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.

DPM-MVS		85.41 10	86.72 15	83.89 10	91.66 12	91.92 14	90.49 3	78.09 4	86.90 18	73.95 20	74.52 34	82.01 15	79.29 12	90.24 1	90.65 1	89.86 6	90.78 71

DELS-MVS		79.49 31	79.84 41	79.08 28	88.26 37	92.49 8	84.12 25	70.63 27	65.27 80	69.60 36	61.29 62	66.50 60	72.75 42	88.07 3	88.03 2	89.13 12	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
MCST-MVS		85.75 8	86.99 12	84.31 6	94.07 3	92.80 7	88.15 8	79.10 3	85.66 23	70.72 30	76.50 32	80.45 20	82.17 3	88.35 2	87.49 3	91.63 2	97.65 4

CNVR-MVS		85.96 7	87.58 10	84.06 8	92.58 5	92.40 10	87.62 10	77.77 5	88.44 14	75.93 17	79.49 25	81.97 16	81.65 4	87.04 6	86.58 4	88.79 17	97.18 7

gm-plane-assit		54.99 174	57.99 170	51.49 183	69.27 142	54.42 207	32.32 210	42.59 197	21.18 211	13.71 207	23.61 198	43.84 144	60.21 116	87.09 5	86.55 5	90.81 4	89.28 88

SMA-MVS		85.24 11	88.27 8	81.72 15	91.74 11	90.71 20	86.71 12	73.16 19	90.56 9	74.33 19	83.07 18	85.88 4	77.16 19	86.28 9	85.58 6	87.23 57	95.77 14
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
CHOSEN 1792x2688		72.55 69	71.98 78	73.22 60	86.57 46	92.41 9	75.63 71	66.77 51	62.08 86	52.32 88	30.27 188	50.74 129	66.14 82	86.22 11	85.41 7	91.90 1	96.75 12

DeepPCF-MVS	76.94 1	83.08 19	87.77 9	77.60 35	90.11 19	90.96 19	78.48 54	72.63 22	93.10 4	65.84 42	80.67 23	81.55 18	74.80 29	85.94 12	85.39 8	83.75 138	96.77 11

HPM-MVS++		85.64 9	88.43 6	82.39 12	92.65 4	90.24 26	85.83 16	74.21 11	90.68 8	75.63 18	86.77 13	84.15 7	78.68 15	86.33 8	85.26 9	87.32 53	95.60 18

gg-mvs-nofinetune		62.34 134	66.19 117	57.86 159	76.15 103	88.61 39	71.18 108	41.24 204	25.74 207	13.16 209	22.91 201	63.97 68	54.52 145	85.06 15	85.25 10	90.92 3	91.78 60

CSCG		82.90 20	84.52 22	81.02 18	91.85 10	93.43 6	87.14 11	74.01 14	81.96 33	76.14 15	70.84 38	82.49 12	69.71 61	82.32 41	85.18 11	87.26 56	95.40 22

SED-MVS		88.94 1	90.98 1	86.56 1	92.53 6	95.09 1	88.55 5	76.83 7	94.16 1	86.57 1	90.85 6	87.07 1	86.18 1	86.36 7	85.08 12	88.67 21	98.21 3

MVS_0304		79.43 33	82.20 31	76.20 42	84.22 53	91.79 16	81.82 37	63.81 70	76.83 51	61.71 56	66.37 49	75.52 31	76.38 23	85.54 13	85.03 13	89.28 11	94.32 34

DeepC-MVS_fast	75.41 2	81.69 24	82.10 33	81.20 17	91.04 16	87.81 51	83.42 27	74.04 13	83.77 27	71.09 28	66.88 47	72.44 38	79.48 10	85.08 14	84.97 14	88.12 39	93.78 41
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
NCCC		84.16 16	85.46 20	82.64 11	92.34 7	90.57 23	86.57 13	76.51 8	86.85 20	72.91 23	77.20 31	78.69 26	79.09 14	84.64 19	84.88 15	88.44 29	95.41 21

CANet		80.90 28	82.93 29	78.53 31	86.83 45	92.26 11	81.19 42	66.95 49	81.60 36	69.90 33	66.93 46	74.80 32	76.79 20	84.68 18	84.77 16	89.50 9	95.50 19

DVP-MVS		88.07 2	90.73 2	84.97 4	91.98 9	95.01 2	87.86 9	76.88 6	93.90 2	85.15 2	90.11 8	86.90 2	79.46 11	86.26 10	84.67 17	88.50 28	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
MSP-MVS		87.87 3	90.57 3	84.73 5	89.38 27	91.60 17	88.24 7	74.15 12	93.55 3	82.28 3	94.99 1	83.21 10	85.96 2	87.67 4	84.67 17	88.32 31	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
3Dnovator	70.49 5	78.42 39	76.77 56	80.35 20	91.43 14	90.27 25	81.84 36	70.79 26	72.10 57	71.95 24	50.02 97	67.86 56	77.47 18	82.89 31	84.24 19	88.61 24	89.99 80

DeepC-MVS	74.46 3	80.30 30	81.05 36	79.42 24	87.42 41	88.50 40	83.23 28	73.27 18	82.78 30	71.01 29	62.86 57	69.93 51	74.80 29	84.30 20	84.20 20	86.79 68	94.77 27
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
PHI-MVS		79.43 33	84.06 24	74.04 56	86.15 48	91.57 18	80.85 46	68.90 40	82.22 32	51.81 91	78.10 27	74.28 33	70.39 58	84.01 23	84.00 21	86.14 79	94.24 35

MVSTER		76.92 49	79.92 40	73.42 59	74.98 111	82.97 90	78.15 57	63.41 74	78.02 46	64.41 46	67.54 44	72.80 37	71.05 53	83.29 29	83.73 22	88.53 27	91.12 66

ACMMP_NAP		83.54 17	86.37 17	80.25 21	89.57 26	90.10 28	85.27 20	71.66 23	87.38 15	73.08 22	84.23 17	80.16 21	75.31 25	84.85 16	83.64 23	86.57 70	94.21 37

SteuartSystems-ACMMP		82.51 21	85.35 21	79.20 26	90.25 17	89.39 33	84.79 21	70.95 25	82.86 29	68.32 38	86.44 14	77.19 27	73.07 39	83.63 24	83.64 23	87.82 40	94.34 33
Skip Steuart: Steuart Systems R&D Blog.
MVS_Test		75.22 56	76.69 57	73.51 57	79.30 77	88.82 37	80.06 49	58.74 111	69.77 64	57.50 74	59.78 69	61.35 78	75.31 25	82.07 44	83.60 25	90.13 5	91.41 63

DPE-MVS		87.60 4	90.44 4	84.29 7	92.09 8	93.44 5	88.69 4	75.11 9	93.06 5	80.80 5	94.23 2	86.70 3	81.44 5	84.84 17	83.52 26	87.64 46	97.28 5
Kehua Chen, Zhenlong Yuan, Tianlu Mao, Zhaoqi Wang: Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization. AAAI2025
MVS_111021_HR		77.42 46	78.40 47	76.28 41	86.95 43	90.68 21	77.41 64	70.56 30	66.21 74	62.48 52	66.17 50	63.98 67	72.08 47	82.87 32	83.15 27	88.24 34	95.71 16

MAR-MVS		77.19 48	78.37 48	75.81 46	89.87 21	90.58 22	79.33 53	65.56 60	77.62 49	58.33 69	59.24 70	67.98 54	74.83 28	82.37 40	83.12 28	86.95 63	87.67 105
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
CDPH-MVS		79.39 36	82.13 32	76.19 43	89.22 30	88.34 42	84.20 24	71.00 24	79.67 43	56.97 75	77.77 28	72.24 42	68.50 73	81.33 51	82.74 29	87.23 57	92.84 50

canonicalmvs		77.65 43	79.59 42	75.39 47	81.52 64	89.83 32	81.32 41	60.74 103	80.05 41	66.72 40	68.43 42	65.09 63	74.72 31	78.87 70	82.73 30	87.32 53	92.16 55

xxxxxxxxxxxxxcwj		84.33 14	83.20 26	85.64 2	94.57 1	94.55 3	91.01 1	79.94 1	89.15 11	79.85 6	92.37 3	44.71 142	79.75 7	83.52 25	82.72 31	88.75 19	95.37 23

SF-MVS		87.30 5	88.71 5	85.64 2	94.57 1	94.55 3	91.01 1	79.94 1	89.15 11	79.85 6	92.37 3	83.29 9	79.75 7	83.52 25	82.72 31	88.75 19	95.37 23

train_agg		83.35 18	86.93 14	79.17 27	89.70 23	88.41 41	85.60 19	72.89 21	86.31 21	66.58 41	90.48 7	82.24 14	73.06 40	83.10 30	82.64 33	87.21 61	95.30 25

APDe-MVS		86.37 6	88.41 7	84.00 9	91.43 14	91.83 15	88.34 6	74.67 10	91.19 6	81.76 4	91.13 5	81.94 17	80.07 6	83.38 27	82.58 34	87.69 44	96.78 10

FMVSNet3		70.41 81	71.89 80	68.68 88	70.89 133	79.42 123	75.63 71	60.97 99	65.32 77	51.06 93	47.37 108	62.05 72	64.90 87	82.49 36	82.27 35	88.64 23	84.34 130

QAPM		77.50 45	77.43 50	77.59 36	91.52 13	92.00 13	81.41 40	70.63 27	66.22 73	58.05 70	54.70 79	71.79 44	74.49 32	82.46 37	82.04 36	89.46 10	92.79 52

3Dnovator+	70.16 6	77.87 42	77.29 52	78.55 30	89.25 29	88.32 43	80.09 48	67.95 44	74.89 56	71.83 26	52.05 91	70.68 48	76.27 24	82.27 42	82.04 36	85.92 84	90.77 72

EPNet		79.28 37	82.25 30	75.83 45	88.31 36	90.14 27	79.43 52	68.07 43	81.76 35	61.26 58	77.26 30	70.08 50	70.06 59	82.43 39	82.00 38	87.82 40	92.09 56
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
IB-MVS	64.48 11	69.02 87	68.97 98	69.09 86	81.75 63	89.01 36	64.50 146	64.91 64	56.65 105	62.59 51	47.89 104	45.23 140	51.99 150	69.18 165	81.88 39	88.77 18	92.93 49
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
casdiffmvs		75.20 57	75.69 63	74.63 55	79.26 78	89.07 35	78.47 55	63.59 73	67.05 71	63.79 47	55.72 76	60.32 84	73.58 35	82.16 43	81.78 40	89.08 13	93.72 43

HQP-MVS		78.26 40	80.91 37	75.17 50	85.67 50	84.33 82	83.01 30	69.38 35	79.88 42	55.83 76	79.85 24	64.90 65	70.81 54	82.46 37	81.78 40	86.30 75	93.18 48

PVSNet_BlendedMVS		76.84 50	78.47 45	74.95 51	82.37 58	89.90 30	75.45 75	65.45 61	74.99 54	70.66 31	63.07 55	58.27 96	67.60 76	84.24 21	81.70 42	88.18 35	97.10 8

PVSNet_Blended		76.84 50	78.47 45	74.95 51	82.37 58	89.90 30	75.45 75	65.45 61	74.99 54	70.66 31	63.07 55	58.27 96	67.60 76	84.24 21	81.70 42	88.18 35	97.10 8

SD-MVS		84.31 15	86.96 13	81.22 16	88.98 31	88.68 38	85.65 17	73.85 15	89.09 13	79.63 8	87.34 12	84.84 5	73.71 34	82.66 34	81.60 44	85.48 101	94.51 31
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
DI_MVS_plusplus_trai		73.94 63	74.85 66	72.88 62	76.57 100	86.80 60	80.41 47	61.47 94	62.35 85	59.44 67	47.91 103	68.12 53	72.24 45	82.84 33	81.50 45	87.15 62	94.42 32

TSAR-MVS + GP.		82.27 23	85.98 18	77.94 33	80.72 71	88.25 44	81.12 43	67.71 45	87.10 16	73.31 21	85.23 15	83.68 8	76.64 21	80.43 60	81.47 46	88.15 37	95.66 17

APD-MVS		84.83 12	87.00 11	82.30 13	89.61 25	89.21 34	86.51 14	73.64 16	90.98 7	77.99 12	89.89 9	80.04 23	79.18 13	82.00 46	81.37 47	86.88 65	95.49 20
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
TSAR-MVS + ACMM		81.59 26	85.84 19	76.63 39	89.82 22	86.53 63	86.32 15	66.72 52	85.96 22	65.43 43	88.98 11	82.29 13	67.57 78	82.06 45	81.33 48	83.93 136	93.75 42

CS-MVS		75.18 58	78.59 44	71.20 70	77.74 86	87.69 52	73.93 89	58.81 110	69.17 69	55.73 77	67.86 43	66.89 57	72.87 41	82.50 35	81.29 49	88.15 37	94.71 29

CLD-MVS		77.36 47	77.29 52	77.45 37	82.21 60	88.11 46	81.92 35	68.96 39	77.97 47	69.62 35	62.08 58	59.44 89	73.57 36	81.75 48	81.27 50	88.41 30	90.39 76
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
ETV-MVS		76.25 52	80.22 39	71.63 69	78.23 82	87.95 50	72.75 92	60.27 107	77.50 50	57.73 71	71.53 37	66.60 59	73.16 38	80.99 56	81.23 51	87.63 47	95.73 15

HFP-MVS		82.48 22	84.12 23	80.56 19	90.15 18	87.55 54	84.28 23	69.67 34	85.22 24	77.95 13	84.69 16	75.94 30	75.04 27	81.85 47	81.17 52	86.30 75	92.40 54

zzz-MVS		81.65 25	83.10 27	79.97 23	88.14 38	87.62 53	83.96 26	69.90 31	86.92 17	77.67 14	72.47 36	78.74 25	74.13 33	81.59 50	81.15 53	86.01 83	93.19 47

ACMMPR		80.62 29	82.98 28	77.87 34	88.41 33	87.05 59	83.02 29	69.18 37	83.91 26	68.35 37	82.89 19	73.64 35	72.16 46	80.78 58	81.13 54	86.10 80	91.43 61

baseline1		71.47 74	72.02 77	70.82 73	80.56 72	84.51 78	76.61 68	66.93 50	56.22 109	48.66 104	55.40 77	60.43 83	62.55 101	83.35 28	80.99 55	89.60 7	83.28 139

FMVSNet2		68.06 95	68.57 100	67.45 98	69.49 138	78.65 128	74.54 82	60.23 108	56.29 108	49.64 102	42.13 135	57.08 101	63.43 94	81.15 54	80.99 55	87.37 50	83.73 133

MP-MVS		80.94 27	83.49 25	77.96 32	88.48 32	88.16 45	82.82 32	69.34 36	80.79 39	69.67 34	82.35 20	77.13 28	71.60 51	80.97 57	80.96 57	85.87 87	94.06 38
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
TSAR-MVS + MP.		84.39 13	86.58 16	81.83 14	88.09 39	86.47 64	85.63 18	73.62 17	90.13 10	79.24 9	89.67 10	82.99 11	77.72 17	81.22 52	80.92 58	86.68 69	94.66 30
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
OpenMVS	67.62 8	74.92 59	73.91 69	76.09 44	90.10 20	90.38 24	78.01 58	66.35 54	66.09 75	62.80 49	46.33 120	64.55 66	71.77 49	79.92 63	80.88 59	87.52 49	89.20 89

CANet_DTU		72.84 67	76.63 58	68.43 91	76.81 97	86.62 62	75.54 74	54.71 157	72.06 58	43.54 124	67.11 45	58.46 93	72.40 44	81.13 55	80.82 60	87.57 48	90.21 78

ET-MVSNet_ETH3D		71.38 76	74.70 67	67.51 97	51.61 203	88.06 48	77.29 65	60.95 102	63.61 82	48.36 106	66.60 48	60.67 82	79.55 9	73.56 121	80.58 61	87.30 55	89.80 82

PGM-MVS		79.42 35	81.84 34	76.60 40	88.38 35	86.69 61	82.97 31	65.75 58	80.39 40	64.94 44	81.95 22	72.11 43	71.41 52	80.45 59	80.55 62	86.18 77	90.76 73

X-MVS		78.16 41	80.55 38	75.38 48	87.99 40	86.27 66	81.05 44	68.98 38	78.33 45	61.07 60	75.25 33	72.27 39	67.52 79	80.03 62	80.52 63	85.66 98	91.20 65

diffmvs		74.32 60	75.42 64	73.04 61	75.60 108	87.27 56	78.20 56	62.96 78	68.66 70	61.89 54	59.79 68	59.84 87	71.80 48	78.30 77	79.87 64	87.80 42	94.23 36

CP-MVS		79.44 32	81.51 35	77.02 38	86.95 43	85.96 70	82.00 34	68.44 42	81.82 34	67.39 39	77.43 29	73.68 34	71.62 50	79.56 66	79.58 65	85.73 91	92.51 53

OPM-MVS		72.74 68	70.93 87	74.85 53	85.30 51	84.34 81	82.82 32	69.79 32	49.96 133	55.39 82	54.09 86	60.14 86	70.04 60	80.38 61	79.43 66	85.74 90	88.20 101

baseline2		71.22 78	73.01 75	69.13 84	75.76 106	86.34 65	71.23 106	62.78 84	62.62 83	52.85 87	57.32 72	54.31 115	63.27 97	79.74 64	79.31 67	88.89 15	91.43 61

GG-mvs-BLEND		54.54 177	77.58 49	27.67 206	0.03 220	90.09 29	77.20 66	0.02 217	66.83 72	0.05 221	59.90 67	73.33 36	0.04 216	78.40 75	79.30 68	88.65 22	95.20 26

HyFIR lowres test		68.39 92	68.28 103	68.52 90	80.85 68	88.11 46	71.08 110	58.09 116	54.87 120	47.80 109	27.55 194	55.80 107	64.97 86	79.11 68	79.14 69	88.31 32	93.35 44

PVSNet_Blended_VisFu		71.76 73	73.54 72	69.69 80	79.01 79	87.16 58	72.05 95	61.80 91	56.46 107	59.66 66	53.88 87	62.48 70	59.08 125	81.17 53	78.90 70	86.53 72	94.74 28

MSLP-MVS++		78.57 38	77.33 51	80.02 22	88.39 34	84.79 76	84.62 22	66.17 56	75.96 53	78.40 10	61.59 60	71.47 45	73.54 37	78.43 74	78.88 71	88.97 14	90.18 79

EIA-MVS		73.48 64	76.05 60	70.47 76	78.12 83	87.21 57	71.78 98	60.63 104	69.66 65	55.56 80	64.86 52	60.69 81	69.53 64	77.35 83	78.59 72	87.22 59	94.01 39

GA-MVS		64.55 119	65.76 121	63.12 124	69.68 137	81.56 101	69.59 119	58.16 115	45.23 154	35.58 166	47.01 115	41.82 147	59.41 121	79.62 65	78.54 73	86.32 74	86.56 112

IS_MVSNet		67.29 102	71.98 78	61.82 136	76.92 96	84.32 83	65.90 144	58.22 114	55.75 113	39.22 146	54.51 82	62.47 71	45.99 171	78.83 71	78.52 74	84.70 122	89.47 86

ACMMP		77.61 44	79.59 42	75.30 49	85.87 49	85.58 71	81.42 39	67.38 48	79.38 44	62.61 50	78.53 26	65.79 62	68.80 72	78.56 73	78.50 75	85.75 88	90.80 70
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 83	69.00 97	71.91 68	85.20 52	85.35 72	77.84 61	61.77 92	58.01 101	55.40 81	41.26 139	58.34 95	61.69 106	81.70 49	78.29 76	89.56 8	80.02 156

GBi-Net		69.21 84	70.40 89	67.81 94	69.49 138	78.65 128	74.54 82	60.97 99	65.32 77	51.06 93	47.37 108	62.05 72	63.43 94	77.49 79	78.22 77	87.37 50	83.73 133

test1		69.21 84	70.40 89	67.81 94	69.49 138	78.65 128	74.54 82	60.97 99	65.32 77	51.06 93	47.37 108	62.05 72	63.43 94	77.49 79	78.22 77	87.37 50	83.73 133

FMVSNet1		63.48 127	63.07 134	63.97 118	65.31 166	76.37 150	71.77 100	57.90 120	43.32 161	45.66 113	35.06 175	49.43 131	58.57 127	77.49 79	78.22 77	84.59 125	81.60 152

test-LLR		68.23 94	71.61 82	64.28 116	71.37 128	81.32 105	63.98 151	61.03 97	58.62 98	42.96 129	52.74 88	61.65 76	57.74 134	75.64 98	78.09 80	88.61 24	93.21 45

TESTMET0.1,1		67.38 101	71.61 82	62.45 132	66.05 161	81.32 105	63.98 151	55.36 149	58.62 98	42.96 129	52.74 88	61.65 76	57.74 134	75.64 98	78.09 80	88.61 24	93.21 45

PCF-MVS	70.85 4	75.73 54	76.55 59	74.78 54	83.67 54	88.04 49	81.47 38	70.62 29	69.24 68	57.52 73	60.59 66	69.18 52	70.65 56	77.11 84	77.65 82	84.75 121	94.01 39
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
thisisatest0530		68.38 93	70.98 86	65.35 105	72.61 121	84.42 79	68.21 127	57.98 117	59.77 94	50.80 96	54.63 80	58.48 92	57.92 131	76.99 86	77.47 83	84.60 124	85.07 124

tttt0517		67.99 96	70.61 88	64.94 108	71.94 126	83.96 85	67.62 131	57.98 117	59.30 96	49.90 101	54.50 83	57.98 99	57.92 131	76.48 89	77.47 83	84.24 131	84.58 127

MVS_111021_LR		74.26 61	75.95 61	72.27 65	79.43 76	85.04 74	72.71 93	65.27 63	70.92 60	63.58 48	69.32 40	60.31 85	69.43 66	77.01 85	77.15 85	83.22 145	91.93 59

Effi-MVS+		70.42 79	71.23 84	69.47 81	78.04 84	85.24 73	75.57 73	58.88 109	59.56 95	48.47 105	52.73 90	54.94 112	69.69 62	78.34 76	77.06 86	86.18 77	90.73 74

Anonymous202405211			66.35 116		78.00 85	84.41 80	74.85 79	63.18 76	51.00 129		31.37 185	53.73 120	69.67 63	76.28 90	76.84 87	83.21 147	90.85 69

AdaColmap		76.23 53	73.55 71	79.35 25	89.38 27	85.00 75	79.99 50	73.04 20	76.60 52	71.17 27	55.18 78	57.99 98	77.87 16	76.82 87	76.82 88	84.67 123	86.45 113

Anonymous20231211		68.44 91	66.37 115	70.86 72	77.58 90	83.49 87	75.15 78	61.89 89	52.54 126	58.50 68	28.89 190	56.78 102	69.29 69	74.96 105	76.61 89	82.73 151	91.36 64

MIMVSNet		57.78 167	59.71 162	55.53 171	54.79 198	77.10 145	63.89 153	45.02 188	46.59 148	36.79 159	28.36 192	40.77 155	45.84 172	74.97 103	76.58 90	86.87 66	73.60 176

DCV-MVSNet		69.13 86	69.07 96	69.21 83	77.65 89	77.52 141	74.68 80	57.85 121	54.92 118	55.34 83	55.74 75	55.56 110	66.35 81	75.05 102	76.56 91	83.35 142	88.13 102

Fast-Effi-MVS+		67.59 97	67.56 107	67.62 96	73.67 116	81.14 107	71.12 109	54.79 156	58.88 97	50.61 98	46.70 118	47.05 136	69.12 70	76.06 94	76.44 92	86.43 73	86.65 111

CostFormer		72.18 70	73.90 70	70.18 78	79.47 75	86.19 69	76.94 67	48.62 176	66.07 76	60.40 65	54.14 85	65.82 61	67.98 74	75.84 96	76.41 93	87.67 45	92.83 51

baseline		72.89 66	74.46 68	71.07 71	75.99 104	87.50 55	74.57 81	60.49 105	70.72 61	57.60 72	60.63 65	60.97 80	70.79 55	75.27 101	76.33 94	86.94 64	89.79 83

LGP-MVS_train		72.02 72	73.18 74	70.67 75	82.13 61	80.26 116	79.58 51	63.04 77	70.09 62	51.98 89	65.06 51	55.62 109	62.49 102	75.97 95	76.32 95	84.80 120	88.93 92

Vis-MVSNet		65.53 112	69.83 93	60.52 142	70.80 134	84.59 77	66.37 143	55.47 148	48.40 140	40.62 142	57.67 71	58.43 94	45.37 174	77.49 79	76.24 96	84.47 127	85.99 119
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
CDS-MVSNet		64.22 121	65.89 120	62.28 134	70.05 135	80.59 112	69.91 118	57.98 117	43.53 160	46.58 111	48.22 102	50.76 128	46.45 168	75.68 97	76.08 97	82.70 152	86.34 115
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
test-mter		64.06 123	69.24 95	58.01 157	59.07 190	77.40 142	59.13 173	48.11 179	55.64 114	39.18 147	51.56 93	58.54 91	55.38 142	73.52 122	76.00 98	87.22 59	92.05 58

thres100view900		67.14 104	66.09 118	68.38 92	77.70 87	83.84 86	74.52 85	66.33 55	49.16 137	43.40 126	43.24 124	41.34 148	62.59 100	79.31 67	75.92 99	85.73 91	89.81 81

FC-MVSNet-train		68.83 89	68.29 102	69.47 81	78.35 81	79.94 117	64.72 145	66.38 53	54.96 117	54.51 85	56.75 73	47.91 135	66.91 80	75.57 100	75.75 100	85.92 84	87.12 107

EG-PatchMatch MVS		58.73 162	58.03 169	59.55 148	72.32 122	80.49 113	63.44 157	55.55 146	32.49 196	38.31 151	28.87 191	37.22 169	42.84 179	74.30 114	75.70 101	84.84 116	77.14 165

PMMVS		70.37 82	75.06 65	64.90 109	71.46 127	81.88 96	64.10 148	55.64 144	71.31 59	46.69 110	70.69 39	58.56 90	69.53 64	79.03 69	75.63 102	81.96 163	88.32 100

thres200		65.58 110	64.74 124	66.56 101	77.52 92	81.61 98	73.44 91	62.95 79	46.23 149	42.45 133	42.76 126	41.18 150	58.12 129	76.24 91	75.59 103	84.89 114	89.58 84

ACMM	66.70 10	70.42 79	68.49 101	72.67 63	82.85 55	77.76 139	77.70 62	64.76 65	64.61 81	60.74 64	49.29 98	53.97 118	65.86 83	74.97 103	75.57 104	84.13 135	83.29 138
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
GeoE		68.96 88	69.32 94	68.54 89	76.61 99	83.12 89	71.78 98	56.87 133	60.21 93	54.86 84	45.95 121	54.79 114	64.27 90	74.59 107	75.54 105	86.84 67	91.01 68

TSAR-MVS + COLMAP		73.09 65	76.86 55	68.71 87	74.97 112	82.49 95	74.51 86	61.83 90	83.16 28	49.31 103	82.22 21	51.62 126	68.94 71	78.76 72	75.52 106	82.67 153	84.23 131

ACMP	68.86 7	72.15 71	72.25 76	72.03 66	80.96 67	80.87 110	77.93 59	64.13 68	69.29 66	60.79 63	64.04 53	53.54 121	63.91 92	73.74 120	75.27 107	84.45 128	88.98 91
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
UGNet		67.57 99	71.69 81	62.76 129	69.88 136	82.58 93	66.43 141	58.64 112	54.71 121	51.87 90	61.74 59	62.01 75	45.46 173	74.78 106	74.99 108	84.24 131	91.02 67
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
Vis-MVSNet (Re-imp)		62.25 137	68.74 99	54.68 174	73.70 115	78.74 127	56.51 179	57.49 126	55.22 115	26.86 187	54.56 81	61.35 78	31.06 189	73.10 125	74.90 109	82.49 156	83.31 137

v2v482		63.68 126	62.85 138	64.65 111	68.01 147	80.46 114	71.90 96	57.60 124	44.26 157	42.82 131	39.80 148	38.62 164	61.56 107	73.06 126	74.86 110	86.03 82	88.90 94

UniMVSNet_NR-MVSNet		62.30 136	63.51 130	60.89 140	69.48 141	77.83 137	64.07 149	63.94 69	50.03 132	31.17 179	44.82 122	41.12 151	51.37 153	71.02 147	74.81 111	85.30 103	84.95 125

test_part1		66.32 105	63.35 131	69.77 79	77.40 93	78.35 132	77.85 60	56.25 137	44.52 156	62.15 53	33.05 179	53.91 119	62.38 104	72.19 137	74.65 112	82.59 154	86.81 109

thres400		65.18 115	64.44 126	66.04 102	76.40 101	82.63 92	71.52 103	64.27 67	44.93 155	40.69 141	41.86 136	40.79 154	58.12 129	77.67 78	74.64 113	85.26 104	88.56 97

tfpn200view9		65.90 109	64.96 122	67.00 100	77.70 87	81.58 100	71.71 101	62.94 81	49.16 137	43.40 126	43.24 124	41.34 148	61.42 108	76.24 91	74.63 114	84.84 116	88.52 98

TranMVSNet+NR-MVSNet		60.38 152	61.30 150	59.30 151	68.34 145	75.57 158	63.38 158	63.78 71	46.74 146	27.73 185	42.56 130	36.84 171	47.66 163	70.36 156	74.59 115	84.91 113	82.46 145

v1144		63.00 131	62.39 142	63.70 121	67.72 150	80.27 115	71.23 106	56.40 134	42.51 162	40.81 140	38.12 156	37.73 165	60.42 115	74.46 109	74.55 116	85.64 99	89.12 90

v10		63.00 131	62.22 143	63.90 120	67.88 149	77.78 138	71.59 102	54.34 158	45.37 153	42.76 132	38.53 151	38.93 162	61.05 111	74.39 111	74.52 117	85.75 88	86.04 117

IterMVS-LS		66.08 108	66.56 114	65.51 104	73.67 116	74.88 160	70.89 113	53.55 163	50.42 131	48.32 107	50.59 95	55.66 108	61.83 105	73.93 116	74.42 118	84.82 119	86.01 118
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
v8		63.44 128	62.58 140	64.43 113	68.28 146	78.07 134	71.82 97	54.85 154	46.70 147	45.20 116	39.40 149	40.91 153	60.54 113	72.85 130	74.39 119	85.92 84	85.76 121

V42		62.86 133	62.97 135	62.74 130	60.84 184	78.99 126	71.46 104	57.13 131	46.85 145	44.28 122	38.87 150	40.73 156	57.63 136	72.60 134	74.14 120	85.09 109	88.63 96

v1192		62.25 137	61.64 147	62.96 125	66.88 155	79.72 119	69.96 117	55.77 142	41.58 167	39.42 144	37.05 161	35.96 178	60.50 114	74.30 114	74.09 121	85.24 105	88.76 95

OMC-MVS		74.03 62	75.82 62	71.95 67	79.56 74	80.98 108	75.35 77	63.21 75	84.48 25	61.83 55	61.54 61	66.89 57	69.41 67	76.60 88	74.07 122	82.34 159	86.15 116

NR-MVSNet		61.08 148	62.09 145	59.90 145	71.96 125	75.87 152	63.60 155	61.96 87	49.31 135	27.95 184	42.76 126	33.85 188	48.82 160	74.35 112	74.05 123	85.13 106	84.45 128

thisisatest0515		59.37 157	60.68 155	57.84 160	64.39 170	75.65 157	58.56 175	53.86 161	41.55 168	42.12 134	40.40 145	39.59 160	47.09 166	71.69 144	73.79 124	81.02 171	82.08 149

IterMVS		61.87 143	63.55 129	59.90 145	67.29 154	72.20 172	67.34 135	48.56 177	47.48 143	37.86 155	47.07 113	48.27 132	54.08 146	72.12 138	73.71 125	84.30 130	83.99 132
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
Baseline_NR-MVSNet		59.47 156	60.28 157	58.54 156	66.69 156	73.90 166	61.63 165	62.90 82	49.15 139	26.87 186	35.18 174	37.62 166	48.20 161	69.67 162	73.61 126	84.92 111	82.82 143

v144192		62.05 141	61.46 149	62.73 131	66.59 159	79.87 118	69.30 121	55.88 140	41.50 169	39.41 145	37.23 159	36.45 173	59.62 119	72.69 133	73.51 127	85.61 100	88.93 92

ACMH	59.42 14	61.59 145	59.22 164	64.36 115	78.92 80	78.26 133	67.65 130	67.48 47	39.81 175	30.98 181	38.25 154	34.59 184	61.37 110	70.55 154	73.47 128	79.74 179	79.59 157
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
v1921920		61.66 144	61.10 152	62.31 133	66.32 160	79.57 121	68.41 126	55.49 147	41.03 170	38.69 149	36.64 167	35.27 181	59.60 120	73.23 124	73.41 129	85.37 102	88.51 99

MDTV_nov1_ep13		65.21 114	67.28 109	62.79 127	70.91 132	81.72 97	69.28 122	49.50 175	58.08 100	43.94 123	50.50 96	56.02 105	58.86 126	70.72 150	73.37 130	84.24 131	80.52 155

v1240		61.09 147	60.55 156	61.72 137	65.92 164	79.28 124	67.16 136	54.91 153	39.79 176	38.10 152	36.08 169	34.64 183	59.15 124	72.86 129	73.36 131	85.10 107	87.84 103

CPTT-MVS		75.43 55	77.13 54	73.44 58	81.43 65	82.55 94	80.96 45	64.35 66	77.95 48	61.39 57	69.20 41	70.94 47	69.38 68	73.89 117	73.32 132	83.14 148	92.06 57

pm-mvs1		59.21 158	59.58 163	58.77 155	67.97 148	77.07 146	64.12 147	57.20 129	34.73 192	36.86 157	35.34 172	40.54 158	43.34 178	74.32 113	73.30 133	83.13 149	81.77 151

UniMVSNet (Re)		60.62 150	62.93 137	57.92 158	67.64 151	77.90 136	61.75 164	61.24 96	49.83 134	29.80 183	42.57 129	40.62 157	43.36 177	70.49 155	73.27 134	83.76 137	85.81 120

CR-MVSNet		62.31 135	64.75 123	59.47 149	68.63 144	71.29 178	67.53 132	43.18 194	55.83 111	41.40 135	41.04 141	55.85 106	57.29 137	72.76 131	73.27 134	78.77 184	83.23 140

PatchT		60.46 151	63.85 128	56.51 168	65.95 163	75.68 156	47.34 193	41.39 201	53.89 124	41.40 135	37.84 157	50.30 130	57.29 137	72.76 131	73.27 134	85.67 95	83.23 140

DU-MVS		60.87 149	61.82 146	59.76 147	66.69 156	75.87 152	64.07 149	61.96 87	49.31 135	31.17 179	42.76 126	36.95 170	51.37 153	69.67 162	73.20 137	83.30 144	84.95 125

tfpnnormal		58.97 159	56.48 174	61.89 135	71.27 130	76.21 151	66.65 140	61.76 93	32.90 195	36.41 161	27.83 193	29.14 201	50.64 157	73.06 126	73.05 138	84.58 126	83.15 142

MSDG		65.57 111	61.57 148	70.24 77	82.02 62	76.47 148	74.46 88	68.73 41	56.52 106	50.33 99	38.47 152	41.10 152	62.42 103	72.12 138	72.94 139	83.47 141	73.37 178

EPP-MVSNet		67.58 98	71.10 85	63.48 122	75.71 107	83.35 88	66.85 137	57.83 122	53.02 125	41.15 138	55.82 74	67.89 55	56.01 140	74.40 110	72.92 140	83.33 143	90.30 77

TransMVSNet (Re)		57.83 165	56.90 172	58.91 154	72.26 123	74.69 163	63.57 156	61.42 95	32.30 197	32.65 175	33.97 177	35.96 178	39.17 185	73.84 119	72.84 141	84.37 129	74.69 171

pmmvs5		59.72 154	60.24 158	59.11 153	62.77 178	77.33 144	63.17 159	54.00 160	40.21 174	37.23 156	40.41 144	35.99 177	51.75 151	72.55 135	72.74 142	85.72 93	82.45 146

v148		62.00 142	61.19 151	62.96 125	67.46 153	79.49 122	67.87 128	57.66 123	42.30 163	45.02 118	38.20 155	38.89 163	54.77 144	69.83 161	72.60 143	84.96 110	87.01 108

TAPA-MVS	67.10 9	71.45 75	73.47 73	69.10 85	77.04 95	80.78 111	73.81 90	62.10 86	80.80 38	51.28 92	60.91 63	63.80 69	67.98 74	74.59 107	72.42 144	82.37 158	80.97 153
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
IterMVS-SCA-FT		60.21 153	62.97 135	57.00 166	66.64 158	71.84 173	67.53 132	46.93 184	47.56 142	36.77 160	46.85 117	48.21 133	52.51 149	70.36 156	72.40 145	71.63 202	83.53 136

UA-Net		64.62 117	68.23 104	60.42 143	77.53 91	81.38 103	60.08 170	57.47 127	47.01 144	44.75 119	60.68 64	71.32 46	41.84 181	73.27 123	72.25 146	80.83 173	71.68 183

EPNet_dtu		66.17 107	70.13 92	61.54 138	81.04 66	77.39 143	68.87 124	62.50 85	69.78 63	33.51 174	63.77 54	56.22 104	37.65 187	72.20 136	72.18 147	85.69 94	79.38 158
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
tpmrst		67.15 103	68.12 105	66.03 103	76.21 102	80.98 108	71.27 105	45.05 187	60.69 91	50.63 97	46.95 116	54.15 117	65.30 84	71.80 142	71.77 148	87.72 43	90.48 75

thres600view7		63.77 125	63.14 133	64.51 112	75.49 109	81.61 98	69.59 119	62.95 79	43.96 159	38.90 148	41.09 140	40.24 159	55.25 143	76.24 91	71.54 149	84.89 114	87.30 106

tpm cat1		67.47 100	67.05 110	67.98 93	76.63 98	81.51 102	74.49 87	47.65 181	61.18 88	61.12 59	42.51 131	53.02 124	64.74 89	70.11 159	71.50 150	83.22 145	89.49 85

pmmvs4		63.14 129	62.46 141	63.94 119	66.03 162	76.40 149	66.82 138	57.60 124	56.74 104	50.26 100	40.81 143	37.51 167	59.26 123	71.75 143	71.48 151	83.68 140	82.53 144

Fast-Effi-MVS+-dtu		63.05 130	64.72 125	61.11 139	71.21 131	76.81 147	70.72 114	43.13 196	52.51 127	35.34 167	46.55 119	46.36 137	61.40 109	71.57 145	71.44 152	84.84 116	87.79 104

FMVSNet5		58.86 160	60.24 158	57.25 164	52.66 202	66.25 190	63.77 154	52.86 168	57.85 102	37.92 154	36.12 168	52.22 125	51.37 153	70.88 149	71.43 153	84.92 111	66.91 194

EPMVS		66.21 106	67.49 108	64.73 110	75.81 105	84.20 84	68.94 123	44.37 191	61.55 87	48.07 108	49.21 100	54.87 113	62.88 98	71.82 141	71.40 154	88.28 33	79.37 159

PatchmatchNet		65.43 113	67.71 106	62.78 128	73.49 118	82.83 91	66.42 142	45.40 186	60.40 92	45.27 115	49.22 99	57.60 100	60.01 117	70.61 151	71.38 155	86.08 81	81.91 150
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
USDC		59.69 155	60.03 160	59.28 152	64.04 171	71.84 173	63.15 160	55.36 149	54.90 119	35.02 168	48.34 101	29.79 200	58.16 128	70.60 152	71.33 156	79.99 177	73.42 177

tpm		64.85 116	66.02 119	63.48 122	74.52 113	78.38 131	70.98 112	44.99 189	51.61 128	43.28 128	47.66 106	53.18 122	60.57 112	70.58 153	71.30 157	86.54 71	89.45 87

test0.0.03 1		57.35 169	59.89 161	54.38 177	71.37 128	73.45 168	52.71 185	61.03 97	46.11 150	26.33 188	41.73 137	44.08 143	29.72 191	71.43 146	70.90 158	85.10 107	71.56 184

SCA		63.90 124	66.67 111	60.66 141	73.75 114	71.78 175	59.87 171	43.66 192	61.13 89	45.03 117	51.64 92	59.45 88	57.92 131	70.96 148	70.80 159	83.71 139	80.92 154

TAMVS		58.86 160	60.91 153	56.47 169	62.38 180	77.57 140	58.97 174	52.98 166	38.76 179	36.17 162	42.26 134	47.94 134	46.45 168	70.23 158	70.79 160	81.86 164	78.82 161

PLC	64.00 12	68.54 90	66.66 112	70.74 74	80.28 73	74.88 160	72.64 94	63.70 72	69.26 67	55.71 78	47.24 111	55.31 111	70.42 57	72.05 140	70.67 161	81.66 166	77.19 164
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
anonymousdsp		54.99 174	57.24 171	52.36 180	53.82 200	71.75 176	51.49 186	48.14 178	33.74 193	33.66 173	38.34 153	36.13 176	47.54 164	64.53 181	70.60 162	79.53 181	85.59 123

CNLPA		71.37 77	70.27 91	72.66 64	80.79 70	81.33 104	71.07 111	65.75 58	82.36 31	64.80 45	42.46 132	56.49 103	72.70 43	73.00 128	70.52 163	80.84 172	85.76 121

v7n		57.04 170	56.64 173	57.52 162	62.85 177	74.75 162	61.76 163	51.80 171	35.58 191	36.02 164	32.33 182	33.61 189	50.16 158	67.73 169	70.34 164	82.51 155	82.12 148

dps		64.08 122	63.22 132	65.08 107	75.27 110	79.65 120	66.68 139	46.63 185	56.94 103	55.67 79	43.96 123	43.63 145	64.00 91	69.50 164	69.82 165	82.25 160	79.02 160

UniMVSNet_ETH3D		57.83 165	56.46 175	59.43 150	63.24 175	73.22 169	67.70 129	55.58 145	36.17 187	36.84 158	32.64 180	35.14 182	51.50 152	65.81 173	69.81 166	81.73 165	82.44 147

ACMH+	60.36 13	61.16 146	58.38 166	64.42 114	77.37 94	74.35 165	68.45 125	62.81 83	45.86 151	38.48 150	35.71 170	37.35 168	59.81 118	67.24 170	69.80 167	79.58 180	78.32 162

CHOSEN 280x420		62.23 139	66.57 113	57.17 165	59.88 187	68.92 184	61.20 167	42.28 198	54.17 122	39.57 143	47.78 105	64.97 64	62.68 99	73.85 118	69.52 168	77.43 188	86.75 110

pmmvs6		54.20 179	53.54 181	54.97 172	63.22 176	72.98 170	60.17 169	52.32 170	26.77 206	34.30 171	23.29 200	36.23 175	40.33 184	68.77 166	68.76 169	79.47 182	78.00 163

LS3D		64.54 120	62.14 144	67.34 99	80.85 68	75.79 154	69.99 116	65.87 57	60.77 90	44.35 121	42.43 133	45.95 139	65.01 85	69.88 160	68.69 170	77.97 187	71.43 185

Anonymous20231206		52.23 184	52.80 186	51.56 182	64.70 169	69.41 182	51.01 187	58.60 113	36.63 184	22.44 194	21.80 203	31.42 195	30.52 190	66.79 171	67.83 171	82.10 162	75.73 167

RPMNet		58.63 163	62.80 139	53.76 179	67.59 152	71.29 178	54.60 182	38.13 206	55.83 111	35.70 165	41.58 138	53.04 123	47.89 162	66.10 172	67.38 172	78.65 186	84.40 129

pmmvs-eth3d		55.20 171	53.95 180	56.65 167	57.34 196	67.77 186	57.54 177	53.74 162	40.93 171	41.09 139	31.19 186	29.10 202	49.07 159	65.54 174	67.28 173	81.14 169	75.81 166

PatchMatch-RL		62.22 140	60.69 154	64.01 117	68.74 143	75.75 155	59.27 172	60.35 106	56.09 110	53.80 86	47.06 114	36.45 173	64.80 88	68.22 167	67.22 174	77.10 189	74.02 173

MVS-HIRNet		53.86 181	53.02 183	54.85 173	60.30 186	72.36 171	44.63 201	42.20 199	39.45 177	43.47 125	21.66 204	34.00 187	55.47 141	65.42 175	67.16 175	83.02 150	71.08 187

testgi		48.51 194	50.53 192	46.16 195	64.78 167	67.15 189	41.54 203	54.81 155	29.12 202	17.03 199	32.07 183	31.98 191	20.15 205	65.26 176	67.00 176	78.67 185	61.10 205

ADS-MVSNet		58.40 164	59.16 165	57.52 162	65.80 165	74.57 164	60.26 168	40.17 205	50.51 130	38.01 153	40.11 147	44.72 141	59.36 122	64.91 177	66.55 177	81.53 167	72.72 181

PEN-MVS		51.04 185	52.94 184	48.82 187	61.45 183	66.00 191	48.68 190	57.20 129	36.87 182	15.36 203	36.98 162	32.72 190	28.77 195	57.63 193	66.37 178	81.44 168	74.00 174

Effi-MVS+-dtu		64.58 118	64.08 127	65.16 106	73.04 120	75.17 159	70.68 115	56.23 138	54.12 123	44.71 120	47.42 107	51.10 127	63.82 93	68.08 168	66.32 179	82.47 157	86.38 114

MDTV_nov1_ep13_2view		54.47 178	54.61 177	54.30 178	60.50 185	73.82 167	57.92 176	43.38 193	39.43 178	32.51 176	33.23 178	34.05 186	47.26 165	62.36 185	66.21 180	84.24 131	73.19 179

WR-MVS		51.02 186	54.56 178	46.90 193	63.84 172	69.23 183	44.78 200	56.38 135	38.19 180	14.19 205	37.38 158	36.82 172	22.39 201	60.14 189	66.20 181	79.81 178	73.95 175

test20.03		47.23 197	48.69 197	45.53 197	63.28 174	64.39 195	41.01 204	56.93 132	29.16 201	15.21 204	23.90 197	30.76 198	17.51 208	64.63 180	65.26 182	79.21 183	62.71 202

DTE-MVSNet		49.82 190	51.92 190	47.37 192	61.75 182	64.38 196	45.89 199	57.33 128	36.11 188	12.79 210	36.87 163	31.93 193	25.73 198	58.01 191	65.22 183	80.75 174	70.93 188

TinyColmap		52.66 183	50.09 194	55.65 170	59.72 188	64.02 198	57.15 178	52.96 167	40.28 173	32.51 176	32.42 181	20.97 211	56.65 139	63.95 183	65.15 184	74.91 195	63.87 199

CMPMVS	43.63 17	57.67 168	55.43 176	60.28 144	72.01 124	79.00 125	62.77 161	53.23 165	41.77 166	45.42 114	30.74 187	39.03 161	53.01 148	64.81 179	64.65 185	75.26 194	68.03 192
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
CVMVSNet		54.92 176	58.16 167	51.13 184	62.61 179	68.44 185	55.45 181	52.38 169	42.28 164	21.45 195	47.10 112	46.10 138	37.96 186	64.42 182	63.81 186	76.92 190	75.01 170

pmnet_mix02		53.92 180	53.30 182	54.65 176	61.89 181	71.33 177	54.54 183	54.17 159	40.38 172	34.65 169	34.76 176	30.68 199	40.44 183	60.97 187	63.71 187	82.19 161	71.24 186

CP-MVSNet		50.57 187	52.60 188	48.21 190	58.77 192	65.82 192	48.17 191	56.29 136	37.41 181	16.59 200	37.14 160	31.95 192	29.21 192	56.60 196	63.71 187	80.22 175	75.56 168

PS-CasMVS		50.17 188	52.02 189	48.02 191	58.60 193	65.54 193	48.04 192	56.19 139	36.42 186	16.42 202	35.68 171	31.33 196	28.85 194	56.42 198	63.54 189	80.01 176	75.18 169

LTVRE_ROB	47.26 16	49.41 192	49.91 195	48.82 187	64.76 168	69.79 181	49.05 189	47.12 183	20.36 213	16.52 201	36.65 166	26.96 204	50.76 156	60.47 188	63.16 190	64.73 205	72.00 182
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
WR-MVS_H		49.62 191	52.63 187	46.11 196	58.80 191	67.58 187	46.14 198	54.94 151	36.51 185	13.63 208	36.75 165	35.67 180	22.10 202	56.43 197	62.76 191	81.06 170	72.73 180

COLMAP_ROB	51.17 15	55.13 172	52.90 185	57.73 161	73.47 119	67.21 188	62.13 162	55.82 141	47.83 141	34.39 170	31.60 184	34.24 185	44.90 175	63.88 184	62.52 192	75.67 192	63.02 201
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
TDRefinement		52.70 182	51.02 191	54.66 175	57.41 195	65.06 194	61.47 166	54.94 151	44.03 158	33.93 172	30.13 189	27.57 203	46.17 170	61.86 186	62.48 193	74.01 198	66.06 195

FC-MVSNet-test		47.24 196	54.37 179	38.93 202	59.49 189	58.25 205	34.48 209	53.36 164	45.66 152	6.66 215	50.62 94	42.02 146	16.62 209	58.39 190	61.21 194	62.99 206	64.40 198

MIMVSNet1		40.84 202	43.46 200	37.79 203	32.14 211	58.92 204	39.24 206	50.83 173	27.00 205	11.29 212	16.76 210	26.53 206	17.75 207	57.14 195	61.12 195	75.46 193	56.78 206

N_pmnet		47.67 195	47.00 199	48.45 189	54.72 199	62.78 199	46.95 195	51.25 172	36.01 189	26.09 189	26.59 196	25.93 208	35.50 188	55.67 200	59.01 196	76.22 191	63.04 200

pmmvs3		41.86 201	42.29 203	41.36 199	39.80 209	52.66 208	38.93 207	35.85 210	23.40 210	20.22 197	19.30 205	20.84 212	40.56 182	55.98 199	58.79 197	72.80 200	65.03 197

SixPastTwentyTwo		49.11 193	49.22 196	48.99 186	58.54 194	64.14 197	47.18 194	47.75 180	31.15 199	24.42 190	41.01 142	26.55 205	44.04 176	54.76 201	58.70 198	71.99 201	68.21 190

MDA-MVSNet-bldmvs		44.15 199	42.27 204	46.34 194	38.34 210	62.31 200	46.28 196	55.74 143	29.83 200	20.98 196	27.11 195	16.45 216	41.98 180	41.11 208	57.47 199	74.72 196	61.65 204

ambc			42.30 202		50.36 204	49.51 209	35.47 208		32.04 198	23.53 191	17.36 207	8.95 218	29.06 193	64.88 178	56.26 200	61.29 207	67.12 193

new-patchmatchnet		42.21 200	42.97 201	41.33 200	53.05 201	59.89 202	39.38 205	49.61 174	28.26 204	12.10 211	22.17 202	21.54 210	19.22 206	50.96 203	56.04 201	74.61 197	61.92 203

EU-MVSNet		44.84 198	47.85 198	41.32 201	49.26 205	56.59 206	43.07 202	47.64 182	33.03 194	13.82 206	36.78 164	30.99 197	24.37 199	53.80 202	55.57 202	69.78 203	68.21 190

PM-MVS		50.11 189	50.38 193	49.80 185	47.23 208	62.08 201	50.91 188	44.84 190	41.90 165	36.10 163	35.22 173	26.05 207	46.83 167	57.64 192	55.42 203	72.90 199	74.32 172

RPSCF		55.07 173	58.06 168	51.57 181	48.87 206	58.95 203	53.68 184	41.26 203	62.42 84	45.88 112	54.38 84	54.26 116	53.75 147	57.15 194	53.53 204	66.01 204	65.75 196

Gipuma		24.91 207	24.61 209	25.26 207	31.47 212	21.59 214	18.06 214	37.53 207	25.43 208	10.03 213	4.18 216	4.25 220	14.85 210	43.20 206	47.03 205	39.62 212	26.55 213
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
new_pmnet		33.19 204	35.52 206	30.47 205	27.55 215	45.31 211	29.29 211	30.92 211	29.00 203	9.88 214	18.77 206	17.64 215	26.77 197	44.07 204	45.98 206	58.41 209	47.87 207

FPMVS		39.11 203	36.39 205	42.28 198	55.97 197	45.94 210	46.23 197	41.57 200	35.73 190	22.61 192	23.46 199	19.82 213	28.32 196	43.57 205	40.67 207	58.96 208	45.54 208

PMVS	27.44 18	32.08 205	29.07 208	35.60 204	48.33 207	24.79 213	26.97 212	41.34 202	20.45 212	22.50 193	17.11 209	18.64 214	20.44 204	41.99 207	38.06 208	54.02 210	42.44 209
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
PMMVS2		20.45 208	22.31 210	18.27 211	20.52 216	26.73 212	14.85 216	28.43 213	13.69 214	0.79 220	10.35 212	9.10 217	3.83 215	27.64 211	32.87 209	41.17 211	35.81 210

MVE	15.98 19	14.37 211	16.36 211	12.04 213	7.72 218	20.24 216	5.90 220	29.05 212	8.28 217	3.92 217	4.72 215	2.42 221	9.57 213	18.89 213	31.46 210	16.07 218	28.53 212
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
test_method		28.15 206	34.48 207	20.76 208	6.76 219	21.18 215	21.03 213	18.41 214	36.77 183	17.52 198	15.67 211	31.63 194	24.05 200	41.03 209	26.69 211	36.82 213	68.38 189

tmp_tt				16.09 212	13.07 217	8.12 220	13.61 217	2.08 216	55.09 116	30.10 182	40.26 146	22.83 209	5.35 214	29.91 210	25.25 212	32.33 214

E-PMN		15.08 209	11.65 212	19.08 209	28.73 213	12.31 218	6.95 219	36.87 209	10.71 216	3.63 218	5.13 213	2.22 223	13.81 212	11.34 214	18.50 213	24.49 215	21.32 214

EMVS		14.40 210	10.71 213	18.70 210	28.15 214	12.09 219	7.06 218	36.89 208	11.00 215	3.56 219	4.95 214	2.27 222	13.91 211	10.13 215	16.06 214	22.63 216	18.51 215

testmvs		0.05 212	0.08 214	0.01 214	0.00 221	0.01 221	0.03 222	0.01 218	0.05 218	0.00 222	0.14 218	0.01 224	0.03 218	0.05 216	0.05 215	0.01 219	0.24 217

test123		0.05 212	0.08 214	0.01 214	0.00 221	0.01 221	0.01 223	0.00 219	0.05 218	0.00 222	0.16 217	0.00 225	0.04 216	0.02 217	0.05 215	0.00 220	0.26 216

uanet_test		0.00 214	0.00 216	0.00 216	0.00 221	0.00 223	0.00 224	0.00 219	0.00 220	0.00 222	0.00 219	0.00 225	0.00 219	0.00 218	0.00 217	0.00 220	0.00 218

sosnet-low-res		0.00 214	0.00 216	0.00 216	0.00 221	0.00 223	0.00 224	0.00 219	0.00 220	0.00 222	0.00 219	0.00 225	0.00 219	0.00 218	0.00 217	0.00 220	0.00 218

sosnet		0.00 214	0.00 216	0.00 216	0.00 221	0.00 223	0.00 224	0.00 219	0.00 220	0.00 222	0.00 219	0.00 225	0.00 219	0.00 218	0.00 217	0.00 220	0.00 218

RE-MVS-def										31.47 178

9.14												84.47 6

SR-MVS					86.33 47			67.54 46				80.78 19

our_test_3					63.32 173	71.07 180	55.90 180

MTAPA										78.32 11		79.42 24

MTMP										76.04 16		76.65 29

Patchmatch-RL test							2.17 221

XVS					82.43 56	86.27 66	75.70 69			61.07 60		72.27 39				85.67 95

X-MVStestdata					82.43 56	86.27 66	75.70 69			61.07 60		72.27 39				85.67 95

abl_6				79.06 29	89.68 24	92.14 12	77.70 62	69.68 33	86.87 19	71.88 25	74.29 35	80.06 22	76.56 22			88.84 16	95.82 13

mPP-MVS					86.96 42							70.61 49

NP-MVS									81.60 36

Patchmtry						78.06 135	67.53 132	43.18 194		41.40 135

DeepMVS_CX						19.81 217	17.01 215	10.02 215	23.61 209	5.85 216	17.21 208	8.03 219	21.13 203	22.60 212		21.42 217	30.01 211

High-resolution multi-view benchmark