68.54 122	76.61 127	83.12 128	71.78 140	56.87 182	60.21 121	54.86 118	45.95 164	54.79 154	64.27 128	74.59 144	75.54 142	86.84 103	91.01 97

MVSTER		76.92 52	79.92 42	73.42 77	74.98 144	82.97 129	78.15 76	63.41 92	78.02 51	64.41 57	67.54 50	72.80 40	71.05 74	83.29 33	83.73 25	88.53 45	91.12 95

PatchmatchNet		65.43 154	67.71 146	62.78 168	73.49 154	82.83 130	66.42 187	45.40 240	60.40 119	45.27 156	49.22 140	57.60 125	60.01 156	70.61 193	71.38 198	86.08 123	81.91 200
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
viewdifsd2359ckpt11		69.15 122	68.30 139	70.14 108	73.44 156	82.79 131	72.24 133	61.20 132	54.59 163	61.70 83	53.16 116	52.89 164	67.57 111	71.81 182	72.73 182	84.66 174	90.10 113

viewmsd2359difaftdt		69.14 123	68.29 140	70.13 109	73.44 156	82.79 131	72.24 133	61.20 132	54.60 162	61.68 84	53.16 116	52.87 165	67.58 110	71.82 180	72.73 182	84.66 174	90.10 113

thres400		65.18 156	64.44 168	66.04 140	76.40 129	82.63 133	71.52 145	64.27 71	44.93 199	40.69 185	41.86 180	40.79 201	58.12 168	77.67 106	74.64 151	85.26 153	88.56 132

UGNet		67.57 139	71.69 112	62.76 169	69.88 174	82.58 134	66.43 186	58.64 152	54.71 161	51.87 130	61.74 66	62.01 83	45.46 225	74.78 143	74.99 145	84.24 183	91.02 96
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
CPTT-MVS		75.43 61	77.13 60	73.44 75	81.43 69	82.55 135	80.96 51	64.35 70	77.95 53	61.39 87	69.20 44	70.94 50	69.38 97	73.89 154	73.32 171	83.14 201	92.06 85

TSAR-MVS + COLMAP		73.09 89	76.86 61	68.71 120	74.97 145	82.49 136	74.51 117	61.83 122	83.16 31	49.31 144	82.22 24	51.62 167	68.94 101	78.76 94	75.52 143	82.67 206	84.23 174

PMMVS		70.37 113	75.06 76	64.90 149	71.46 165	81.88 137	64.10 194	55.64 192	71.31 66	46.69 151	70.69 42	58.56 112	69.53 93	79.03 85	75.63 139	81.96 216	88.32 137

MDTV_nov1_ep13		65.21 155	67.28 149	62.79 167	70.91 170	81.72 138	69.28 167	49.50 227	58.08 132	43.94 166	50.50 136	56.02 142	58.86 165	70.72 192	73.37 169	84.24 183	80.52 206

thres600view7		63.77 166	63.14 178	64.51 152	75.49 139	81.61 139	69.59 164	62.95 98	43.96 202	38.90 194	41.09 184	40.24 210	55.25 185	76.24 126	71.54 192	84.89 163	87.30 144

thres200		65.58 151	64.74 166	66.56 139	77.52 119	81.61 139	73.44 126	62.95 98	46.23 193	42.45 176	42.76 169	41.18 197	58.12 168	76.24 126	75.59 140	84.89 163	89.58 119

tfpn200view9		65.90 150	64.96 164	67.00 138	77.70 114	81.58 141	71.71 143	62.94 100	49.16 181	43.40 169	43.24 167	41.34 195	61.42 146	76.24 126	74.63 152	84.84 165	88.52 133

GA-MVS		64.55 160	65.76 163	63.12 164	69.68 175	81.56 142	69.59 164	58.16 155	45.23 198	35.58 218	47.01 157	41.82 192	59.41 160	79.62 80	78.54 103	86.32 115	86.56 149

tpm cat1		67.47 140	67.05 152	67.98 129	76.63 126	81.51 143	74.49 118	47.65 235	61.18 115	61.12 90	42.51 174	53.02 163	64.74 126	70.11 201	71.50 193	83.22 198	89.49 120

UA-Net		64.62 158	68.23 143	60.42 186	77.53 118	81.38 144	60.08 222	57.47 168	47.01 188	44.75 161	60.68 71	71.32 49	41.84 233	73.27 163	72.25 187	80.83 226	71.68 235

CNLPA		71.37 108	70.27 126	72.66 86	80.79 74	81.33 145	71.07 153	65.75 62	82.36 34	64.80 56	42.46 175	56.49 139	72.70 51	73.00 168	70.52 206	80.84 225	85.76 161

test-LLR		68.23 132	71.61 113	64.28 156	71.37 166	81.32 146	63.98 197	61.03 134	58.62 126	42.96 172	52.74 118	61.65 84	57.74 175	75.64 134	78.09 111	88.61 41	93.21 54

TESTMET0.1,1		67.38 141	71.61 113	62.45 172	66.05 203	81.32 146	63.98 197	55.36 198	58.62 126	42.96 172	52.74 118	61.65 84	57.74 175	75.64 134	78.09 111	88.61 41	93.21 54

Fast-Effi-MVS+		67.59 137	67.56 147	67.62 132	73.67 152	81.14 148	71.12 151	54.79 206	58.88 125	50.61 138	46.70 160	47.05 181	69.12 100	76.06 129	76.44 129	86.43 114	86.65 148

tpmrst		67.15 143	68.12 144	66.03 141	76.21 130	80.98 149	71.27 147	45.05 241	60.69 118	50.63 137	46.95 158	54.15 157	65.30 121	71.80 183	71.77 189	87.72 71	90.48 106

OMC-MVS		74.03 78	75.82 69	71.95 96	79.56 84	80.98 149	75.35 108	63.21 93	84.48 27	61.83 81	61.54 68	66.89 62	69.41 96	76.60 122	74.07 161	82.34 212	86.15 154

ACMP	68.86 7	72.15 100	72.25 102	72.03 94	80.96 71	80.87 151	77.93 79	64.13 74	69.29 76	60.79 95	64.04 60	53.54 160	63.91 130	73.74 157	75.27 144	84.45 180	88.98 126
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
TAPA-MVS	67.10 9	71.45 106	73.47 94	69.10 115	77.04 123	80.78 152	73.81 124	62.10 118	80.80 41	51.28 132	60.91 70	63.80 77	67.98 105	74.59 144	72.42 185	82.37 211	80.97 204
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
CDS-MVSNet		64.22 162	65.89 162	62.28 174	70.05 173	80.59 153	69.91 162	57.98 158	43.53 203	46.58 152	48.22 144	50.76 171	46.45 220	75.68 133	76.08 134	82.70 205	86.34 152
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
usedtu_dtu_shiyan1		62.43 177	64.08 169	60.50 185	59.68 232	80.58 154	66.18 189	61.75 126	53.08 167	36.05 214	36.33 214	41.74 193	51.86 196	77.70 105	77.95 115	87.47 81	81.17 203

EG-PatchMatch MVS		58.73 207	58.03 214	59.55 192	72.32 160	80.49 155	63.44 203	55.55 194	32.49 248	38.31 201	28.87 242	37.22 221	42.84 231	74.30 151	75.70 138	84.84 165	77.14 216

v2v482		63.68 167	62.85 183	64.65 151	68.01 186	80.46 156	71.90 138	57.60 165	44.26 200	42.82 174	39.80 194	38.62 216	61.56 145	73.06 166	74.86 147	86.03 125	88.90 129

v1144		63.00 172	62.39 187	63.70 161	67.72 189	80.27 157	71.23 148	56.40 183	42.51 205	40.81 184	38.12 202	37.73 217	60.42 154	74.46 146	74.55 154	85.64 142	89.12 125

LGP-MVS_train		72.02 102	73.18 95	70.67 104	82.13 63	80.26 158	79.58 58	63.04 95	70.09 69	51.98 129	65.06 57	55.62 148	62.49 141	75.97 130	76.32 132	84.80 170	88.93 127

FC-MVSNet-train		68.83 127	68.29 140	69.47 111	78.35 109	79.94 159	64.72 191	66.38 57	54.96 157	54.51 119	56.75 90	47.91 179	66.91 116	75.57 137	75.75 137	85.92 126	87.12 145

v144192		62.05 185	61.46 194	62.73 171	66.59 201	79.87 160	69.30 166	55.88 188	41.50 212	39.41 191	37.23 205	36.45 225	59.62 158	72.69 173	73.51 166	85.61 143	88.93 127

v1192		62.25 181	61.64 192	62.96 165	66.88 197	79.72 161	69.96 161	55.77 190	41.58 210	39.42 190	37.05 207	35.96 230	60.50 153	74.30 151	74.09 160	85.24 154	88.76 130

dps		64.08 163	63.22 177	65.08 147	75.27 141	79.65 162	66.68 184	46.63 239	56.94 140	55.67 113	43.96 166	43.63 190	64.00 129	69.50 206	69.82 208	82.25 213	79.02 211

v1921920		61.66 189	61.10 197	62.31 173	66.32 202	79.57 163	68.41 171	55.49 196	41.03 213	38.69 195	36.64 213	35.27 233	59.60 159	73.23 164	73.41 168	85.37 149	88.51 134

v148		62.00 186	61.19 196	62.96 165	67.46 195	79.49 164	67.87 173	57.66 164	42.30 206	45.02 160	38.20 201	38.89 215	54.77 186	69.83 203	72.60 184	84.96 159	87.01 146

FMVSNet3		70.41 112	71.89 110	68.68 121	70.89 171	79.42 165	75.63 102	60.97 136	65.32 90	51.06 133	47.37 150	62.05 80	64.90 124	82.49 39	82.27 39	88.64 40	84.34 173

v1240		61.09 192	60.55 201	61.72 178	65.92 206	79.28 166	67.16 181	54.91 203	39.79 219	38.10 202	36.08 216	34.64 235	59.15 163	72.86 169	73.36 170	85.10 156	87.84 141

dmvs_re		67.60 136	67.21 151	68.06 128	74.07 149	79.01 167	73.31 128	68.74 42	58.27 131	42.07 178	49.72 138	43.96 188	60.66 150	76.79 120	78.04 114	89.51 18	84.69 169

CMPMVS	43.63 17	57.67 215	55.43 226	60.28 188	72.01 162	79.00 168	62.77 212	53.23 216	41.77 209	45.42 155	30.74 238	39.03 213	53.01 193	64.81 230	64.65 237	75.26 251	68.03 246
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
V42		62.86 174	62.97 180	62.74 170	60.84 227	78.99 169	71.46 146	57.13 179	46.85 189	44.28 164	38.87 196	40.73 203	57.63 177	72.60 174	74.14 159	85.09 158	88.63 131

Vis-MVSNet (Re-imp)		62.25 181	68.74 135	54.68 223	73.70 151	78.74 170	56.51 231	57.49 167	55.22 155	26.86 240	54.56 104	61.35 86	31.06 243	73.10 165	74.90 146	82.49 208	83.31 182

GBi-Net		69.21 120	70.40 124	67.81 130	69.49 176	78.65 171	74.54 113	60.97 136	65.32 90	51.06 133	47.37 150	62.05 80	63.43 132	77.49 108	78.22 108	87.37 84	83.73 176

test1		69.21 120	70.40 124	67.81 130	69.49 176	78.65 171	74.54 113	60.97 136	65.32 90	51.06 133	47.37 150	62.05 80	63.43 132	77.49 108	78.22 108	87.37 84	83.73 176

FMVSNet2		68.06 133	68.57 136	67.45 135	69.49 176	78.65 171	74.54 113	60.23 148	56.29 147	49.64 143	42.13 179	57.08 130	63.43 132	81.15 58	80.99 67	87.37 84	83.73 176

tpm		64.85 157	66.02 161	63.48 162	74.52 148	78.38 174	70.98 154	44.99 243	51.61 172	43.28 171	47.66 148	53.18 161	60.57 151	70.58 195	71.30 200	86.54 110	89.45 122

ACMH	59.42 14	61.59 190	59.22 209	64.36 155	78.92 102	78.26 175	67.65 175	67.48 51	39.81 218	30.98 234	38.25 200	34.59 236	61.37 148	70.55 196	73.47 167	79.74 232	79.59 208
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
v8		63.44 169	62.58 185	64.43 153	68.28 184	78.07 176	71.82 139	54.85 204	46.70 191	45.20 158	39.40 195	40.91 200	60.54 152	72.85 170	74.39 157	85.92 126	85.76 161

Patchmtry						78.06 177	67.53 177	43.18 249		41.40 179

UniMVSNet (Re)		60.62 195	62.93 182	57.92 202	67.64 191	77.90 178	61.75 216	61.24 131	49.83 178	29.80 236	42.57 172	40.62 204	43.36 229	70.49 197	73.27 173	83.76 189	85.81 160

UniMVSNet_NR-MVSNet		62.30 180	63.51 174	60.89 182	69.48 179	77.83 179	64.07 195	63.94 82	50.03 176	31.17 232	44.82 165	41.12 198	51.37 201	71.02 189	74.81 149	85.30 152	84.95 167

v10		63.00 172	62.22 188	63.90 160	67.88 188	77.78 180	71.59 144	54.34 208	45.37 197	42.76 175	38.53 197	38.93 214	61.05 149	74.39 148	74.52 155	85.75 131	86.04 156

ACMM	66.70 10	70.42 110	68.49 137	72.67 85	82.85 57	77.76 181	77.70 83	64.76 69	64.61 96	60.74 96	49.29 139	53.97 158	65.86 120	74.97 140	75.57 141	84.13 187	83.29 183
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
TAMVS		58.86 205	60.91 198	56.47 217	62.38 222	77.57 182	58.97 226	52.98 217	38.76 228	36.17 212	42.26 178	47.94 178	46.45 220	70.23 200	70.79 203	81.86 217	78.82 212

DCV-MVSNet		69.13 124	69.07 132	69.21 113	77.65 116	77.52 183	74.68 111	57.85 162	54.92 158	55.34 117	55.74 95	55.56 149	66.35 118	75.05 139	76.56 128	83.35 195	88.13 140

test-mter		64.06 164	69.24 131	58.01 201	59.07 234	77.40 184	59.13 225	48.11 233	55.64 153	39.18 193	51.56 129	58.54 113	55.38 184	73.52 161	76.00 135	87.22 94	92.05 86

EPNet_dtu		66.17 148	70.13 127	61.54 179	81.04 70	77.39 185	68.87 169	62.50 114	69.78 71	33.51 226	63.77 61	56.22 141	37.65 239	72.20 176	72.18 188	85.69 137	79.38 209
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
pmmvs5		59.72 199	60.24 203	59.11 197	62.77 220	77.33 186	63.17 205	54.00 211	40.21 217	37.23 206	40.41 189	35.99 229	51.75 197	72.55 175	72.74 181	85.72 136	82.45 196

MIMVSNet		57.78 212	59.71 207	55.53 220	54.79 243	77.10 187	63.89 199	45.02 242	46.59 192	36.79 209	28.36 244	40.77 202	45.84 224	74.97 140	76.58 127	86.87 102	73.60 227

pm-mvs1		59.21 203	59.58 208	58.77 199	67.97 187	77.07 188	64.12 193	57.20 176	34.73 243	36.86 207	35.34 219	40.54 205	43.34 230	74.32 150	73.30 172	83.13 202	81.77 201

Fast-Effi-MVS+-dtu		63.05 171	64.72 167	61.11 181	71.21 169	76.81 189	70.72 156	43.13 251	52.51 171	35.34 219	46.55 161	46.36 182	61.40 147	71.57 187	71.44 195	84.84 165	87.79 142

MSDG		65.57 152	61.57 193	70.24 106	82.02 64	76.47 190	74.46 119	68.73 43	56.52 145	50.33 139	38.47 198	41.10 199	62.42 142	72.12 177	72.94 178	83.47 194	73.37 229

pmmvs4		63.14 170	62.46 186	63.94 159	66.03 204	76.40 191	66.82 183	57.60 165	56.74 141	50.26 140	40.81 188	37.51 219	59.26 162	71.75 185	71.48 194	83.68 193	82.53 194

FMVSNet1		63.48 168	63.07 179	63.97 158	65.31 208	76.37 192	71.77 142	57.90 161	43.32 204	45.66 154	35.06 222	49.43 174	58.57 166	77.49 108	78.22 108	84.59 177	81.60 202

blend_shiyan4		66.60 146	67.24 150	65.85 142	68.02 185	76.25 193	75.94 99	58.03 157	64.52 97	53.78 122	52.14 124	60.47 94	53.51 190	67.10 214	66.76 221	85.79 130	83.46 180

tfpnnormal		58.97 204	56.48 224	61.89 176	71.27 168	76.21 194	66.65 185	61.76 125	32.90 246	36.41 211	27.83 245	29.14 253	50.64 207	73.06 166	73.05 177	84.58 178	83.15 187

DU-MVS		60.87 194	61.82 191	59.76 191	66.69 198	75.87 195	64.07 195	61.96 119	49.31 179	31.17 232	42.76 169	36.95 222	51.37 201	69.67 204	73.20 176	83.30 197	84.95 167

NR-MVSNet		61.08 193	62.09 190	59.90 189	71.96 163	75.87 195	63.60 201	61.96 119	49.31 179	27.95 237	42.76 169	33.85 240	48.82 210	74.35 149	74.05 162	85.13 155	84.45 171

LS3D		64.54 161	62.14 189	67.34 136	80.85 72	75.79 197	69.99 160	65.87 61	60.77 117	44.35 163	42.43 176	45.95 184	65.01 122	69.88 202	68.69 213	77.97 240	71.43 237

PatchMatch-RL		62.22 184	60.69 199	64.01 157	68.74 181	75.75 198	59.27 224	60.35 145	56.09 149	53.80 121	47.06 156	36.45 225	64.80 125	68.22 210	67.22 217	77.10 243	74.02 224

PatchT		60.46 196	63.85 172	56.51 216	65.95 205	75.68 199	47.34 248	41.39 256	53.89 166	41.40 179	37.84 203	50.30 173	57.29 178	72.76 171	73.27 173	85.67 138	83.23 185

thisisatest0515		59.37 202	60.68 200	57.84 204	64.39 212	75.65 200	58.56 227	53.86 212	41.55 211	42.12 177	40.40 190	39.59 211	47.09 218	71.69 186	73.79 163	81.02 224	82.08 199

TranMVSNet+NR-MVSNet		60.38 197	61.30 195	59.30 195	68.34 183	75.57 201	63.38 204	63.78 86	46.74 190	27.73 238	42.56 173	36.84 223	47.66 215	70.36 198	74.59 153	84.91 162	82.46 195

wanda-best-256-512		57.69 213	57.90 216	57.46 208	48.58 254	75.44 202	63.15 206	57.47 168	39.27 222	38.64 196	34.66 224	40.34 206	51.44 199	66.38 216	66.54 223	85.46 145	82.64 189

FE-blended-shiyan7		57.69 213	57.90 216	57.46 208	48.58 254	75.44 202	63.15 206	57.47 168	39.27 222	38.64 196	34.66 224	40.34 206	51.44 199	66.38 216	66.54 223	85.46 145	82.64 189

usedtu_blend_shiyan5		62.84 175	63.39 175	62.21 175	48.58 254	75.44 202	74.43 120	57.47 168	39.26 225	53.78 122	52.14 124	60.47 94	53.51 190	66.38 216	66.54 223	85.46 145	83.46 180

FE-MVSNET3		61.91 187	63.26 176	60.33 187	48.58 254	75.44 202	63.15 206	57.47 168	39.27 222	53.78 122	52.14 124	60.47 94	53.51 190	66.38 216	66.54 223	85.46 145	82.59 191

blended_shiyan8		57.49 217	57.71 219	57.24 211	48.52 258	75.34 206	62.85 210	57.32 175	38.77 227	38.43 199	34.41 227	40.31 208	50.92 204	66.25 221	66.37 228	85.37 149	82.55 193

blended_shiyan6		57.50 216	57.73 218	57.23 212	48.51 259	75.34 206	62.85 210	57.33 173	38.78 226	38.38 200	34.46 226	40.29 209	50.91 205	66.27 220	66.37 228	85.37 149	82.59 191

Effi-MVS+-dtu		64.58 159	64.08 169	65.16 146	73.04 158	75.17 208	70.68 158	56.23 186	54.12 165	44.71 162	47.42 149	51.10 169	63.82 131	68.08 211	66.32 231	82.47 209	86.38 151

IterMVS-LS		66.08 149	66.56 156	65.51 143	73.67 152	74.88 209	70.89 155	53.55 214	50.42 175	48.32 148	50.59 134	55.66 147	61.83 143	73.93 153	74.42 156	84.82 169	86.01 157
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
PLC	64.00 12	68.54 128	66.66 154	70.74 103	80.28 79	74.88 209	72.64 132	63.70 88	69.26 77	55.71 112	47.24 153	55.31 150	70.42 83	72.05 179	70.67 204	81.66 219	77.19 215
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
v7n		57.04 219	56.64 223	57.52 206	62.85 219	74.75 211	61.76 215	51.80 222	35.58 242	36.02 215	32.33 233	33.61 241	50.16 208	67.73 212	70.34 207	82.51 207	82.12 198

gbinet_0.2-2-1-0.02		56.72 220	57.64 220	55.64 219	45.57 262	74.69 212	62.04 214	57.17 178	35.71 241	35.71 216	33.73 229	41.66 194	48.54 211	66.06 223	66.43 227	84.83 168	85.22 165

TransMVSNet (Re)		57.83 210	56.90 222	58.91 198	72.26 161	74.69 212	63.57 202	61.42 130	32.30 249	32.65 228	33.97 228	35.96 230	39.17 237	73.84 156	72.84 180	84.37 181	74.69 222

ADS-MVSNet		58.40 209	59.16 210	57.52 206	65.80 207	74.57 214	60.26 220	40.17 260	50.51 174	38.01 203	40.11 193	44.72 186	59.36 161	64.91 228	66.55 222	81.53 220	72.72 232

ACMH+	60.36 13	61.16 191	58.38 211	64.42 154	77.37 122	74.35 215	68.45 170	62.81 103	45.86 195	38.48 198	35.71 217	37.35 220	59.81 157	67.24 213	69.80 210	79.58 233	78.32 213

Baseline_NR-MVSNet		59.47 201	60.28 202	58.54 200	66.69 198	73.90 216	61.63 217	62.90 102	49.15 183	26.87 239	35.18 221	37.62 218	48.20 213	69.67 204	73.61 165	84.92 160	82.82 188

MDTV_nov1_ep13_2view		54.47 228	54.61 227	54.30 227	60.50 228	73.82 217	57.92 228	43.38 248	39.43 221	32.51 229	33.23 230	34.05 238	47.26 217	62.36 236	66.21 232	84.24 183	73.19 230

test0.0.03 1		57.35 218	59.89 206	54.38 226	71.37 166	73.45 218	52.71 238	61.03 134	46.11 194	26.33 241	41.73 181	44.08 187	29.72 245	71.43 188	70.90 201	85.10 156	71.56 236

UniMVSNet_ETH3D		57.83 210	56.46 225	59.43 194	63.24 217	73.22 219	67.70 174	55.58 193	36.17 237	36.84 208	32.64 231	35.14 234	51.50 198	65.81 224	69.81 209	81.73 218	82.44 197

pmmvs6		54.20 229	53.54 231	54.97 221	63.22 218	72.98 220	60.17 221	52.32 221	26.77 259	34.30 223	23.29 256	36.23 227	40.33 236	68.77 208	68.76 212	79.47 235	78.00 214

MVS-HIRNet		53.86 231	53.02 233	54.85 222	60.30 229	72.36 221	44.63 257	42.20 254	39.45 220	43.47 168	21.66 260	34.00 239	55.47 183	65.42 226	67.16 218	83.02 203	71.08 240

IterMVS		61.87 188	63.55 173	59.90 189	67.29 196	72.20 222	67.34 180	48.56 231	47.48 187	37.86 205	47.07 155	48.27 175	54.08 188	72.12 177	73.71 164	84.30 182	83.99 175
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
IterMVS-SCA-FT		60.21 198	62.97 180	57.00 214	66.64 200	71.84 223	67.53 177	46.93 238	47.56 186	36.77 210	46.85 159	48.21 176	52.51 194	70.36 198	72.40 186	71.63 259	83.53 179

USDC		59.69 200	60.03 205	59.28 196	64.04 213	71.84 223	63.15 206	55.36 198	54.90 159	35.02 220	48.34 143	29.79 252	58.16 167	70.60 194	71.33 199	79.99 230	73.42 228

SCA		63.90 165	66.67 153	60.66 183	73.75 150	71.78 225	59.87 223	43.66 247	61.13 116	45.03 159	51.64 128	59.45 110	57.92 172	70.96 190	70.80 202	83.71 191	80.92 205

anonymousdsp		54.99 224	57.24 221	52.36 230	53.82 245	71.75 226	51.49 240	48.14 232	33.74 244	33.66 225	38.34 199	36.13 228	47.54 216	64.53 232	70.60 205	79.53 234	85.59 163

dtuonly		62.74 176	63.91 171	61.36 180	61.12 226	71.54 227	70.69 157	50.99 224	52.81 169	40.13 187	42.43 176	51.07 170	62.78 137	71.77 184	71.63 191	82.47 209	86.15 154

pmnet_mix02		53.92 230	53.30 232	54.65 225	61.89 223	71.33 228	54.54 236	54.17 210	40.38 215	34.65 221	34.76 223	30.68 251	40.44 235	60.97 238	63.71 239	82.19 214	71.24 239

CR-MVSNet		62.31 179	64.75 165	59.47 193	68.63 182	71.29 229	67.53 177	43.18 249	55.83 150	41.40 179	41.04 185	55.85 143	57.29 178	72.76 171	73.27 173	78.77 237	83.23 185

RPMNet		58.63 208	62.80 184	53.76 228	67.59 193	71.29 229	54.60 235	38.13 261	55.83 150	35.70 217	41.58 182	53.04 162	47.89 214	66.10 222	67.38 215	78.65 239	84.40 172

our_test_3					63.32 215	71.07 231	55.90 232

LTVRE_ROB	47.26 16	49.41 244	49.91 246	48.82 237	64.76 210	69.79 232	49.05 243	47.12 237	20.36 266	16.52 257	36.65 212	26.96 257	50.76 206	60.47 239	63.16 242	64.73 262	72.00 234
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
Anonymous20231206		52.23 234	52.80 236	51.56 232	64.70 211	69.41 233	51.01 241	58.60 153	36.63 234	22.44 248	21.80 259	31.42 247	30.52 244	66.79 215	67.83 214	82.10 215	75.73 218

WR-MVS		51.02 236	54.56 228	46.90 244	63.84 214	69.23 234	44.78 256	56.38 184	38.19 229	14.19 261	37.38 204	36.82 224	22.39 257	60.14 240	66.20 233	79.81 231	73.95 226

FE-MVSNET2		50.42 238	51.98 240	48.61 239	44.79 263	68.96 235	52.01 239	55.50 195	32.55 247	19.88 253	21.60 261	28.20 255	35.80 240	68.31 209	71.76 190	83.69 192	72.45 233

CHOSEN 280x420		62.23 183	66.57 155	57.17 213	59.88 230	68.92 236	61.20 219	42.28 253	54.17 164	39.57 188	47.78 147	64.97 70	62.68 138	73.85 155	69.52 211	77.43 241	86.75 147

CVMVSNet		54.92 226	58.16 212	51.13 234	62.61 221	68.44 237	55.45 234	52.38 220	42.28 207	21.45 249	47.10 154	46.10 183	37.96 238	64.42 233	63.81 238	76.92 244	75.01 221

dtuonlycased		50.09 241	48.12 249	52.39 229	52.04 248	68.20 238	55.54 233	49.33 228	36.78 232	32.91 227	24.24 252	39.38 212	48.29 212	46.71 258	50.09 259	76.23 245	71.43 237

pmmvs-eth3d		55.20 221	53.95 230	56.65 215	57.34 240	67.77 239	57.54 229	53.74 213	40.93 214	41.09 183	31.19 237	29.10 254	49.07 209	65.54 225	67.28 216	81.14 222	75.81 217

WR-MVS_H		49.62 243	52.63 237	46.11 247	58.80 235	67.58 240	46.14 254	54.94 201	36.51 235	13.63 264	36.75 211	35.67 232	22.10 258	56.43 249	62.76 243	81.06 223	72.73 231

COLMAP_ROB	51.17 15	55.13 222	52.90 235	57.73 205	73.47 155	67.21 241	62.13 213	55.82 189	47.83 185	34.39 222	31.60 235	34.24 237	44.90 227	63.88 235	62.52 244	75.67 249	63.02 256
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
testgi		48.51 246	50.53 243	46.16 246	64.78 209	67.15 242	41.54 260	54.81 205	29.12 254	17.03 255	32.07 234	31.98 243	20.15 261	65.26 227	67.00 219	78.67 238	61.10 261

FMVSNet5		58.86 205	60.24 203	57.25 210	52.66 247	66.25 243	63.77 200	52.86 219	57.85 136	37.92 204	36.12 215	52.22 166	51.37 201	70.88 191	71.43 196	84.92 160	66.91 248

PEN-MVS		51.04 235	52.94 234	48.82 237	61.45 225	66.00 244	48.68 244	57.20 176	36.87 231	15.36 259	36.98 208	32.72 242	28.77 249	57.63 245	66.37 228	81.44 221	74.00 225

CP-MVSNet		50.57 237	52.60 238	48.21 241	58.77 236	65.82 245	48.17 245	56.29 185	37.41 230	16.59 256	37.14 206	31.95 244	29.21 246	56.60 248	63.71 239	80.22 228	75.56 219

PS-CasMVS		50.17 239	52.02 239	48.02 242	58.60 237	65.54 246	48.04 247	56.19 187	36.42 236	16.42 258	35.68 218	31.33 248	28.85 248	56.42 250	63.54 241	80.01 229	75.18 220

TDRefinement		52.70 232	51.02 242	54.66 224	57.41 239	65.06 247	61.47 218	54.94 201	44.03 201	33.93 224	30.13 240	27.57 256	46.17 222	61.86 237	62.48 245	74.01 255	66.06 249

test20.03		47.23 249	48.69 248	45.53 248	63.28 216	64.39 248	41.01 261	56.93 181	29.16 253	15.21 260	23.90 253	30.76 250	17.51 264	64.63 231	65.26 234	79.21 236	62.71 258

DTE-MVSNet		49.82 242	51.92 241	47.37 243	61.75 224	64.38 249	45.89 255	57.33 173	36.11 238	12.79 266	36.87 209	31.92 245	25.73 254	58.01 243	65.22 235	80.75 227	70.93 241

PatchmatchNet2					56.14 241	64.21 250	48.11 246
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo. CVPR 2021
SixPastTwentyTwo		49.11 245	49.22 247	48.99 236	58.54 238	64.14 251	47.18 249	47.75 234	31.15 251	24.42 244	41.01 186	26.55 258	44.04 228	54.76 254	58.70 251	71.99 258	68.21 244

TinyColmap		52.66 233	50.09 245	55.65 218	59.72 231	64.02 252	57.15 230	52.96 218	40.28 216	32.51 229	32.42 232	20.97 267	56.65 180	63.95 234	65.15 236	74.91 252	63.87 254

N_pmnet		47.67 247	47.00 251	48.45 240	54.72 244	62.78 253	46.95 250	51.25 223	36.01 239	26.09 243	26.59 249	25.93 262	35.50 242	55.67 252	59.01 249	76.22 247	63.04 255

MDA-MVSNet-bldmvs		44.15 252	42.27 257	46.34 245	38.34 265	62.31 254	46.28 252	55.74 191	29.83 252	20.98 251	27.11 248	16.45 273	41.98 232	41.11 264	57.47 252	74.72 253	61.65 260

PM-MVS		50.11 240	50.38 244	49.80 235	47.23 261	62.08 255	50.91 242	44.84 244	41.90 208	36.10 213	35.22 220	26.05 260	46.83 219	57.64 244	55.42 256	72.90 256	74.32 223

FE-MVSNET		44.36 251	46.68 252	41.65 250	37.55 266	61.05 256	42.06 259	54.34 208	27.09 257	9.86 271	20.55 262	25.56 263	28.72 250	60.12 241	66.83 220	77.36 242	65.56 251

new-patchmatchnet		42.21 253	42.97 254	41.33 252	53.05 246	59.89 257	39.38 262	49.61 226	28.26 256	12.10 267	22.17 258	21.54 266	19.22 262	50.96 257	56.04 254	74.61 254	61.92 259

usedtu_dtu_shiyan2		40.99 255	42.22 258	39.56 254	22.63 272	59.44 258	46.80 251	43.69 246	19.05 268	21.04 250	16.27 270	23.77 264	27.46 252	53.16 256	55.09 257	75.73 248	68.78 242

RPSCF		55.07 223	58.06 213	51.57 231	48.87 253	58.95 259	53.68 237	41.26 258	62.42 108	45.88 153	54.38 109	54.26 156	53.75 189	57.15 246	53.53 258	66.01 261	65.75 250

MIMVSNet1		40.84 256	43.46 253	37.79 256	32.14 267	58.92 260	39.24 263	50.83 225	27.00 258	11.29 268	16.76 268	26.53 259	17.75 263	57.14 247	61.12 247	75.46 250	56.78 262

FC-MVSNet-test		47.24 248	54.37 229	38.93 255	59.49 233	58.25 261	34.48 266	53.36 215	45.66 196	6.66 272	50.62 133	42.02 191	16.62 265	58.39 242	61.21 246	62.99 263	64.40 253

EU-MVSNet		44.84 250	47.85 250	41.32 253	49.26 252	56.59 262	43.07 258	47.64 236	33.03 245	13.82 262	36.78 210	30.99 249	24.37 255	53.80 255	55.57 255	69.78 260	68.21 244

gm-plane-assit		54.99 224	57.99 215	51.49 233	69.27 180	54.42 263	32.32 267	42.59 252	21.18 264	13.71 263	23.61 254	43.84 189	60.21 155	87.09 6	86.55 5	90.81 4	89.28 123

pmmvs3		41.86 254	42.29 256	41.36 251	39.80 264	52.66 264	38.93 264	35.85 265	23.40 263	20.22 252	19.30 263	20.84 268	40.56 234	55.98 251	58.79 250	72.80 257	65.03 252

ambc			42.30 255		50.36 251	49.51 265	35.47 265		32.04 250	23.53 245	17.36 265	8.95 277	29.06 247	64.88 229	56.26 253	61.29 264	67.12 247

FPMVS		39.11 257	36.39 259	42.28 249	55.97 242	45.94 266	46.23 253	41.57 255	35.73 240	22.61 246	23.46 255	19.82 269	28.32 251	43.57 261	40.67 263	58.96 265	45.54 264

new_pmnet		33.19 258	35.52 260	30.47 258	27.55 271	45.31 267	29.29 268	30.92 266	29.00 255	9.88 270	18.77 264	17.64 271	26.77 253	44.07 260	45.98 261	58.41 266	47.87 263

WB-MVS		30.42 260	32.63 262	27.84 259	51.51 250	41.64 268	17.75 272	55.06 200	20.11 267	2.46 277	26.13 251	16.63 272	3.90 273	44.91 259	44.54 262	36.34 271	34.48 268

PMMVS2		20.45 263	22.31 265	18.27 265	20.52 273	26.73 269	14.85 274	28.43 268	13.69 271	0.79 278	10.35 272	9.10 276	3.83 275	27.64 267	32.87 265	41.17 268	35.81 266

PMVS	27.44 18	32.08 259	29.07 263	35.60 257	48.33 260	24.79 270	26.97 269	41.34 257	20.45 265	22.50 247	17.11 267	18.64 270	20.44 260	41.99 263	38.06 264	54.02 267	42.44 265
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
Gipuma		24.91 262	24.61 264	25.26 261	31.47 268	21.59 271	18.06 271	37.53 262	25.43 261	10.03 269	4.18 276	4.25 280	14.85 266	43.20 262	47.03 260	39.62 269	26.55 272
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
test_method		28.15 261	34.48 261	20.76 262	6.76 276	21.18 272	21.03 270	18.41 269	36.77 233	17.52 254	15.67 271	31.63 246	24.05 256	41.03 265	26.69 267	36.82 270	68.38 243

MVE	15.98 19	14.37 266	16.36 267	12.04 267	7.72 275	20.24 273	5.90 278	29.05 267	8.28 275	3.92 274	4.72 275	2.42 281	9.57 269	18.89 269	31.46 266	16.07 276	28.53 270
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
DeepMVS_CX						19.81 274	17.01 273	10.02 270	23.61 262	5.85 273	17.21 266	8.03 278	21.13 259	22.60 268		21.42 275	30.01 269

E-PMN		15.08 264	11.65 269	19.08 263	28.73 269	12.31 275	6.95 277	36.87 264	10.71 274	3.63 275	5.13 273	2.22 284	13.81 268	11.34 271	18.50 269	24.49 273	21.32 273

EMVS		14.40 265	10.71 271	18.70 264	28.15 270	12.09 276	7.06 276	36.89 263	11.00 273	3.56 276	4.95 274	2.27 283	13.91 267	10.13 273	16.06 270	22.63 274	18.51 274

tmp_tt				16.09 266	13.07 274	8.12 277	13.61 275	2.08 271	55.09 156	30.10 235	40.26 191	22.83 265	5.35 271	29.91 266	25.25 268	32.33 272

VLMVS_CLIP		11.46 267	18.27 266	3.50 268	3.73 277	5.54 278	2.13 280	0.48 272	18.85 269	0.26 280	28.51 243	9.68 275	7.31 270	17.28 270	13.56 271	7.11 277	34.49 267

VLMVS		9.08 268	15.28 268	1.84 269	1.39 278	3.31 279	1.20 281	0.09 274	18.54 270	0.39 279	27.68 246	12.43 274	3.90 273	9.16 274	8.34 273	4.04 278	27.51 271

MVS_clip		6.46 269	10.77 270	1.43 270	0.96 279	2.36 280	0.77 282	0.18 273	11.97 272	0.04 282	16.38 269	7.57 279	5.17 272	10.69 272	8.74 272	1.48 279	17.71 275

MVS_baseline		1.61 270	2.81 272	0.21 271	0.06 280	0.07 281	0.02 284	0.00 277	2.84 276	0.00 283	4.11 277	2.29 282	1.18 276	1.23 275	1.30 274	0.00 281	7.85 276

testmvs		0.05 271	0.08 273	0.01 272	0.00 282	0.01 282	0.03 283	0.01 276	0.05 277	0.00 283	0.14 279	0.01 285	0.03 279	0.05 276	0.05 275	0.01 280	0.24 278

test123		0.05 271	0.08 273	0.01 272	0.00 282	0.01 282	0.01 285	0.00 277	0.05 277	0.00 283	0.16 278	0.00 286	0.04 277	0.02 277	0.05 275	0.00 281	0.26 277

uanet_test		0.00 273	0.00 275	0.00 274	0.00 282	0.00 284	0.00 286	0.00 277	0.00 279	0.00 283	0.00 280	0.00 286	0.00 280	0.00 278	0.00 277	0.00 281	0.00 279

sosnet-low-res		0.00 273	0.00 275	0.00 274	0.00 282	0.00 284	0.00 286	0.00 277	0.00 279	0.00 283	0.00 280	0.00 286	0.00 280	0.00 278	0.00 277	0.00 281	0.00 279

sosnet		0.00 273	0.00 275	0.00 274	0.00 282	0.00 284	0.00 286	0.00 277	0.00 279	0.00 283	0.00 280	0.00 286	0.00 280	0.00 278	0.00 277	0.00 281	0.00 279

PatchmatchNet1												25.98 261	35.57 241	55.54 253	59.02 248	76.23 245	62.78 257
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo. CVPR 2021
PatchmatchNet3										26.10 242	26.55 250
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo. CVPR 2021
TestfortrainingZip							88.32 10	77.84 4		88.26 1						90.10 7

RE-MVS-def										31.47 231

9.14												84.47 9

SR-MVS					86.33 50			67.54 50				80.78 25

MTAPA										78.32 14		79.42 29

MTMP										76.04 18		76.65 33

Patchmatch-RL test							2.17 279

mPP-MVS					86.96 45							70.61 52

NP-MVS									81.60 39

High-resolution multi-view benchmark