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.

MSP-MVS		77.54 1	84.41 1	69.54 5	79.93 3	86.08 3	77.20 7	60.31 4	88.62 1	62.54 1	86.67 3	83.77 1	58.04 31	75.84 5	75.69 6	79.21 19	94.17 1

SF-MVS		76.41 2	80.45 4	71.69 1	82.90 1	86.54 1	82.08 1	64.58 1	81.67 9	59.82 3	86.26 4	77.90 6	61.11 13	71.81 25	70.75 32	79.63 12	88.22 25

DVP-MVS		76.38 3	82.99 2	68.68 6	71.93 17	78.65 22	77.61 4	55.44 17	88.04 2	60.25 2	92.24 1	77.08 8	69.84 1	75.48 6	75.69 6	76.99 52	93.75 2

DPE-MVS		75.74 4	82.82 3	67.49 10	77.07 6	82.01 6	77.05 8	57.70 10	86.55 4	55.44 14	90.50 2	82.52 2	60.33 18	72.99 13	72.98 14	77.33 45	92.19 6

DPM-MVS		74.63 5	78.53 9	70.07 3	76.10 8	82.56 5	79.30 3	59.89 6	80.49 12	57.75 9	66.98 24	76.16 11	65.95 2	79.35 1	78.47 1	81.45 5	85.71 45

APDe-MVS		74.59 6	80.23 5	68.01 9	76.51 7	80.20 14	77.39 5	58.18 8	85.31 5	56.84 11	84.89 6	76.08 12	60.66 16	71.85 24	71.76 19	78.47 28	91.49 9

MCST-MVS		74.06 7	77.71 12	69.79 4	78.95 4	81.99 7	76.33 9	62.16 3	75.89 20	52.96 24	64.37 30	73.30 20	65.66 3	77.49 2	77.43 3	82.67 1	93.51 3

CNVR-MVS		73.87 8	78.60 8	68.35 8	73.32 12	81.97 8	76.19 10	59.29 7	80.12 13	56.70 12	67.09 23	76.48 9	64.26 5	75.88 4	75.75 5	80.32 7	92.93 4

SMA-MVS		73.31 9	79.53 6	66.05 12	71.25 19	80.13 15	74.99 11	56.09 13	84.14 6	54.48 17	73.74 15	80.23 3	61.43 10	74.96 7	74.09 10	78.08 35	89.42 14

xxxxxxxxxxxxxcwj		73.17 10	74.44 19	71.69 1	82.90 1	86.54 1	82.08 1	64.58 1	81.67 9	59.82 3	86.26 4	35.82 141	61.11 13	71.81 25	70.75 32	79.63 12	88.22 25

CSCG		72.98 11	76.86 14	68.46 7	78.23 5	81.74 9	77.26 6	60.00 5	75.61 23	59.06 5	62.72 32	77.42 7	56.63 44	74.24 9	77.18 4	79.56 14	89.13 18

HPM-MVS++		72.44 12	78.73 7	65.11 13	71.88 18	77.31 31	71.98 19	55.67 15	83.11 8	53.59 21	75.90 11	78.49 5	61.00 15	73.99 10	73.31 13	76.55 55	88.97 19

APD-MVS		71.86 13	77.91 11	64.80 15	70.39 24	75.69 41	74.02 13	56.14 12	83.59 7	52.92 25	84.67 7	73.46 19	59.30 25	69.47 40	69.66 40	76.02 61	88.84 20
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
ACMMP_NAP		71.50 14	77.27 13	64.77 16	69.64 26	79.26 16	73.53 14	54.73 23	79.32 15	54.23 18	74.81 12	74.61 16	59.40 24	73.00 12	72.17 17	77.10 51	87.72 30

NCCC		71.36 15	75.44 16	66.60 11	72.46 15	79.18 18	74.16 12	57.83 9	76.93 18	54.19 19	63.47 31	71.08 25	61.30 12	73.56 11	73.70 11	79.69 11	90.19 11

train_agg		70.74 16	76.53 15	63.98 18	70.33 25	75.16 45	72.33 18	55.78 14	75.74 21	50.41 33	80.08 10	73.15 21	57.75 36	71.96 23	70.94 29	77.25 49	88.69 22

TSAR-MVS + MP.		70.28 17	75.09 17	64.66 17	69.34 28	64.61 123	72.60 17	56.29 11	80.73 11	58.36 7	84.56 8	75.22 14	55.37 50	69.11 47	69.45 42	75.97 63	81.97 74
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
DeepPCF-MVS	62.48 1	70.07 18	78.36 10	60.39 42	62.38 58	76.96 34	65.54 54	52.23 32	87.46 3	49.07 34	74.05 14	76.19 10	59.01 27	72.79 17	71.61 21	74.13 103	89.49 13

SteuartSystems-ACMMP		69.78 19	74.76 18	63.98 18	73.45 11	78.56 23	73.13 16	55.24 20	70.68 33	48.93 36	70.43 19	69.10 27	54.00 55	72.78 19	72.98 14	79.14 20	88.74 21
Skip Steuart: Steuart Systems R&D Blog.
HFP-MVS		68.75 20	72.84 21	63.98 18	68.87 32	75.09 46	71.87 20	51.22 37	73.50 27	58.17 8	68.05 22	68.67 28	57.79 35	70.49 33	69.23 44	75.98 62	84.84 52

SD-MVS		68.30 21	72.58 23	63.31 24	69.24 29	67.85 96	70.81 25	53.65 28	79.64 14	58.52 6	74.31 13	75.37 13	53.52 61	65.63 71	63.56 100	74.13 103	81.73 78
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
zzz-MVS		67.78 22	72.46 24	62.33 29	66.09 38	74.21 51	70.05 27	51.54 35	77.27 16	54.61 16	60.30 40	71.51 24	56.73 42	69.19 45	68.63 53	74.96 83	86.11 42

DELS-MVS		67.36 23	70.34 37	63.89 21	69.12 30	81.55 10	70.82 24	55.02 21	53.38 73	48.83 37	56.45 46	59.35 54	60.05 22	74.93 8	74.78 8	79.51 15	91.95 7
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
MP-MVS		67.34 24	73.08 20	60.64 39	66.20 37	76.62 36	69.22 31	50.92 39	70.07 34	48.81 38	69.66 20	70.12 26	53.68 58	68.41 52	69.13 46	74.98 82	87.53 32
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
DeepC-MVS	60.65 2	67.33 25	71.52 31	62.44 27	59.79 77	74.84 48	68.89 32	55.56 16	73.91 26	53.50 22	55.00 51	65.63 33	60.08 21	71.99 22	71.33 25	76.85 53	87.94 29
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
HQP-MVS		67.22 26	72.08 26	61.56 34	66.76 35	73.58 58	71.41 21	52.98 30	69.92 36	43.85 56	70.58 18	58.75 56	56.76 41	72.90 15	71.88 18	77.57 41	86.94 36

CANet		67.21 27	71.83 28	61.83 30	64.51 44	79.25 17	66.72 47	48.73 55	68.49 41	50.63 32	61.40 36	66.47 31	61.44 9	69.31 44	69.90 36	78.94 23	88.00 27

CDPH-MVS		67.03 28	71.64 29	61.65 33	69.10 31	76.84 35	71.35 23	55.42 18	67.02 44	42.83 61	65.27 28	64.60 37	53.16 64	69.70 39	71.40 23	78.02 37	86.67 37

MAR-MVS		66.85 29	69.81 38	63.39 23	73.56 10	80.51 13	69.87 28	51.51 36	67.78 43	46.44 43	51.09 64	61.60 49	60.38 17	72.67 20	73.61 12	78.59 25	81.44 82
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
DeepC-MVS_fast	60.18 3	66.84 30	70.69 35	62.36 28	62.76 52	73.21 61	67.96 35	52.31 31	72.26 30	51.03 27	56.50 45	64.26 38	63.37 6	71.64 27	70.85 30	76.70 54	86.10 43
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
TSAR-MVS + GP.		66.77 31	72.21 25	60.44 41	61.23 64	70.00 80	64.26 59	47.79 67	72.98 28	56.32 13	71.35 17	72.33 22	55.68 49	65.49 72	66.66 67	77.35 43	86.62 38

MVS_0304		66.31 32	71.61 30	60.14 43	62.59 56	78.98 20	67.13 43	45.75 91	64.35 49	45.23 50	60.69 38	67.67 30	61.73 8	71.09 29	71.03 27	78.41 31	87.44 33

ACMMPR		66.20 33	71.51 32	60.00 45	65.34 42	74.04 53	69.39 30	50.92 39	71.97 31	46.04 45	66.79 25	65.68 32	53.07 65	68.93 49	69.12 47	75.21 76	84.05 58

3Dnovator	58.39 4	65.97 34	66.85 50	64.94 14	73.72 9	79.03 19	67.73 38	54.25 24	61.52 52	52.79 26	42.27 90	60.73 52	62.01 7	71.29 28	71.75 20	79.12 21	81.34 85

TSAR-MVS + ACMM		65.95 35	72.83 22	57.93 54	69.35 27	65.85 115	73.36 15	39.84 141	76.00 19	48.69 39	82.54 9	75.03 15	49.38 91	65.33 74	63.42 102	66.94 161	81.67 79

canonicalmvs		65.55 36	70.75 34	59.49 48	62.11 60	78.26 26	66.52 48	43.82 111	71.54 32	47.84 41	61.30 37	61.68 47	58.48 30	67.56 59	69.67 39	78.16 34	85.25 50

QAPM		65.47 37	67.82 43	62.72 26	72.56 13	81.17 12	67.43 41	55.38 19	56.07 65	43.29 59	43.60 85	65.38 35	59.10 26	72.20 21	70.76 31	78.56 26	85.59 48

PGM-MVS		65.35 38	70.43 36	59.43 49	65.78 40	73.75 55	69.41 29	48.18 63	68.80 40	45.37 49	65.88 27	64.04 39	52.68 70	68.94 48	68.68 52	75.18 77	82.93 65

PHI-MVS		65.17 39	72.07 27	57.11 64	63.02 51	77.35 30	67.04 44	48.14 65	68.03 42	37.56 87	66.00 26	65.39 34	53.19 63	70.68 30	70.57 35	73.72 110	86.46 41

CLD-MVS		64.69 40	67.25 45	61.69 32	68.22 34	78.33 24	63.09 62	47.59 70	69.64 37	53.98 20	54.87 52	53.94 70	57.87 32	72.79 17	71.34 24	79.40 17	69.87 151
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
MVS_111021_HR		64.66 41	67.11 48	61.80 31	71.04 20	77.91 27	62.75 65	54.78 22	51.43 76	47.54 42	53.77 55	54.85 67	56.84 39	70.59 31	71.50 22	77.86 38	89.70 12

EPNet		64.39 42	70.93 33	56.77 66	60.58 71	75.77 38	59.28 85	50.58 43	69.93 35	40.73 76	68.59 21	61.60 49	53.72 56	68.65 50	68.07 55	75.75 68	83.87 60
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
CP-MVS		64.37 43	69.48 39	58.39 51	62.21 59	71.81 72	67.27 42	49.51 49	69.40 39	45.76 47	60.41 39	64.96 36	51.84 72	67.33 63	67.57 61	73.78 109	84.89 51

casdiffmvs		63.87 44	67.08 49	60.12 44	60.90 67	78.29 25	67.91 36	48.01 66	55.89 67	44.97 51	50.45 66	56.94 61	59.54 23	70.17 36	69.81 37	79.41 16	87.99 28

MVS_Test		63.75 45	67.24 46	59.68 47	60.01 73	76.99 33	68.13 34	45.17 93	57.45 60	43.74 57	53.07 58	56.16 65	61.33 11	70.27 34	71.11 26	79.72 10	85.63 47

X-MVS		63.53 46	68.62 40	57.60 58	64.77 43	73.06 62	65.82 52	50.53 44	65.77 46	42.02 69	58.20 42	63.42 42	47.83 102	68.25 56	68.50 54	74.61 93	83.16 64

ACMMP		63.27 47	67.85 42	57.93 54	62.64 55	72.30 69	68.23 33	48.77 54	66.50 45	43.05 60	62.07 33	57.84 60	49.98 83	66.58 67	66.46 72	74.93 84	83.17 62
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
ETV-MVS		62.88 48	68.18 41	56.70 67	58.47 85	74.89 47	60.26 78	43.96 108	58.27 59	42.37 67	61.47 35	56.56 62	57.80 33	68.00 57	68.74 50	77.34 44	89.33 17

CS-MVS		62.84 49	67.51 44	57.38 61	60.49 72	75.48 44	62.36 67	43.72 113	53.40 72	41.28 74	57.32 44	58.61 58	57.80 33	69.85 38	69.49 41	78.64 24	88.45 23

AdaColmap		62.79 50	62.63 65	62.98 25	70.82 21	72.90 65	67.84 37	54.09 26	65.14 47	50.71 30	41.78 92	47.64 97	60.17 20	67.41 62	66.83 65	74.28 98	76.69 108

3Dnovator+	55.76 7	62.70 51	65.10 57	59.90 46	65.89 39	72.15 70	62.94 64	49.82 48	62.77 51	49.06 35	43.62 84	61.47 51	58.60 29	68.51 51	66.75 66	73.08 124	80.40 93

OpenMVS	55.62 8	62.57 52	63.76 62	61.19 36	72.13 16	78.84 21	64.42 57	50.51 45	56.44 62	45.67 48	36.88 118	56.51 63	56.66 43	68.28 55	68.96 48	77.73 40	80.44 92

PVSNet_BlendedMVS		62.53 53	66.37 52	58.05 52	58.17 86	75.70 39	61.30 71	48.67 58	58.67 56	50.93 28	55.43 49	49.39 86	53.01 66	69.46 41	66.55 69	76.24 59	89.39 15

PVSNet_Blended		62.53 53	66.37 52	58.05 52	58.17 86	75.70 39	61.30 71	48.67 58	58.67 56	50.93 28	55.43 49	49.39 86	53.01 66	69.46 41	66.55 69	76.24 59	89.39 15

MVSTER		62.51 55	67.22 47	57.02 65	55.05 107	69.23 88	63.02 63	46.88 80	61.11 54	43.95 55	59.20 41	58.86 55	56.80 40	69.13 46	70.98 28	76.41 57	82.04 71

CHOSEN 1792x2688		62.48 56	64.06 61	60.64 39	72.50 14	84.18 4	62.43 66	53.77 27	47.90 88	39.85 80	25.15 178	44.76 110	53.72 56	77.29 3	77.61 2	81.60 4	91.53 8

CostFormer		62.45 57	65.68 55	58.67 50	63.29 48	77.65 28	67.62 39	38.42 150	54.04 70	46.00 46	48.27 74	57.89 59	56.97 38	67.03 65	67.79 60	79.74 9	87.09 35

PCF-MVS	55.99 6	62.31 58	66.60 51	57.32 62	59.12 84	73.68 57	67.53 40	48.71 56	61.35 53	42.83 61	51.33 63	63.48 41	53.48 62	65.64 70	64.87 84	72.22 129	85.83 44
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
diffmvs		62.30 59	66.05 54	57.92 56	57.08 91	75.60 43	66.90 45	47.06 78	55.45 69	43.37 58	53.45 57	55.60 66	57.21 37	66.57 68	68.00 57	75.89 66	87.70 31

DI_MVS_plusplus_trai		61.86 60	65.26 56	57.90 57	57.93 89	74.51 50	66.30 49	46.49 86	49.96 80	41.62 72	42.69 88	61.77 46	58.74 28	70.25 35	69.32 43	76.31 58	88.30 24

MSLP-MVS++		61.81 61	62.19 70	61.37 35	68.33 33	63.08 137	70.75 26	38.89 147	63.96 50	57.51 10	48.59 72	61.66 48	53.67 59	62.04 110	59.92 145	79.03 22	76.08 111

OPM-MVS		61.59 62	62.30 69	60.76 38	66.53 36	73.35 60	71.41 21	54.18 25	40.82 116	41.57 73	45.70 80	54.84 68	54.43 54	69.92 37	69.19 45	76.45 56	82.25 68

MS-PatchMatch		61.41 63	61.88 73	60.85 37	70.57 23	75.98 37	66.29 50	46.91 79	50.56 78	48.28 40	36.30 121	51.64 74	50.95 78	72.89 16	70.65 34	82.13 3	75.17 117

EIA-MVS		60.56 64	64.29 60	56.20 71	59.14 83	72.68 66	59.55 83	43.56 115	51.78 75	41.01 75	55.47 48	51.93 73	55.87 46	65.01 77	66.57 68	78.06 36	86.60 40

ACMP	56.21 5	59.78 65	61.81 75	57.41 60	61.15 65	68.88 90	65.98 51	48.85 53	58.56 58	44.19 54	48.89 70	46.31 103	48.56 96	63.61 93	64.49 92	75.75 68	81.91 75
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
LGP-MVS_train		59.69 66	62.59 66	56.31 69	61.94 61	68.15 95	66.90 45	48.15 64	59.75 55	38.47 83	50.38 67	48.34 94	46.87 107	65.39 73	64.93 83	75.51 72	81.21 87

Effi-MVS+		59.63 67	61.78 76	57.12 63	61.56 62	71.63 73	63.61 60	47.59 70	47.18 89	37.79 84	45.29 81	49.93 83	56.27 45	67.45 60	67.06 63	75.91 64	83.93 59

CPTT-MVS		59.54 68	64.47 59	53.79 81	54.99 109	67.63 99	65.48 55	44.59 100	64.81 48	37.74 85	51.55 61	59.90 53	49.77 87	61.83 112	61.26 130	70.18 143	84.31 57

baseline2		59.20 69	61.72 77	56.27 70	59.61 79	74.12 52	58.65 88	49.42 50	48.10 86	40.12 79	49.10 69	44.15 112	51.24 75	66.65 66	67.88 59	78.56 26	82.06 70

baseline		58.65 70	61.99 71	54.75 76	54.70 111	71.85 71	60.20 79	43.91 109	55.99 66	40.13 78	53.50 56	50.91 80	55.76 47	61.29 120	61.73 122	73.83 107	78.68 101

PVSNet_Blended_VisFu		58.56 71	62.33 68	54.16 78	56.90 92	73.92 54	57.72 91	46.16 89	44.23 96	42.73 64	46.26 77	51.06 79	46.28 110	67.99 58	65.38 78	75.18 77	87.44 33

ACMM	53.73 9	57.91 72	58.27 91	57.49 59	63.10 49	66.45 109	65.65 53	49.02 52	53.69 71	42.67 65	36.41 120	46.07 106	50.38 81	64.74 80	64.63 89	74.14 102	75.91 112
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
CANet_DTU		57.87 73	63.63 63	51.15 95	52.18 117	70.20 79	58.14 90	37.32 157	56.49 61	31.06 118	57.38 43	50.05 82	53.67 59	64.98 78	65.04 82	74.57 94	81.29 86

ET-MVSNet_ETH3D		57.84 74	61.91 72	53.09 83	32.91 197	74.53 49	63.51 61	46.80 82	46.52 91	36.14 92	56.00 47	46.20 104	64.41 4	60.75 128	66.99 64	74.79 85	82.35 66

tpm cat1		57.41 75	58.26 92	56.42 68	60.80 69	72.56 67	64.35 58	38.43 149	49.18 84	46.36 44	36.69 119	43.50 115	54.47 52	61.39 118	62.64 110	74.11 105	81.81 76

IB-MVS	53.15 10	57.33 76	59.02 83	55.37 73	60.83 68	77.11 32	54.51 115	50.10 47	43.22 100	42.82 63	40.50 97	37.61 133	44.67 122	59.27 142	69.81 37	79.29 18	85.59 48
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
tpmrst		57.23 77	59.08 82	55.06 74	59.91 75	70.65 77	60.71 74	35.38 168	47.91 87	42.58 66	39.78 101	45.45 108	54.44 53	62.19 107	62.82 107	77.37 42	84.73 53

baseline1		57.21 78	60.53 79	53.33 82	62.50 57	69.86 82	57.33 95	50.59 42	43.39 99	30.00 124	48.60 71	51.09 78	42.36 134	69.38 43	68.03 56	77.20 50	73.39 124

HyFIR lowres test		57.12 79	59.11 81	54.80 75	61.55 63	77.55 29	59.02 86	45.00 95	41.84 113	33.93 104	22.44 185	49.16 89	51.02 77	68.39 53	68.71 51	78.26 33	85.70 46

MVS_111021_LR		57.06 80	60.60 78	52.93 84	56.25 96	65.14 121	55.16 113	41.21 133	52.32 74	44.89 52	53.92 54	49.27 88	52.16 71	61.46 116	60.54 138	67.92 154	81.53 81

DCV-MVSNet		56.80 81	58.96 84	54.28 77	59.96 74	66.74 107	60.37 77	44.87 97	41.01 115	36.81 90	47.57 75	47.87 96	48.23 99	64.41 83	65.17 80	75.45 73	79.95 95

Anonymous20231211		56.40 82	57.00 101	55.70 72	59.78 78	72.49 68	61.29 73	46.83 81	40.50 117	40.46 77	22.12 187	49.73 84	51.07 76	64.39 84	65.30 79	74.74 87	84.44 56

PMMVS		55.74 83	62.68 64	47.64 124	44.34 166	65.58 119	47.22 150	37.96 153	56.43 63	34.11 102	61.51 34	47.41 98	54.55 51	65.88 69	62.49 114	67.67 156	79.48 96

Fast-Effi-MVS+		55.73 84	58.26 92	52.76 85	54.33 112	68.19 94	57.05 96	34.66 170	46.92 90	38.96 82	40.53 96	41.55 124	55.69 48	65.31 75	65.99 73	75.90 65	79.34 97

FC-MVSNet-train		55.68 85	57.00 101	54.13 79	63.37 46	66.16 111	46.77 153	52.14 33	42.36 107	37.67 86	48.50 73	41.42 126	51.28 74	61.58 115	63.22 104	73.56 112	75.76 114

FMVSNet3		55.66 86	59.68 80	50.96 97	50.59 131	66.49 108	57.57 92	46.61 83	49.30 81	28.77 129	39.61 102	51.42 75	43.85 127	68.29 54	68.80 49	78.35 32	73.86 119

OMC-MVS		55.48 87	61.85 74	48.04 123	41.55 173	60.32 154	56.80 100	31.78 190	75.67 22	42.30 68	51.52 62	54.15 69	49.91 85	60.28 133	57.59 152	65.91 164	73.42 122

tpm		54.94 88	57.86 97	51.54 93	59.48 81	67.04 103	58.34 89	34.60 172	41.93 112	34.41 99	42.40 89	47.14 99	49.07 94	61.46 116	61.67 126	73.31 119	83.39 61

GBi-Net		54.66 89	58.42 89	50.26 105	49.36 140	65.81 116	56.80 100	46.61 83	49.30 81	28.77 129	39.61 102	51.42 75	42.71 130	64.25 86	65.54 75	77.32 46	73.03 127

test1		54.66 89	58.42 89	50.26 105	49.36 140	65.81 116	56.80 100	46.61 83	49.30 81	28.77 129	39.61 102	51.42 75	42.71 130	64.25 86	65.54 75	77.32 46	73.03 127

test-LLR		54.62 91	58.66 87	49.89 111	51.68 123	65.89 113	47.88 144	46.35 87	42.51 104	29.84 125	41.41 93	48.87 90	45.20 115	62.91 101	64.43 93	78.43 29	84.62 54

TSAR-MVS + COLMAP		54.37 92	62.43 67	44.98 138	34.33 193	58.94 161	54.11 118	34.15 181	74.06 25	34.57 98	71.63 16	42.03 123	47.88 101	61.26 121	57.33 155	64.83 167	71.74 137

EPMVS		54.07 93	56.06 107	51.75 92	56.74 94	70.80 75	55.32 111	34.20 178	46.46 92	36.59 91	40.38 99	42.55 118	49.77 87	61.25 122	60.90 134	77.86 38	70.08 148

v2v482		54.00 94	55.12 113	52.69 87	51.73 122	69.42 87	60.65 75	45.09 94	34.56 148	33.73 107	35.29 124	35.36 144	49.92 84	64.05 90	65.16 81	75.00 81	81.98 73

CNLPA		54.00 94	57.08 100	50.40 104	49.83 137	61.75 145	53.47 121	37.27 158	74.55 24	44.85 53	33.58 136	45.42 109	52.94 69	58.89 144	53.66 173	64.06 170	71.68 138

FMVSNet2		53.94 96	57.29 98	50.03 108	49.36 140	65.81 116	56.80 100	45.95 90	43.13 101	28.04 133	35.68 122	48.18 95	42.71 130	67.23 64	67.95 58	77.32 46	73.03 127

v8		53.77 97	54.82 117	52.54 88	52.12 118	66.95 106	60.56 76	43.23 121	37.17 137	35.35 94	34.96 127	37.50 135	49.51 90	63.67 92	64.59 90	74.48 95	78.91 100

GA-MVS		53.77 97	56.41 106	50.70 99	51.63 125	69.96 81	57.55 93	44.39 101	34.31 149	27.15 135	40.99 95	36.40 139	47.65 104	67.45 60	67.16 62	75.83 67	78.60 102

Effi-MVS+-dtu		53.63 99	54.85 116	52.20 90	59.32 82	61.33 148	56.42 106	40.24 139	43.84 98	34.22 101	39.49 106	46.18 105	53.00 68	58.72 148	57.49 154	69.99 146	76.91 107

thisisatest0530		53.61 100	57.22 99	49.40 116	51.30 127	68.22 93	52.72 129	43.34 119	42.72 103	35.31 95	43.57 86	44.14 113	44.37 125	63.00 99	64.86 85	69.34 149	74.00 118

v1144		53.47 101	54.65 118	52.10 91	51.93 120	69.81 83	59.32 84	44.77 99	33.21 155	32.52 110	33.55 137	34.34 152	49.29 92	64.58 81	64.81 87	74.74 87	82.27 67

v10		53.44 102	54.40 119	52.31 89	52.08 119	66.99 104	59.68 82	43.41 116	35.90 143	34.30 100	33.98 134	35.56 142	50.10 82	64.39 84	64.67 88	74.32 96	79.30 98

PatchmatchNet		53.37 103	55.62 111	50.75 98	55.93 103	70.54 78	51.39 134	36.41 161	44.85 94	37.26 88	39.40 108	42.54 119	47.83 102	60.29 132	60.88 136	75.69 70	70.87 142
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
IterMVS-LS		53.36 104	55.65 110	50.68 101	55.34 105	59.04 159	55.00 114	39.98 140	38.72 125	33.22 108	44.52 83	47.05 100	49.63 89	61.82 113	61.77 121	70.92 138	76.61 110
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
TESTMET0.1,1		53.30 105	58.66 87	47.04 127	44.94 160	65.89 113	47.88 144	35.95 164	42.51 104	29.84 125	41.41 93	48.87 90	45.20 115	62.91 101	64.43 93	78.43 29	84.62 54

tttt0517		53.05 106	56.73 105	48.76 119	50.35 133	67.51 100	51.96 133	43.34 119	42.00 111	33.88 105	43.19 87	43.49 116	44.37 125	62.58 106	64.86 85	68.67 151	73.46 121

MDTV_nov1_ep13		52.99 107	55.59 112	49.95 110	54.08 113	70.69 76	56.47 105	38.42 150	42.78 102	30.19 123	39.56 105	43.31 117	45.78 112	60.07 137	62.11 118	74.74 87	70.62 143

EPP-MVSNet		52.91 108	58.91 85	45.91 132	54.99 109	68.84 91	49.27 140	42.71 128	37.53 131	20.20 161	46.09 78	56.19 64	36.90 145	61.37 119	60.90 134	71.41 133	81.41 83

dps		52.84 109	52.92 130	52.74 86	59.89 76	69.49 86	54.47 116	37.38 156	42.49 106	39.53 81	35.33 123	32.71 157	51.83 73	60.45 129	61.12 131	73.33 118	68.86 157

v1192		52.69 110	53.86 122	51.31 94	51.22 128	69.76 84	57.37 94	44.39 101	32.21 158	31.39 117	32.41 144	32.44 160	49.19 93	64.25 86	64.17 95	74.31 97	81.81 76

V42		52.63 111	55.08 114	49.76 113	44.93 161	67.49 102	60.19 80	42.13 131	37.21 136	34.08 103	34.57 130	37.30 136	47.29 105	63.48 95	64.15 96	69.96 147	81.38 84

MSDG		52.58 112	51.40 143	53.95 80	65.48 41	64.31 131	61.44 70	44.02 106	44.17 97	32.92 109	30.40 156	31.81 164	46.35 109	62.13 108	62.55 112	73.49 114	64.41 165

Fast-Effi-MVS+-dtu		52.47 113	55.89 108	48.48 121	56.25 96	65.07 122	58.75 87	23.79 201	41.27 114	27.07 137	37.95 113	41.34 127	50.85 79	62.90 103	62.34 116	74.17 101	80.37 94

v144192		52.43 114	53.63 124	51.03 96	51.06 129	69.60 85	56.94 98	44.84 98	32.15 159	30.88 119	32.45 143	32.71 157	48.36 97	62.98 100	63.52 101	74.10 106	82.02 72

thres100view900		52.33 115	53.91 121	50.48 103	56.10 98	67.79 97	56.18 108	49.18 51	35.86 145	25.22 143	34.74 128	34.10 153	42.41 133	64.45 82	62.62 111	73.81 108	77.85 103

v1921920		51.95 116	53.19 126	50.51 102	50.82 130	69.14 89	55.45 110	44.34 105	31.53 163	30.53 121	31.96 146	31.67 165	48.31 98	63.12 97	63.28 103	73.59 111	81.60 80

v148		51.72 117	53.15 127	50.05 107	50.15 135	67.51 100	56.98 97	42.85 126	32.60 157	32.41 112	33.88 135	34.71 149	44.45 123	61.06 123	63.00 106	73.45 115	79.24 99

TAPA-MVS	47.92 11	51.66 118	57.88 96	44.40 141	36.46 188	58.42 164	53.82 120	30.83 191	69.51 38	34.97 97	46.90 76	49.67 85	46.99 106	58.00 151	54.64 170	63.33 176	68.00 159
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
IS_MVSNet		51.53 119	57.98 95	44.01 145	55.96 102	66.16 111	47.65 146	42.84 127	39.82 120	19.09 169	44.97 82	50.28 81	27.20 176	63.43 96	63.84 97	71.33 135	77.33 105

v1240		51.42 120	52.66 132	49.97 109	50.31 134	68.70 92	54.05 119	43.85 110	30.78 167	30.22 122	31.43 149	31.03 172	47.98 100	62.62 105	63.16 105	73.40 116	80.93 89

pmmvs4		51.28 121	52.50 134	49.85 112	49.54 139	63.02 138	52.83 128	43.41 116	44.65 95	35.71 93	34.38 131	32.25 161	45.14 118	60.21 136	60.03 142	72.44 128	72.98 130

Vis-MVSNet		51.13 122	58.04 94	43.06 151	47.68 147	67.71 98	49.10 141	39.09 146	37.75 129	22.57 153	51.03 65	48.78 92	32.42 162	62.12 109	61.80 120	67.49 158	77.12 106
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
UGNet		51.04 123	58.79 86	42.00 157	40.59 175	65.32 120	46.65 155	39.26 144	39.90 119	27.30 134	54.12 53	52.03 72	30.93 166	59.85 139	59.62 147	67.23 160	80.70 90
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
tfpn200view9		50.91 124	52.45 135	49.11 118	56.10 98	64.53 126	53.06 125	47.31 75	35.86 145	25.22 143	34.74 128	34.10 153	41.08 136	60.84 125	61.37 128	71.90 132	75.70 115

SCA		50.88 125	53.70 123	47.59 125	55.99 100	55.81 172	43.14 167	33.45 184	45.16 93	37.14 89	41.83 91	43.82 114	44.43 124	60.37 130	60.02 143	71.38 134	68.90 156

gg-mvs-nofinetune		50.82 126	55.83 109	44.97 139	60.63 70	75.69 41	53.40 122	34.48 174	20.05 200	6.93 194	18.27 193	52.70 71	33.57 153	70.50 32	72.93 16	80.84 6	80.68 91

thres200		50.76 127	52.52 133	48.70 120	55.98 101	64.60 124	55.29 112	47.34 73	33.91 152	24.36 146	34.33 132	33.90 155	37.27 143	60.84 125	62.41 115	71.99 130	77.63 104

thres400		50.39 128	52.22 136	48.26 122	55.02 108	66.32 110	52.97 126	48.33 62	32.68 156	22.94 151	33.21 139	33.38 156	37.27 143	62.74 104	61.38 127	73.04 125	75.81 113

EG-PatchMatch MVS		50.23 129	50.89 146	49.47 114	59.54 80	70.88 74	52.46 130	44.01 107	26.22 187	31.91 113	24.97 179	31.45 168	33.48 155	64.79 79	66.51 71	75.40 74	71.39 140

IterMVS		50.23 129	53.27 125	46.68 128	47.59 149	60.58 152	53.10 124	36.62 160	36.07 141	25.89 140	39.42 107	40.05 130	43.65 128	60.22 135	61.35 129	73.23 120	75.23 116
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
FMVSNet1		50.14 131	52.78 131	47.06 126	45.56 157	63.56 134	54.22 117	43.74 112	34.10 151	25.37 142	29.79 162	42.06 122	38.70 139	64.25 86	65.54 75	74.75 86	70.18 147

ACMH	47.82 13	50.10 132	49.60 152	50.69 100	63.36 47	66.99 104	56.83 99	52.13 34	31.06 166	17.74 174	28.22 168	26.24 188	45.17 117	60.88 124	63.80 98	68.91 150	70.00 150
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
EPNet_dtu		49.85 133	56.99 103	41.52 160	52.79 115	57.06 166	41.44 171	43.13 122	56.13 64	19.24 168	52.11 59	48.38 93	22.14 183	58.19 150	58.38 150	70.35 141	68.71 158
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
LS3D		49.59 134	49.75 151	49.40 116	55.88 104	59.86 156	56.31 107	45.33 92	48.57 85	28.32 132	31.54 148	36.81 138	46.27 111	57.17 156	55.88 165	64.29 169	58.42 182

UniMVSNet_NR-MVSNet		49.56 135	53.04 128	45.49 135	51.59 126	64.42 130	46.97 151	51.01 38	37.87 127	16.42 175	39.87 100	34.91 148	33.43 157	59.59 140	62.70 108	73.52 113	71.94 133

CDS-MVSNet		49.25 136	53.97 120	43.75 147	47.53 150	64.53 126	48.59 142	42.27 130	33.77 153	26.64 138	40.46 98	42.26 121	30.01 169	61.77 114	61.71 123	67.48 159	73.28 126
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
PLC	44.22 14	49.14 137	51.75 139	46.10 131	42.78 171	55.60 175	53.11 123	34.46 175	55.69 68	32.47 111	34.16 133	41.45 125	48.91 95	57.13 157	54.09 171	64.84 166	64.10 166
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
ACMH+	47.85 12	49.13 138	48.86 158	49.44 115	56.75 93	62.01 144	56.62 104	47.55 72	37.49 132	23.98 147	26.68 173	29.46 179	43.12 129	57.45 155	58.85 149	68.62 152	70.05 149

NR-MVSNet		48.84 139	51.76 138	45.44 136	57.66 90	60.64 150	47.39 147	47.63 68	37.26 133	13.31 179	37.31 115	29.64 178	33.53 154	63.52 94	62.09 119	73.10 123	71.89 136

CR-MVSNet		48.82 140	51.85 137	45.29 137	46.74 152	55.95 170	52.06 131	34.21 176	42.17 108	31.74 114	32.92 141	42.53 120	45.00 119	58.80 145	61.11 132	61.99 181	69.47 152

thres600view7		48.44 141	50.23 149	46.35 130	54.05 114	64.60 124	50.18 137	47.34 73	31.73 162	20.74 159	32.28 145	32.62 159	33.79 152	60.84 125	56.11 163	71.99 130	73.40 123

test-mter		48.31 142	55.04 115	40.45 163	34.12 194	59.02 160	41.77 170	28.05 195	38.43 126	22.67 152	39.35 109	44.40 111	41.88 135	60.30 131	61.68 125	74.20 99	82.12 69

PatchT		48.11 143	51.27 145	44.43 140	50.13 136	61.58 146	33.59 184	32.92 186	40.38 118	31.74 114	30.60 155	36.93 137	45.00 119	58.80 145	61.11 132	73.19 121	69.47 152

TranMVSNet+NR-MVSNet		48.06 144	51.36 144	44.21 143	50.38 132	62.09 143	47.28 148	50.88 41	36.11 140	13.25 180	37.51 114	31.60 167	30.70 167	59.34 141	62.53 113	72.81 126	70.31 145

TransMVSNet (Re)		47.46 145	48.94 157	45.74 134	57.96 88	64.29 132	48.26 143	48.47 61	26.33 186	19.33 166	29.45 165	31.28 171	25.31 180	63.05 98	62.70 108	75.10 80	65.47 163

DU-MVS		47.33 146	50.86 147	43.20 150	44.43 164	60.64 150	46.97 151	47.63 68	37.26 133	16.42 175	37.31 115	31.39 169	33.43 157	57.53 153	59.98 144	70.35 141	71.94 133

v7n		47.22 147	48.38 159	45.87 133	48.20 146	63.58 133	50.69 135	40.93 137	26.60 185	26.44 139	26.52 174	29.65 177	38.19 141	58.22 149	60.23 141	70.79 139	73.83 120

UA-Net		47.19 148	53.02 129	40.38 164	55.31 106	60.02 155	38.41 177	38.68 148	36.42 139	22.47 155	51.95 60	58.72 57	25.62 179	54.11 169	53.40 174	61.79 182	56.51 185

Baseline_NR-MVSNet		47.14 149	50.83 148	42.84 153	44.43 164	63.31 136	44.50 163	50.36 46	37.71 130	11.25 185	30.84 152	32.09 162	30.96 165	57.53 153	63.73 99	75.53 71	70.60 144

pmmvs5		47.02 150	50.02 150	43.51 149	43.48 169	62.65 140	47.24 149	37.78 155	30.59 168	24.80 145	35.26 125	30.43 173	34.36 150	59.05 143	60.28 140	73.40 116	71.92 135

UniMVSNet (Re)		46.89 151	51.65 141	41.34 161	45.60 156	62.71 139	44.05 164	47.10 77	37.24 135	13.55 178	36.90 117	34.54 151	26.76 177	57.56 152	59.90 146	70.98 137	72.69 131

thisisatest0515		46.88 152	49.57 153	43.74 148	45.33 159	60.46 153	46.19 157	41.06 136	30.34 169	29.73 127	32.50 142	31.63 166	35.43 148	58.75 147	61.71 123	64.70 168	71.59 139

tfpnnormal		46.61 153	46.82 166	46.37 129	52.70 116	62.31 141	50.39 136	47.17 76	25.74 189	21.80 156	23.13 183	24.15 196	33.45 156	60.28 133	60.77 137	72.70 127	71.39 140

pm-mvs1		46.14 154	49.34 156	42.41 154	48.93 143	62.22 142	44.98 161	42.68 129	27.66 179	20.76 158	29.88 161	34.96 147	26.41 178	60.03 138	60.42 139	70.70 140	70.20 146

IterMVS-SCA-FT		45.87 155	51.55 142	39.24 167	46.22 153	59.43 157	52.89 127	31.93 187	36.01 142	23.68 148	38.86 110	39.88 132	39.05 138	56.25 162	58.17 151	41.70 202	72.25 132

MIMVSNet		45.62 156	49.56 154	41.02 162	38.17 179	64.43 129	49.48 139	35.43 167	36.53 138	20.06 163	22.58 184	35.16 146	28.75 174	61.97 111	62.20 117	74.20 99	64.07 167

gm-plane-assit		45.41 157	48.03 161	42.34 155	56.49 95	40.48 200	24.54 204	34.15 181	14.44 206	6.59 195	17.82 194	35.32 145	49.82 86	72.93 14	74.11 9	82.47 2	81.12 88

ADS-MVSNet		45.39 158	46.42 167	44.19 144	48.74 145	57.52 165	43.91 165	31.93 187	35.89 144	27.11 136	30.12 157	32.06 163	45.30 113	53.13 175	55.19 167	68.15 153	61.07 175

GG-mvs-BLEND		44.87 159	64.59 58	21.86 200	0.01 215	73.70 56	55.99 109	0.01 212	50.70 77	0.01 215	49.18 68	63.61 40	0.01 211	63.83 91	64.50 91	75.13 79	86.62 38

pmmvs-eth3d		44.67 160	45.27 172	43.98 146	42.56 172	55.72 174	44.97 162	40.81 138	31.96 161	29.13 128	26.09 176	25.27 193	36.69 146	55.13 166	56.62 160	69.68 148	66.12 162

MDTV_nov1_ep13_2view		44.44 161	45.75 170	42.91 152	46.13 154	63.43 135	46.53 156	34.20 178	29.08 174	19.95 164	26.23 175	27.89 183	35.88 147	53.36 174	56.43 161	74.74 87	63.86 168

CMPMVS	33.64 16	44.39 162	46.41 168	42.03 156	44.21 167	56.50 168	46.73 154	26.48 200	34.20 150	35.14 96	24.22 180	34.64 150	40.52 137	56.50 161	56.07 164	59.12 186	62.74 171
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
Vis-MVSNet (Re-imp)		44.31 163	51.67 140	35.72 177	51.82 121	55.24 176	34.57 183	41.63 132	39.10 123	8.84 192	45.93 79	46.63 102	14.45 193	54.09 170	57.03 157	63.00 177	63.65 169

TAMVS		44.27 164	49.35 155	38.35 171	44.74 162	61.04 149	39.07 175	31.82 189	29.95 171	18.34 172	33.55 137	39.94 131	30.01 169	56.85 159	57.58 153	66.13 163	66.54 160

MVS-HIRNet		43.98 165	43.63 176	44.39 142	47.66 148	59.31 158	32.66 190	33.88 183	30.15 170	33.75 106	16.82 199	28.39 182	45.25 114	53.92 173	55.00 169	73.16 122	61.80 172

UniMVSNet_ETH3D		43.97 166	46.01 169	41.59 158	38.31 178	56.20 169	49.69 138	38.18 152	28.18 175	19.88 165	27.82 170	30.20 174	33.41 159	54.18 168	56.30 162	70.05 145	69.17 154

RPMNet		43.70 167	48.17 160	38.48 170	45.52 158	55.95 170	37.66 178	26.63 199	42.17 108	25.47 141	29.59 164	37.61 133	33.87 151	50.85 179	52.02 178	61.75 183	69.00 155

PatchMatch-RL		43.37 168	44.93 173	41.56 159	37.94 180	51.70 178	40.02 173	35.75 165	39.04 124	30.71 120	35.14 126	27.43 185	46.58 108	51.99 176	50.55 182	58.38 188	58.64 180

FMVSNet5		43.29 169	47.07 164	38.87 168	30.46 199	50.99 180	45.87 158	37.19 159	42.17 108	19.32 167	26.77 172	40.51 128	30.26 168	56.82 160	55.81 166	70.10 144	56.46 186

test0.0.03 1		43.07 170	46.95 165	38.54 169	51.68 123	58.77 162	35.28 179	46.35 87	32.05 160	12.44 181	28.53 167	35.52 143	14.40 194	57.12 158	56.93 158	71.11 136	59.69 176

anonymousdsp		43.03 171	47.19 163	38.18 172	36.00 190	56.92 167	38.44 176	34.56 173	24.22 191	22.53 154	29.69 163	29.92 175	35.21 149	53.96 172	58.98 148	62.32 180	76.66 109

USDC		42.80 172	45.57 171	39.58 165	34.55 192	51.13 179	42.61 168	36.21 162	39.59 121	23.65 149	33.13 140	20.87 200	37.86 142	55.35 165	57.16 156	62.61 178	61.75 173

CHOSEN 280x420		42.39 173	47.40 162	36.54 175	33.56 195	39.66 203	40.67 172	26.88 198	34.66 147	18.03 173	30.09 158	45.59 107	44.82 121	54.46 167	54.00 172	55.28 195	73.32 125

pmmvs6		41.90 174	44.01 175	39.43 166	44.45 163	58.77 162	41.92 169	39.22 145	21.74 193	19.08 170	17.40 197	31.33 170	24.28 182	55.94 163	56.67 159	67.60 157	66.24 161

Anonymous20231206		40.63 175	43.29 177	37.53 173	48.88 144	55.81 172	34.99 180	44.98 96	28.16 176	10.16 189	17.26 198	27.50 184	18.28 187	54.00 171	55.07 168	67.85 155	65.23 164

CVMVSNet		38.91 176	44.49 174	32.40 186	34.57 191	47.20 191	34.81 181	34.20 178	31.45 164	8.95 191	38.86 110	36.38 140	24.30 181	47.77 183	46.94 194	57.59 190	62.85 170

COLMAP_ROB	34.79 15	38.65 177	40.72 180	36.23 176	36.41 189	49.22 187	45.51 160	27.60 197	37.81 128	20.54 160	23.37 182	24.25 195	28.11 175	51.02 178	48.55 185	59.22 185	50.82 196
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
PEN-MVS		38.23 178	41.72 179	34.15 179	40.56 176	50.07 183	33.17 187	44.35 104	27.64 181	5.54 201	30.84 152	26.67 186	14.99 191	45.64 186	52.38 177	66.29 162	58.83 179

WR-MVS		37.61 179	42.15 178	32.31 188	43.64 168	51.85 177	29.39 196	43.35 118	27.65 180	4.40 203	29.90 160	29.80 176	10.46 197	46.73 185	51.98 179	62.60 179	57.16 183

TinyColmap		37.18 180	37.37 193	36.95 174	31.17 198	45.21 194	39.71 174	34.65 171	29.83 172	20.20 161	18.54 192	13.72 208	38.27 140	50.33 180	51.57 180	57.71 189	52.42 193

CP-MVSNet		37.09 181	40.62 181	32.99 181	37.56 182	48.25 188	32.75 188	43.05 123	27.88 178	5.93 197	31.27 150	25.82 191	15.09 189	43.37 193	48.82 183	63.54 174	58.90 177

DTE-MVSNet		36.91 182	40.44 182	32.79 184	40.74 174	47.55 190	30.71 194	44.39 101	27.03 183	4.32 204	30.88 151	25.99 189	12.73 195	45.58 187	50.80 181	63.86 171	55.23 189

PS-CasMVS		36.84 183	40.23 185	32.89 182	37.44 183	48.09 189	32.68 189	42.97 125	27.36 182	5.89 198	30.08 159	25.48 192	14.96 192	43.28 194	48.71 184	63.39 175	58.63 181

WR-MVS_H		36.29 184	40.35 184	31.55 190	37.80 181	49.94 185	30.57 195	41.11 135	26.90 184	4.14 205	30.72 154	28.85 180	10.45 198	42.47 195	47.99 189	65.24 165	55.54 187

SixPastTwentyTwo		36.11 185	37.80 189	34.13 180	37.13 186	46.72 192	34.58 182	34.96 169	21.20 196	11.66 182	29.15 166	19.88 201	29.77 171	44.93 188	48.34 186	56.67 192	54.41 191

test20.03		36.00 186	38.92 186	32.60 185	45.92 155	50.99 180	28.05 200	43.69 114	21.62 194	6.03 196	17.61 196	25.91 190	8.34 204	51.26 177	52.60 176	63.58 172	52.46 192

TDRefinement		35.76 187	38.23 187	32.88 183	19.09 208	46.04 193	43.29 166	29.49 192	33.49 154	19.04 171	22.29 186	17.82 203	29.69 173	48.60 182	47.24 192	56.65 193	52.12 194

LTVRE_ROB	32.83 17	35.10 188	37.46 190	32.35 187	43.12 170	49.99 184	28.52 198	33.23 185	12.73 207	8.18 193	27.71 171	21.34 199	32.64 161	46.92 184	48.11 187	48.41 199	55.45 188
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
PM-MVS		34.96 189	38.17 188	31.22 191	22.78 203	40.82 199	33.56 185	23.61 202	29.16 173	21.43 157	28.00 169	21.43 198	31.90 163	44.33 191	42.12 197	54.07 197	61.34 174

testgi		34.51 190	37.42 191	31.12 192	47.37 151	50.34 182	24.38 205	41.21 133	20.32 198	5.64 200	20.56 188	26.55 187	8.06 205	49.28 181	52.65 175	60.05 184	42.23 202

MDA-MVSNet-bldmvs		34.31 191	34.11 197	34.54 178	24.73 201	49.66 186	33.42 186	43.03 124	21.59 195	11.10 186	19.81 190	12.68 209	31.41 164	35.59 200	48.05 188	63.56 173	51.39 195

N_pmnet		34.09 192	35.74 195	32.17 189	37.25 185	43.17 197	32.26 192	35.57 166	26.22 187	10.60 188	20.44 189	19.38 202	20.20 185	44.59 190	47.00 193	57.13 191	49.35 199

RPSCF		33.61 193	40.43 183	25.65 196	16.00 210	32.41 205	31.73 193	13.33 209	50.13 79	23.12 150	31.56 147	40.09 129	32.73 160	41.14 199	37.05 200	36.99 205	50.63 197

EU-MVSNet		33.00 194	36.49 194	28.92 193	33.10 196	42.86 198	29.32 197	35.99 163	22.94 192	5.83 199	25.29 177	24.43 194	15.21 188	41.22 198	41.65 199	54.08 196	57.01 184

pmmvs3		31.22 195	33.62 198	28.43 194	22.82 202	40.26 202	26.40 201	22.05 204	16.89 204	10.99 187	14.72 201	16.26 204	29.70 172	44.82 189	47.39 191	58.61 187	54.98 190

FC-MVSNet-test		30.97 196	37.38 192	23.49 199	37.42 184	33.68 204	19.43 207	39.27 143	31.37 165	1.67 211	38.56 112	28.85 180	6.06 208	41.40 196	43.80 196	37.10 204	44.03 201

new-patchmatchnet		30.47 197	32.80 200	27.75 195	36.81 187	43.98 195	24.85 203	39.29 142	20.52 197	4.06 206	15.94 200	16.05 205	9.57 199	41.32 197	42.05 198	51.94 198	49.74 198

MIMVSNet1		29.60 198	33.37 199	25.20 198	19.52 206	43.94 196	26.29 202	37.92 154	19.95 201	3.79 207	12.64 205	21.99 197	7.70 206	43.83 192	46.32 195	55.97 194	44.92 200

FPMVS		26.87 199	28.19 201	25.32 197	27.09 200	29.49 206	32.28 191	17.79 206	28.09 177	11.33 183	19.38 191	14.69 206	20.88 184	35.11 201	32.82 202	42.56 201	37.75 203

PMVS	18.18 18	21.95 200	22.85 202	20.90 201	21.92 204	14.78 208	19.95 206	17.31 207	15.69 205	11.32 184	13.70 202	13.91 207	15.02 190	34.92 202	31.72 203	39.85 203	35.20 204
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
new_pmnet		19.10 201	22.71 203	14.89 203	10.93 212	24.08 207	14.22 208	13.94 208	18.68 202	2.93 208	12.84 204	11.27 210	11.94 196	30.57 204	30.58 204	35.38 206	30.93 205

Gipuma		17.16 202	17.83 204	16.36 202	18.76 209	12.15 211	11.97 209	27.78 196	17.94 203	4.86 202	2.53 211	2.73 215	8.90 202	34.32 203	36.09 201	25.92 207	19.06 207
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
PMMVS2		12.25 203	14.17 205	10.00 206	11.39 211	14.35 209	8.21 210	19.29 205	9.31 208	0.19 214	7.38 207	6.19 213	1.10 210	19.26 205	21.13 206	19.85 208	21.56 206

E-PMN		10.66 204	8.30 207	13.42 204	19.91 205	7.87 212	4.30 213	29.47 193	8.37 211	1.70 210	3.67 208	1.29 218	9.12 201	8.98 209	13.59 207	16.03 209	14.30 210

EMVS		10.15 205	7.67 208	13.05 205	19.22 207	7.77 213	4.48 211	29.34 194	8.65 210	1.67 211	3.55 209	1.36 217	9.15 200	8.15 210	11.79 209	14.44 210	12.43 211

MVE	10.35 19	9.76 206	11.08 206	8.22 207	4.43 213	13.04 210	3.36 214	23.57 203	5.74 212	1.76 209	3.09 210	1.75 216	6.78 207	12.78 207	23.04 205	9.44 211	18.09 208
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
testmvs		0.01 207	0.01 209	0.00 209	0.00 216	0.00 216	0.00 217	0.00 213	0.01 213	0.00 216	0.02 212	0.00 219	0.00 213	0.01 211	0.01 210	0.00 214	0.03 212

test123		0.01 207	0.01 209	0.00 209	0.00 216	0.00 216	0.00 217	0.00 213	0.01 213	0.00 216	0.02 212	0.00 219	0.01 211	0.00 212	0.01 210	0.00 214	0.03 212

uanet_test		0.00 209	0.00 211	0.00 209	0.00 216	0.00 216	0.00 217	0.00 213	0.00 215	0.00 216	0.00 214	0.00 219	0.00 213	0.00 212	0.00 212	0.00 214	0.00 214

sosnet-low-res		0.00 209	0.00 211	0.00 209	0.00 216	0.00 216	0.00 217	0.00 213	0.00 215	0.00 216	0.00 214	0.00 219	0.00 213	0.00 212	0.00 212	0.00 214	0.00 214

sosnet		0.00 209	0.00 211	0.00 209	0.00 216	0.00 216	0.00 217	0.00 213	0.00 215	0.00 216	0.00 214	0.00 219	0.00 213	0.00 212	0.00 212	0.00 214	0.00 214

9.14												80.07 4

SR-MVS					63.74 45			48.51 60				73.80 17

Anonymous202405211			56.81 104		60.91 66	73.48 59	59.82 81	48.68 57	39.26 122		24.00 181	46.77 101	50.73 80	65.28 76	65.72 74	75.37 75	83.17 62

our_test_3					49.68 138	61.50 147	45.84 159

test_part1																	92.54 5

ambc			35.52 196		38.36 177	40.40 201	28.38 199		25.20 190	14.87 177	13.22 203	7.54 212	19.34 186	55.63 164	47.79 190	47.91 200	58.89 178

MTAPA										54.82 15		71.98 23

MTMP										50.64 31		68.31 29

Patchmatch-RL test							0.69 216

tmp_tt				4.41 208	2.56 214	1.81 215	2.61 215	0.27 211	20.12 199	9.81 190	17.69 195	9.04 211	1.96 209	12.88 206	12.11 208	9.23 212

XVS					62.70 53	73.06 62	61.80 68			42.02 69		63.42 42				74.68 91

X-MVStestdata					62.70 53	73.06 62	61.80 68			42.02 69		63.42 42				74.68 91

abl_6				63.79 22	70.80 22	81.22 11	65.26 56	53.25 29	77.02 17	53.02 23	65.14 29	73.74 18	60.30 19			80.13 8	90.27 10

mPP-MVS					63.08 50							62.34 45

NP-MVS									72.62 29

Patchmtry						64.49 128	52.06 131	34.21 176		31.74 114

DeepMVS_CX						5.87 214	4.32 212	1.74 210	9.04 209	1.30 213	7.97 206	3.16 214	8.56 203	9.74 208		6.30 213	14.51 209

High-resolution multi-view benchmark