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.

MTAPA										65.14 1		80.20 13

MPTG		74.25 15	77.97 19	69.91 10	73.43 18	74.06 38	79.69 13	56.44 13	80.74 9	64.98 2	68.72 25	79.98 14	62.92 9	78.24 13	77.77 11	81.99 24	86.30 17

HFP-MVS		74.87 10	78.86 15	70.21 6	73.99 15	77.91 13	80.36 10	56.63 11	78.41 15	64.27 3	74.54 15	77.75 22	62.96 8	78.70 8	77.82 9	83.02 7	86.91 15

CSCG		74.68 11	79.22 11	69.40 13	75.69 8	80.01 4	79.12 18	52.83 35	79.34 13	63.99 4	70.49 20	82.02 7	60.35 26	77.48 21	77.22 15	84.38 1	87.97 10

ACMMP_Plus		76.15 4	81.17 4	70.30 5	74.09 14	79.47 5	81.59 5	57.09 9	81.38 6	63.89 5	79.02 9	80.48 12	62.24 13	80.05 3	79.12 3	82.94 9	88.64 4

HPM-MVS++		76.01 5	80.47 7	70.81 3	76.60 4	74.96 30	80.18 11	58.36 3	81.96 5	63.50 6	78.80 10	82.53 6	64.40 4	78.74 7	78.84 4	81.81 26	87.46 12

CNVR-MVS		75.62 7	79.91 9	70.61 4	75.76 6	78.82 9	81.66 4	57.12 8	79.77 12	63.04 7	70.69 19	81.15 9	62.99 7	80.23 2	79.54 2	83.11 6	89.16 3

SD-MVS		74.43 12	78.87 13	69.26 15	74.39 13	73.70 40	79.06 19	55.24 21	81.04 7	62.71 8	80.18 8	82.61 5	61.70 17	75.43 35	73.92 38	82.44 16	85.22 27

TSAR-MVS + MP.		75.22 9	80.06 8	69.56 12	74.61 12	72.74 44	80.59 7	55.70 19	80.80 8	62.65 9	86.25 3	82.92 4	62.07 15	76.89 24	75.66 26	81.77 28	85.19 28

MTMP										62.63 10		78.04 20

ESAPD		78.19 1	83.74 1	71.72 1	79.01 1	81.38 1	83.23 2	58.63 2	83.92 4	62.44 11	87.06 2	85.82 1	64.54 3	79.39 4	77.99 7	82.44 16	90.61 1

NCCC		74.27 14	77.83 20	70.13 8	75.70 7	77.41 17	80.51 9	57.09 9	78.25 16	62.28 12	65.54 32	78.26 19	62.18 14	79.13 5	78.51 5	83.01 8	87.68 11

DeepC-MVS	66.32 2	73.85 18	78.10 18	68.90 17	67.92 43	79.31 6	78.16 23	59.28 1	78.24 17	61.13 13	67.36 31	76.10 26	63.40 6	79.11 6	78.41 6	83.52 4	88.16 8
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
APDe-MVS		77.58 2	82.93 2	71.35 2	77.86 2	80.55 2	83.38 1	57.61 6	85.57 1	61.11 14	86.10 4	82.98 3	64.76 2	78.29 11	76.78 18	83.40 5	90.20 2

HSP-MVS		76.78 3	82.44 3	70.19 7	75.26 10	80.22 3	80.59 7	57.85 5	84.79 3	60.84 15	88.54 1	83.43 2	66.24 1	78.21 14	76.47 20	80.34 37	85.43 26

CLD-MVS		67.02 44	71.57 37	61.71 46	71.01 28	74.81 32	71.62 46	38.91 165	71.86 35	60.70 16	64.97 34	67.88 52	51.88 93	76.77 27	74.98 31	76.11 94	69.75 119
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
MP-MVS		74.31 13	78.87 13	68.99 16	73.49 17	78.56 10	79.25 17	56.51 12	75.33 22	60.69 17	75.30 14	79.12 17	61.81 16	77.78 18	77.93 8	82.18 22	88.06 9

MSLP-MVS++		68.17 38	70.72 43	65.19 33	69.41 36	70.64 50	74.99 35	45.76 66	70.20 41	60.17 18	56.42 61	73.01 37	61.14 19	72.80 46	70.54 50	79.70 42	81.42 45

MCST-MVS		73.67 20	77.39 21	69.33 14	76.26 5	78.19 12	78.77 20	54.54 24	75.33 22	59.99 19	67.96 27	79.23 16	62.43 12	78.00 15	75.71 25	84.02 2	87.30 13

SteuartSystems-ACMMP		75.23 8	79.60 10	70.13 8	76.81 3	78.92 7	81.74 3	57.99 4	75.30 24	59.83 20	75.69 13	78.45 18	60.48 24	80.58 1	79.77 1	83.94 3	88.52 5
Skip Steuart: Steuart Systems R&D Blog.
APD-MVS		75.80 6	80.90 6	69.86 11	75.42 9	78.48 11	81.43 6	57.44 7	80.45 10	59.32 21	85.28 5	80.82 11	63.96 5	76.89 24	76.08 23	81.58 32	88.30 7
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
train_agg		73.89 17	78.25 17	68.80 18	75.25 11	72.27 46	79.75 12	56.05 16	74.87 27	58.97 22	81.83 7	79.76 15	61.05 21	77.39 22	76.01 24	81.71 29	85.61 24

CP-MVS		72.63 22	76.95 23	67.59 21	70.67 29	75.53 28	77.95 25	56.01 17	75.65 21	58.82 23	69.16 24	76.48 24	60.46 25	77.66 19	77.20 16	81.65 30	86.97 14

3Dnovator+	62.63 4	69.51 30	72.62 34	65.88 31	68.21 42	76.47 24	73.50 43	52.74 36	70.85 38	58.65 24	55.97 63	69.95 44	61.11 20	76.80 26	75.09 27	81.09 35	83.23 39

DeepC-MVS_fast	65.08 3	72.00 24	76.11 24	67.21 23	68.93 39	77.46 15	76.54 29	54.35 25	74.92 26	58.64 25	65.18 33	74.04 36	62.62 10	77.92 16	77.02 17	82.16 23	86.21 18
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
ACMMPR		73.79 19	78.41 16	68.40 19	72.35 22	77.79 14	79.32 15	56.38 14	77.67 19	58.30 26	74.16 16	76.66 23	61.40 18	78.32 10	77.80 10	82.68 13	86.51 16

DeepPCF-MVS	66.49 1	74.25 15	80.97 5	66.41 26	67.75 45	78.87 8	75.61 33	54.16 27	84.86 2	58.22 27	77.94 11	81.01 10	62.52 11	78.34 9	77.38 12	80.16 40	88.40 6

PGM-MVS		72.89 21	77.13 22	67.94 20	72.47 21	77.25 18	79.27 16	54.63 23	73.71 29	57.95 28	72.38 17	75.33 28	60.75 22	78.25 12	77.36 14	82.57 15	85.62 23

AdaColmap		67.89 40	68.85 50	66.77 24	73.73 16	74.30 37	75.28 34	53.58 30	70.24 40	57.59 29	51.19 82	59.19 75	60.74 23	75.33 37	73.72 40	79.69 44	77.96 59

TSAR-MVS + GP.		69.71 29	73.92 31	64.80 37	68.27 41	70.56 51	71.90 45	50.75 45	71.38 36	57.46 30	68.68 26	75.42 27	60.10 27	73.47 43	73.99 37	80.32 38	83.97 33

CNLPA		62.78 55	66.31 55	58.65 54	58.47 82	68.41 59	65.98 85	41.22 150	78.02 18	56.04 31	46.65 121	59.50 74	57.50 37	69.67 64	65.27 121	72.70 135	76.67 68

ACMP	61.42 5	68.72 37	71.37 38	65.64 32	69.06 38	74.45 36	75.88 32	53.30 31	68.10 44	55.74 32	61.53 48	62.29 63	56.97 42	74.70 39	74.23 36	82.88 10	84.31 30
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
ACMMP		71.57 25	75.84 25	66.59 25	70.30 33	76.85 23	78.46 22	53.95 28	73.52 30	55.56 33	70.13 21	71.36 41	58.55 33	77.00 23	76.23 22	82.71 12	85.81 22
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
MAR-MVS		68.04 39	70.74 42	64.90 36	71.68 26	76.33 25	74.63 38	50.48 49	63.81 50	55.52 34	54.88 68	69.90 45	57.39 38	75.42 36	74.79 32	79.71 41	80.03 49
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
TSAR-MVS + ACMM		72.56 23	79.07 12	64.96 35	73.24 19	73.16 43	78.50 21	48.80 58	79.34 13	55.32 35	85.04 6	81.49 8	58.57 32	75.06 38	73.75 39	75.35 102	85.61 24

OMC-MVS		65.16 49	71.35 39	57.94 60	52.95 158	68.82 56	69.00 50	38.28 172	79.89 11	55.20 36	62.76 41	68.31 49	56.14 48	71.30 53	68.70 63	76.06 96	79.67 50

MVS_111021_HR		67.62 41	70.39 44	64.39 38	69.77 35	70.45 52	71.44 48	51.72 41	60.77 57	55.06 37	62.14 45	66.40 55	58.13 35	76.13 29	74.79 32	80.19 39	82.04 43

3Dnovator	60.86 6	66.99 45	70.32 45	63.11 43	66.63 48	74.52 33	71.56 47	45.76 66	67.37 46	55.00 38	54.31 72	68.19 50	58.49 34	73.97 42	73.63 41	81.22 34	80.23 48

HQP-MVS		70.88 28	75.02 28	66.05 29	71.69 25	74.47 35	77.51 26	53.17 32	72.89 31	54.88 39	70.03 22	70.48 43	57.26 39	76.02 30	75.01 30	81.78 27	86.21 18

CANet		68.77 35	73.01 32	63.83 41	68.30 40	75.19 29	73.73 42	47.90 59	63.86 49	54.84 40	67.51 29	74.36 34	57.62 36	74.22 41	73.57 42	80.56 36	82.36 40

ACMM	60.30 7	67.58 42	68.82 51	66.13 28	70.59 30	72.01 48	76.54 29	54.26 26	65.64 48	54.78 41	50.35 84	61.72 66	58.74 31	75.79 33	75.03 28	81.88 25	81.17 46
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
abl_6				64.36 39	70.08 34	77.45 16	72.88 44	50.15 50	71.31 37	54.77 42	62.79 40	77.99 21	56.80 44			81.50 33	83.91 34

XVS					70.49 31	76.96 20	74.36 39			54.48 43		74.47 31				82.24 18

X-MVStestdata					70.49 31	76.96 20	74.36 39			54.48 43		74.47 31				82.24 18

X-MVS		71.18 27	75.66 27	65.96 30	71.71 24	76.96 20	77.26 27	55.88 18	72.75 32	54.48 43	64.39 36	74.47 31	54.19 55	77.84 17	77.37 13	82.21 20	85.85 21

PCF-MVS	59.98 8	67.32 43	71.04 41	62.97 44	64.77 56	74.49 34	74.78 37	49.54 52	67.44 45	54.39 46	58.35 56	72.81 38	55.79 51	71.54 51	69.24 58	78.57 50	83.41 37
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
MVS_0304		69.49 31	73.96 30	64.28 40	67.92 43	76.13 26	74.90 36	47.60 60	63.29 52	54.09 47	67.44 30	76.35 25	59.53 29	75.81 32	75.03 28	81.62 31	83.70 36

OPM-MVS		69.33 32	71.05 40	67.32 22	72.34 23	75.70 27	79.57 14	56.34 15	55.21 63	53.81 48	59.51 52	68.96 46	59.67 28	77.61 20	76.44 21	82.19 21	83.88 35

MVS_111021_LR		63.05 53	66.43 54	59.10 53	61.33 68	63.77 101	65.87 86	43.58 104	60.20 58	53.70 49	62.09 46	62.38 62	55.84 50	70.24 61	68.08 67	74.30 107	78.28 58

CDPH-MVS		71.47 26	75.82 26	66.41 26	72.97 20	77.15 19	78.14 24	54.71 22	69.88 42	53.07 50	70.98 18	74.83 30	56.95 43	76.22 28	76.57 19	82.62 14	85.09 29

PVSNet_BlendedMVS		61.63 59	64.82 63	57.91 62	57.21 123	67.55 66	63.47 103	46.08 64	54.72 64	52.46 51	58.59 54	60.73 68	51.82 94	70.46 58	65.20 123	76.44 83	76.50 74

PVSNet_Blended		61.63 59	64.82 63	57.91 62	57.21 123	67.55 66	63.47 103	46.08 64	54.72 64	52.46 51	58.59 54	60.73 68	51.82 94	70.46 58	65.20 123	76.44 83	76.50 74

LGP-MVS_train		68.87 34	72.03 36	65.18 34	69.33 37	74.03 39	76.67 28	53.88 29	68.46 43	52.05 53	63.21 38	63.89 58	56.31 45	75.99 31	74.43 34	82.83 11	84.18 31

PHI-MVS		69.27 33	74.84 29	62.76 45	66.83 47	74.83 31	73.88 41	49.32 54	70.61 39	50.93 54	69.62 23	74.84 29	57.25 40	75.53 34	74.32 35	78.35 54	84.17 32

PVSNet_Blended_VisFu		63.65 51	66.92 52	59.83 51	60.03 74	73.44 42	66.33 79	48.95 56	52.20 77	50.81 55	56.07 62	60.25 71	53.56 60	73.23 45	70.01 55	79.30 46	83.24 38

CPTT-MVS		68.76 36	73.01 32	63.81 42	65.42 54	73.66 41	76.39 31	52.08 37	72.61 33	50.33 56	60.73 49	72.65 39	59.43 30	73.32 44	72.12 44	79.19 48	85.99 20

OpenMVS	57.13 9	62.81 54	65.75 58	59.39 52	66.47 50	69.52 54	64.26 99	43.07 120	61.34 56	50.19 57	47.29 118	64.41 57	54.60 54	70.18 62	68.62 65	77.73 56	78.89 53

MVS_Test		62.40 57	66.23 56	57.94 60	59.77 77	64.77 97	66.50 78	41.76 140	57.26 61	49.33 58	62.68 42	67.47 54	53.50 63	68.57 73	66.25 91	76.77 71	76.58 71

QAPM		65.27 48	69.49 49	60.35 48	65.43 53	72.20 47	65.69 88	47.23 61	63.46 51	49.14 59	53.56 73	71.04 42	57.01 41	72.60 47	71.41 47	77.62 58	82.14 42

DELS-MVS		65.87 46	70.30 46	60.71 47	64.05 63	72.68 45	70.90 49	45.43 70	57.49 60	49.05 60	64.43 35	68.66 47	55.11 53	74.31 40	73.02 43	79.70 42	81.51 44
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
TAPA-MVS	54.74 10	60.85 61	66.61 53	54.12 100	47.38 186	65.33 90	65.35 91	36.51 183	75.16 25	48.82 61	54.70 69	63.51 59	53.31 67	68.36 74	64.97 126	73.37 120	74.27 104
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
tpmp4_e23		56.84 109	57.14 128	56.49 90	62.45 65	62.05 116	67.57 53	41.56 145	54.17 66	48.57 62	49.18 88	46.54 146	50.44 98	61.93 163	58.82 177	68.34 170	67.28 137

canonicalmvs		65.62 47	72.06 35	58.11 56	63.94 64	71.05 49	64.49 97	43.18 115	74.08 28	47.35 63	64.17 37	71.97 40	51.17 96	71.87 49	70.74 48	78.51 52	80.56 47

tpm cat1		53.30 131	53.41 159	53.17 107	58.16 83	59.15 151	63.73 102	38.27 173	50.73 81	46.98 64	45.57 136	44.00 171	49.20 102	55.90 199	54.02 198	62.65 192	64.50 166

DI_MVS_plusplus_trai		61.88 58	65.17 62	58.06 57	60.05 73	65.26 92	66.03 83	44.22 82	55.75 62	46.73 65	54.64 70	68.12 51	54.13 57	69.13 66	66.66 83	77.18 64	76.61 69

Effi-MVS+		63.28 52	65.96 57	60.17 49	64.26 60	68.06 60	68.78 51	45.71 68	54.08 67	46.64 66	55.92 64	63.13 61	55.94 49	70.38 60	71.43 46	79.68 45	78.70 54

v18		58.68 80	60.20 88	56.90 79	57.26 121	63.28 110	66.58 77	42.42 135	48.86 94	46.37 67	49.01 94	53.05 96	52.74 77	67.40 109	65.52 117	76.02 98	74.28 103

v16		58.71 77	60.20 88	56.97 70	57.35 108	63.36 109	66.67 75	42.49 132	48.69 100	46.36 68	48.87 98	52.92 102	52.82 72	67.57 99	65.58 115	76.15 93	74.38 100

TSAR-MVS + COLMAP		62.65 56	69.90 47	54.19 99	46.31 190	66.73 75	65.49 90	41.36 148	76.57 20	46.31 69	76.80 12	56.68 83	53.27 68	69.50 65	66.65 84	72.40 141	76.36 76

EPNet		65.14 50	69.54 48	60.00 50	66.61 49	67.67 65	67.53 54	55.32 20	62.67 54	46.22 70	67.74 28	65.93 56	48.07 109	72.17 48	72.12 44	76.28 86	78.47 56
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
diffmvs		59.53 65	64.04 68	54.26 98	55.09 138	59.86 141	64.80 94	39.55 164	58.39 59	46.21 71	60.48 50	67.82 53	49.27 101	63.53 150	63.32 146	70.64 156	74.89 86

v17		58.69 78	60.19 91	56.94 72	57.38 103	63.37 108	66.67 75	42.47 134	48.52 104	46.10 72	48.90 97	53.00 97	52.84 70	67.58 98	65.60 111	76.19 91	74.38 100

v6		58.89 71	60.54 81	56.96 71	57.34 110	66.13 84	66.71 70	42.84 122	47.85 108	45.80 73	49.04 90	52.95 99	52.79 73	67.53 101	65.59 112	76.26 87	74.73 88

v8		58.88 72	60.57 80	56.92 76	57.35 108	65.69 89	66.69 74	42.64 130	47.89 107	45.77 74	49.04 90	52.98 98	52.77 76	67.51 104	65.57 116	76.26 87	75.30 84

v1neww		58.88 72	60.54 81	56.94 72	57.33 112	66.13 84	66.70 72	42.84 122	47.84 109	45.74 75	49.02 92	52.93 100	52.78 74	67.53 101	65.59 112	76.26 87	74.73 88

v7new		58.88 72	60.54 81	56.94 72	57.33 112	66.13 84	66.70 72	42.84 122	47.84 109	45.74 75	49.02 92	52.93 100	52.78 74	67.53 101	65.59 112	76.26 87	74.73 88

v7		59.19 69	60.62 78	57.53 65	57.96 85	67.19 71	67.09 59	44.28 81	46.84 114	45.45 77	48.19 110	51.06 116	53.62 59	67.84 91	66.59 86	76.79 68	76.60 70

v10		59.17 70	60.60 79	57.50 66	57.95 86	66.73 75	67.09 59	44.11 83	46.85 113	45.42 78	48.18 112	51.07 115	53.63 58	67.84 91	66.59 86	76.79 68	76.92 66

Fast-Effi-MVS+		60.36 62	63.35 69	56.87 82	58.70 79	65.86 88	65.08 92	37.11 178	53.00 74	45.36 79	52.12 78	56.07 88	56.27 46	71.28 54	69.42 57	78.71 49	75.69 80

v2v482		58.69 78	60.12 95	57.03 69	57.16 125	66.05 87	67.17 57	43.52 106	46.33 119	45.19 80	49.46 87	51.02 117	52.51 84	67.30 111	66.03 94	76.61 79	74.62 95

v15		58.43 92	59.75 100	56.88 81	57.45 99	63.44 106	66.84 65	42.65 129	46.24 121	45.07 81	48.68 102	52.07 106	52.63 82	67.84 91	65.70 105	76.65 75	74.31 102

V9		58.45 88	59.75 100	56.92 76	57.51 96	63.49 104	66.86 62	42.73 127	46.07 125	45.05 82	48.45 105	51.99 108	52.66 80	68.04 89	65.75 101	76.72 73	74.50 97

V14		58.44 89	59.75 100	56.90 79	57.48 98	63.46 105	66.85 64	42.68 128	46.16 122	45.03 83	48.57 103	52.04 107	52.65 81	67.93 90	65.72 104	76.69 74	74.40 99

v13		58.44 89	59.72 104	56.94 72	57.55 90	63.51 102	66.86 62	42.81 125	45.90 128	44.98 84	48.17 113	51.87 110	52.68 78	68.20 79	65.78 99	76.78 70	74.63 94

v12		58.44 89	59.74 103	56.92 76	57.54 92	63.50 103	66.84 65	42.77 126	45.96 126	44.95 85	48.31 106	51.94 109	52.67 79	68.14 82	65.75 101	76.75 72	74.55 96

CMPMVS	37.70 17	49.24 163	52.71 163	45.19 165	45.97 192	51.23 188	47.44 181	29.31 210	43.04 153	44.69 86	34.45 196	48.35 127	43.64 128	62.59 154	59.82 172	60.08 197	69.48 125
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
v1		58.56 81	60.06 96	56.83 84	57.36 105	66.19 81	66.80 67	43.10 119	45.87 130	44.68 87	48.73 99	51.83 112	52.38 86	67.45 107	65.65 106	76.63 76	74.66 91

CostFormer		56.57 110	59.13 113	53.60 101	57.52 95	61.12 128	66.94 61	35.95 185	53.44 68	44.68 87	55.87 65	54.44 91	48.21 107	60.37 171	58.33 180	68.27 172	70.33 117

v1141		58.56 81	60.05 97	56.81 85	57.36 105	66.18 82	66.80 67	43.11 117	45.87 130	44.60 89	48.71 100	51.83 112	52.38 86	67.46 105	65.64 109	76.63 76	74.66 91

divwei89l23v2f112		58.56 81	60.05 97	56.81 85	57.36 105	66.18 82	66.80 67	43.11 117	45.89 129	44.60 89	48.71 100	51.84 111	52.38 86	67.45 107	65.65 106	76.63 76	74.66 91

v11		58.19 98	59.47 109	56.70 87	57.54 92	63.42 107	66.28 81	42.49 132	45.62 134	44.59 91	48.16 114	50.78 120	52.84 70	67.80 95	65.76 100	76.49 82	74.76 87

PLC	52.09 14	59.21 68	62.47 71	55.41 95	53.24 157	64.84 96	64.47 98	40.41 158	65.92 47	44.53 92	46.19 129	55.69 89	55.33 52	68.24 78	65.30 120	74.50 105	71.09 111
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
v1144		58.88 72	60.16 92	57.39 67	58.03 84	67.26 69	67.14 58	44.46 80	45.17 136	44.33 93	47.81 115	49.92 125	53.20 69	67.77 96	66.62 85	77.15 65	76.58 71

IB-MVS	54.11 11	58.36 94	60.70 77	55.62 93	58.67 80	68.02 61	61.56 106	43.15 116	46.09 123	44.06 94	44.24 144	50.99 119	48.71 104	66.70 120	70.33 51	77.60 59	78.50 55
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
MSDG		58.46 87	58.97 115	57.85 64	66.27 52	66.23 80	67.72 52	42.33 136	53.43 69	43.68 95	43.39 152	45.35 153	49.75 99	68.66 71	67.77 70	77.38 62	67.96 132

v1192		58.51 84	59.66 105	57.17 68	57.82 87	67.72 63	66.21 82	44.83 76	44.15 143	43.49 96	46.68 120	47.94 129	53.55 61	67.39 110	66.51 88	77.13 66	77.20 64

tpm		48.82 167	51.27 178	45.96 162	54.10 152	47.35 199	56.05 136	30.23 208	46.70 115	43.21 97	52.54 77	47.55 135	37.28 160	54.11 204	50.50 209	54.90 210	60.12 185

v144192		58.23 97	59.40 111	56.87 82	57.56 89	66.89 73	65.70 87	45.01 75	44.06 144	42.88 98	46.61 122	48.09 128	53.49 64	66.94 116	65.90 97	76.61 79	77.29 62

DWT-MVSNet_training		53.80 126	54.31 151	53.21 105	57.65 88	59.04 152	60.65 112	40.11 161	46.35 118	42.77 99	49.07 89	41.07 186	51.06 97	58.62 181	58.96 176	67.00 180	67.06 138

v1921920		57.89 101	59.02 114	56.58 89	57.55 90	66.66 78	64.72 96	44.70 78	43.55 147	42.73 100	46.17 130	46.93 143	53.51 62	66.78 119	65.75 101	76.29 85	77.28 63

V42		56.97 106	60.14 93	53.28 104	48.16 181	62.78 115	66.30 80	37.93 174	47.44 111	42.68 101	48.19 110	52.59 104	51.90 92	67.46 105	65.94 96	72.72 129	76.55 73

v1240		57.55 102	58.63 117	56.29 91	57.30 118	66.48 79	63.77 101	44.56 79	42.77 163	42.48 102	45.64 135	46.28 148	53.46 65	66.32 127	65.80 98	76.16 92	77.13 65

v148		55.58 116	57.61 126	53.20 106	54.59 147	61.86 117	61.18 109	38.70 170	44.30 142	42.25 103	47.53 116	50.24 124	48.73 103	65.15 144	62.61 156	73.79 113	71.61 109

MS-PatchMatch		58.19 98	60.20 88	55.85 92	65.17 55	64.16 99	64.82 93	41.48 147	50.95 79	42.17 104	45.38 137	56.42 84	48.08 108	68.30 75	66.70 82	73.39 119	69.46 127

Effi-MVS+-dtu		60.34 63	62.32 72	58.03 59	64.31 58	67.44 68	65.99 84	42.26 137	49.55 85	42.00 105	48.92 96	59.79 73	56.27 46	68.07 85	67.03 76	77.35 63	75.45 82

EG-PatchMatch MVS		56.98 105	58.24 121	55.50 94	64.66 57	68.62 57	61.48 108	43.63 103	38.44 196	41.44 106	38.05 188	46.18 150	43.95 126	71.71 50	70.61 49	77.87 55	74.08 105

MVSTER		57.19 103	61.11 76	52.62 111	50.82 174	58.79 154	61.55 107	37.86 175	48.81 96	41.31 107	57.43 60	52.10 105	48.60 105	68.19 81	66.75 81	75.56 100	75.68 81

LS3D		60.20 64	61.70 73	58.45 55	64.18 61	67.77 62	67.19 56	48.84 57	61.67 55	41.27 108	45.89 132	51.81 114	54.18 56	68.78 68	66.50 89	75.03 103	69.48 125

PatchmatchNet		49.92 160	51.29 177	48.32 145	51.83 168	51.86 186	53.38 160	37.63 177	47.90 106	40.83 109	48.54 104	45.30 154	45.19 123	56.86 188	53.99 200	61.08 196	54.57 201
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
pmmvs4		54.66 124	56.07 133	53.00 108	54.63 144	57.08 165	60.43 116	44.10 84	51.69 78	40.55 110	46.55 125	44.79 163	45.95 120	62.54 155	63.66 141	72.36 143	66.20 147

dps		50.42 155	51.20 181	49.51 126	55.88 130	56.07 167	53.73 154	38.89 166	43.66 145	40.36 111	45.66 134	37.63 203	45.23 122	59.05 174	56.18 183	62.94 191	60.16 184

IterMVS-LS		58.30 95	61.39 74	54.71 97	59.92 76	58.40 158	59.42 118	43.64 101	48.71 98	40.25 112	57.53 59	58.55 77	52.15 91	65.42 143	65.34 119	72.85 126	75.77 78
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
tpmrst		48.08 174	49.88 189	45.98 161	52.71 159	48.11 197	53.62 157	33.70 196	48.70 99	39.74 113	48.96 95	46.23 149	40.29 143	50.14 213	49.28 211	55.80 207	57.71 192

CR-MVSNet		50.47 154	52.61 164	47.98 150	49.03 180	52.94 180	48.27 175	38.86 167	44.41 139	39.59 114	44.34 143	44.65 166	46.63 117	58.97 176	60.31 169	65.48 183	62.66 173

Patchmtry						47.61 198	48.27 175	38.86 167		39.59 114

PatchT		48.08 174	51.03 182	44.64 169	42.96 203	50.12 191	40.36 210	35.09 187	43.17 152	39.59 114	42.00 174	39.96 193	46.63 117	58.97 176	60.31 169	63.21 190	62.66 173

CHOSEN 1792x2688		55.85 113	58.01 122	53.33 103	57.26 121	62.82 114	63.29 105	41.55 146	46.65 116	38.34 117	34.55 195	53.50 93	52.43 85	67.10 114	67.56 73	67.13 177	73.92 106

MVS-HIRNet		42.24 199	41.15 212	43.51 175	44.06 202	40.74 213	35.77 217	35.35 186	35.38 204	38.34 117	25.63 214	38.55 199	43.48 130	50.77 210	47.03 219	64.07 187	49.98 211

v7n		55.67 114	57.46 127	53.59 102	56.06 129	65.29 91	61.06 111	43.26 114	40.17 183	37.99 119	40.79 179	45.27 155	47.09 113	67.67 97	66.21 92	76.08 95	76.82 67

MDTV_nov1_ep13		50.32 157	52.43 167	47.86 152	49.87 177	54.70 174	58.10 122	34.29 191	45.59 135	37.71 120	47.44 117	47.42 136	41.86 136	58.07 184	55.21 191	65.34 185	58.56 190

PatchMatch-RL		50.11 159	51.56 174	48.43 143	46.23 191	51.94 185	50.21 168	38.62 171	46.62 117	37.51 121	42.43 168	39.38 195	52.24 90	60.98 167	59.56 173	65.76 182	60.01 186

HyFIR lowres test		56.87 108	58.60 118	54.84 96	56.62 128	69.27 55	64.77 95	42.21 138	45.66 133	37.50 122	33.08 197	57.47 82	53.33 66	65.46 142	67.94 68	74.60 104	71.35 110

CANet_DTU		58.88 72	64.68 65	52.12 114	55.77 131	66.75 74	63.92 100	37.04 179	53.32 70	37.45 123	59.81 51	61.81 65	44.43 125	68.25 76	67.47 74	74.12 110	75.33 83

pmmvs-eth3d		51.33 149	52.25 169	50.26 121	50.82 174	54.65 175	56.03 137	43.45 111	43.51 148	37.20 124	39.20 186	39.04 197	42.28 134	61.85 164	62.78 153	71.78 150	64.72 164

GA-MVS		55.67 114	58.33 119	52.58 112	55.23 136	63.09 111	61.08 110	40.15 160	42.95 154	37.02 125	52.61 76	47.68 132	47.51 111	65.92 135	65.35 118	74.49 106	70.68 115

V4		53.60 127	56.73 132	49.93 124	45.54 193	61.64 120	60.65 112	36.99 180	38.74 194	36.33 126	39.64 183	47.12 140	47.05 114	66.89 117	65.64 109	73.04 124	77.48 60

v52		53.60 127	56.74 131	49.93 124	45.54 193	61.64 120	60.65 112	36.99 180	38.75 192	36.32 127	39.64 183	47.13 139	47.05 114	66.89 117	65.65 106	73.04 124	77.48 60

USDC		51.11 150	53.71 155	48.08 149	44.76 196	55.99 168	53.01 161	40.90 152	52.49 75	36.14 128	44.67 142	33.66 210	43.27 132	63.23 151	61.10 162	70.39 159	64.82 163

ACMH+	53.71 12	59.26 67	60.28 85	58.06 57	64.17 62	68.46 58	67.51 55	50.93 44	52.46 76	35.83 129	40.83 178	45.12 156	52.32 89	69.88 63	69.00 61	77.59 60	76.21 77

RPSCF		46.41 187	54.42 149	37.06 204	25.70 232	45.14 209	45.39 195	20.81 226	62.79 53	35.10 130	44.92 141	55.60 90	43.56 129	56.12 196	52.45 205	51.80 216	63.91 169

Fast-Effi-MVS+-dtu		56.30 111	59.29 112	52.82 110	58.64 81	64.89 95	65.56 89	32.89 203	45.80 132	35.04 131	45.89 132	54.14 92	49.41 100	67.16 113	66.45 90	75.37 101	70.69 114

FC-MVSNet-train		58.40 93	63.15 70	52.85 109	64.29 59	61.84 118	55.98 138	46.47 62	53.06 72	34.96 132	61.95 47	56.37 86	39.49 144	68.67 70	68.36 66	75.92 99	71.81 108

PMMVS		49.20 165	54.28 152	43.28 178	34.13 218	45.70 208	48.98 172	26.09 221	46.31 120	34.92 133	55.22 66	53.47 94	47.48 112	59.43 173	59.04 175	68.05 173	60.77 181

v748		52.93 132	55.29 141	50.19 122	51.90 167	61.31 126	56.54 133	40.05 162	39.12 190	34.82 134	39.93 182	43.83 172	43.66 127	64.26 148	63.32 146	74.15 109	75.28 85

IterMVS		53.45 130	57.12 129	49.17 130	49.23 178	60.93 129	59.05 120	34.63 189	44.53 138	33.22 135	51.09 83	51.01 118	48.38 106	62.43 156	60.79 166	70.54 158	69.05 130
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
UGNet		57.03 104	65.25 61	47.44 155	46.54 189	66.73 75	56.30 134	43.28 113	50.06 82	32.99 136	62.57 43	63.26 60	33.31 176	68.25 76	67.58 72	72.20 145	78.29 57
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
COLMAP_ROB	46.52 15	51.99 143	54.86 146	48.63 142	49.13 179	61.73 119	60.53 115	36.57 182	53.14 71	32.95 137	37.10 189	38.68 198	40.49 141	65.72 138	63.08 149	72.11 146	64.60 165
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
anonymousdsp		52.84 133	57.78 124	47.06 156	40.24 209	58.95 153	53.70 155	33.54 198	36.51 203	32.69 138	43.88 146	45.40 152	47.97 110	67.17 112	70.28 52	74.22 108	82.29 41

test-LLR		49.28 162	50.29 185	48.10 148	55.26 134	47.16 200	49.52 169	43.48 109	39.22 188	31.98 139	43.65 150	47.93 130	41.29 139	56.80 189	55.36 189	67.08 178	61.94 176

TESTMET0.1,1		46.09 190	50.29 185	41.18 190	36.91 214	47.16 200	49.52 169	20.32 227	39.22 188	31.98 139	43.65 150	47.93 130	41.29 139	56.80 189	55.36 189	67.08 178	61.94 176

TinyColmap		47.08 183	47.56 197	46.52 160	42.35 205	53.44 179	51.77 165	40.70 156	43.44 150	31.92 141	29.78 205	23.72 225	45.04 124	61.99 162	59.54 174	67.35 176	61.03 180

RPMNet		46.41 187	48.72 192	43.72 174	47.77 184	52.94 180	46.02 192	33.92 193	44.41 139	31.82 142	36.89 190	37.42 204	37.41 154	53.88 205	54.02 198	65.37 184	61.47 178

UA-Net		58.50 85	64.68 65	51.30 116	66.97 46	67.13 72	53.68 156	45.65 69	49.51 87	31.58 143	62.91 39	68.47 48	35.85 166	68.20 79	67.28 75	74.03 111	69.24 129

TDRefinement		49.31 161	52.44 166	45.67 164	30.44 224	59.42 145	59.24 119	39.78 163	48.76 97	31.20 144	35.73 192	29.90 214	42.81 133	64.24 149	62.59 157	70.55 157	66.43 143

GBi-Net		55.20 119	60.25 86	49.31 127	52.42 160	61.44 122	57.03 128	44.04 86	49.18 90	30.47 145	48.28 107	58.19 78	38.22 149	68.05 86	66.96 77	73.69 115	69.65 120

test1		55.20 119	60.25 86	49.31 127	52.42 160	61.44 122	57.03 128	44.04 86	49.18 90	30.47 145	48.28 107	58.19 78	38.22 149	68.05 86	66.96 77	73.69 115	69.65 120

FMVSNet3		54.78 123	59.58 107	49.17 130	52.37 163	61.31 126	56.72 132	44.04 86	49.18 90	30.47 145	48.28 107	58.19 78	38.09 152	65.48 141	65.20 123	73.31 121	69.45 128

FMVSNet2		55.04 122	59.95 99	49.31 127	52.42 160	61.44 122	57.03 128	44.08 85	49.55 85	30.40 148	46.89 119	58.84 76	38.22 149	67.07 115	66.21 92	73.69 115	69.65 120

Vis-MVSNet		58.48 86	65.70 59	50.06 123	53.40 156	67.20 70	60.24 117	43.32 112	48.83 95	30.23 149	62.38 44	61.61 67	40.35 142	71.03 56	69.77 56	72.82 128	79.11 52
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
PM-MVS		44.55 194	48.13 195	40.37 193	32.85 222	46.82 204	46.11 191	29.28 211	40.48 181	29.99 150	39.98 181	34.39 209	41.80 137	56.08 197	53.88 202	62.19 194	65.31 157

LP		40.79 203	41.99 209	39.38 196	40.98 207	46.49 206	42.14 206	33.66 197	35.37 205	29.89 151	29.30 208	27.81 216	32.74 177	52.55 206	52.19 206	56.87 205	50.23 210

MDTV_nov1_ep13_2view		47.62 178	49.72 190	45.18 166	48.05 182	53.70 178	54.90 149	33.80 195	39.90 185	29.79 152	38.85 187	41.89 179	39.17 145	58.99 175	55.55 188	65.34 185	59.17 188

EPMVS		44.66 193	47.86 196	40.92 191	47.97 183	44.70 210	47.58 180	33.27 199	48.11 105	29.58 153	49.65 85	44.38 169	34.65 169	51.71 208	47.90 215	52.49 215	48.57 215

CDS-MVSNet		52.42 137	57.06 130	47.02 157	53.92 154	58.30 160	55.50 141	46.47 62	42.52 166	29.38 154	49.50 86	52.85 103	28.49 191	66.70 120	66.89 80	68.34 170	62.63 175
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
FMVSNet1		54.08 125	58.68 116	48.71 141	50.90 173	61.35 125	56.73 131	43.94 90	45.91 127	29.32 155	42.72 165	56.26 87	37.70 153	68.05 86	66.96 77	73.69 115	69.50 124

ACMH	52.42 13	58.24 96	59.56 108	56.70 87	66.34 51	69.59 53	66.71 70	49.12 55	46.08 124	28.90 156	42.67 166	41.20 185	52.60 83	71.39 52	70.28 52	76.51 81	75.72 79
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
ADS-MVSNet		40.67 205	43.38 207	37.50 203	44.36 198	39.79 216	42.09 207	32.67 205	44.34 141	28.87 157	40.76 180	40.37 191	30.22 184	48.34 223	45.87 221	46.81 224	44.21 219

pmmvs5		47.07 184	51.02 183	42.46 183	45.18 195	51.47 187	48.23 177	33.09 202	38.17 199	28.62 158	46.60 123	43.48 174	30.74 183	58.28 182	58.63 179	68.92 168	60.48 182

test-mter		45.30 191	50.37 184	39.38 196	33.65 220	46.99 202	47.59 179	18.59 229	38.75 192	28.00 159	43.28 155	46.82 145	41.50 138	57.28 187	55.78 186	66.93 181	63.70 170

thres100view900		52.04 141	54.81 147	48.80 139	57.31 114	59.33 146	55.30 148	42.92 121	42.85 157	27.81 160	43.00 161	45.06 158	36.99 161	64.74 146	63.51 143	72.47 140	65.21 160

tfpn200view9		52.53 134	55.51 135	49.06 132	57.31 114	60.24 131	55.42 143	43.77 92	42.85 157	27.81 160	43.00 161	45.06 158	37.32 155	66.38 122	64.54 128	72.71 132	66.54 140

tfpn111		52.44 136	55.38 138	49.01 134	57.31 114	60.24 131	55.42 143	43.77 92	42.85 157	27.51 162	42.03 172	45.06 158	37.32 155	66.38 122	64.54 128	72.71 132	66.54 140

conf200view11		52.51 135	55.51 135	49.01 134	57.31 114	60.24 131	55.42 143	43.77 92	42.85 157	27.51 162	43.00 161	45.06 158	37.32 155	66.38 122	64.54 128	72.71 132	66.54 140

conf0.01		52.02 142	54.62 148	49.00 136	57.30 118	60.17 136	55.42 143	43.76 95	42.85 157	27.49 164	43.12 158	39.71 194	37.32 155	66.26 130	64.54 128	72.72 129	65.66 155

conf0.002		51.76 147	54.13 153	49.00 136	57.28 120	60.15 137	55.42 143	43.75 97	42.85 157	27.49 164	43.13 157	37.12 206	37.32 155	66.23 131	64.17 135	72.72 129	65.24 159

thres200		52.39 138	55.37 140	48.90 138	57.39 102	60.18 134	55.60 140	43.73 98	42.93 155	27.41 166	43.35 153	45.09 157	36.61 164	66.36 125	63.92 140	72.66 136	65.78 153

EPNet_dtu		52.05 140	58.26 120	44.81 168	54.10 152	50.09 192	52.01 164	40.82 155	53.03 73	27.41 166	54.90 67	57.96 81	26.72 196	62.97 152	62.70 155	67.78 174	66.19 148
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
EPP-MVSNet		59.39 66	65.45 60	52.32 113	60.96 70	67.70 64	58.42 121	44.75 77	49.71 84	27.23 168	59.03 53	62.20 64	43.34 131	70.71 57	69.13 59	79.25 47	79.63 51

thres400		52.38 139	55.51 135	48.74 140	57.49 97	60.10 138	55.45 142	43.54 105	42.90 156	26.72 169	43.34 154	45.03 162	36.61 164	66.20 132	64.53 132	72.66 136	66.43 143

IS_MVSNet		57.95 100	64.26 67	50.60 118	61.62 67	65.25 93	57.18 127	45.42 71	50.79 80	26.49 170	57.81 58	60.05 72	34.51 170	71.24 55	70.20 54	78.36 53	74.44 98

tfpnnormal		50.16 158	52.19 170	47.78 153	56.86 127	58.37 159	54.15 152	44.01 89	38.35 198	25.94 171	36.10 191	37.89 200	34.50 171	65.93 134	63.42 144	71.26 153	65.28 158

view600		51.96 144	55.13 143	48.27 146	57.41 101	60.05 139	54.74 150	43.64 101	42.57 165	25.88 172	43.11 159	44.48 167	35.34 167	66.27 128	63.61 142	72.61 139	65.80 152

pm-mvs1		51.02 151	55.55 134	45.73 163	54.16 151	58.52 156	50.92 166	42.56 131	40.32 182	25.67 173	43.66 149	50.34 123	30.06 185	65.85 136	63.97 139	70.99 155	66.21 146

thres600view7		51.91 146	55.14 142	48.14 147	57.43 100	60.18 134	54.60 151	43.73 98	42.61 164	25.20 174	43.10 160	44.47 168	35.19 168	66.36 125	63.28 148	72.66 136	66.01 150

view800		51.55 148	54.89 145	47.66 154	57.37 104	59.77 143	53.62 157	43.72 100	42.22 167	24.94 175	42.80 164	43.81 173	33.94 174	66.09 133	64.38 134	72.39 142	65.14 162

TransMVSNet (Re)		51.92 145	55.38 138	47.88 151	60.95 71	59.90 140	53.95 153	45.14 73	39.47 187	24.85 176	43.87 147	46.51 147	29.15 187	67.55 100	65.23 122	73.26 123	65.16 161

ambc			45.54 203		50.66 176	52.63 183	40.99 209		38.36 197	24.67 177	22.62 219	13.94 232	29.14 188	65.71 139	58.06 181	58.60 201	67.43 134

pmmvs6		48.35 170	51.64 172	44.51 171	51.92 166	57.94 162	49.44 171	42.17 139	34.45 206	24.62 178	28.87 210	46.90 144	29.07 189	64.60 147	63.08 149	69.83 161	65.68 154

UniMVSNet_NR-MVSNet		56.94 107	61.14 75	52.05 115	60.02 75	65.21 94	57.44 125	52.93 34	49.37 88	24.31 179	54.62 71	50.54 121	39.04 146	68.69 69	68.84 62	78.53 51	70.72 112

DU-MVS		55.41 117	59.59 106	50.54 120	54.60 145	62.97 112	57.44 125	51.80 39	48.62 102	24.31 179	51.99 79	47.00 142	39.04 146	68.11 83	67.75 71	76.03 97	70.72 112

tfpn		50.58 153	53.65 157	47.00 158	57.34 110	59.31 147	52.41 162	43.76 95	41.81 171	23.86 181	42.49 167	37.80 201	32.63 179	65.68 140	64.02 138	71.99 148	64.41 167

conf0.05thres1000		50.64 152	53.84 154	46.92 159	57.02 126	59.29 148	52.29 163	43.80 91	39.84 186	23.81 182	39.26 185	43.14 176	32.52 180	65.74 137	64.04 136	72.05 147	65.53 156

MIMVSNet		43.79 196	48.53 193	38.27 200	41.46 206	48.97 195	50.81 167	32.88 204	44.55 137	22.07 183	32.05 198	47.15 138	24.76 199	58.73 178	56.09 185	57.63 204	52.14 204

FMVSNet5		40.96 201	45.81 201	35.29 208	34.30 217	44.55 211	47.28 182	28.84 212	40.76 179	21.62 184	29.85 204	42.44 177	24.77 198	57.53 186	55.00 192	54.93 209	50.56 209

thresconf0.02		48.17 173	51.22 180	44.60 170	55.14 137	55.73 169	48.95 173	41.35 149	43.43 151	21.23 185	42.03 172	37.25 205	31.19 182	62.33 159	60.61 168	69.76 162	57.17 194

tfpn_ndepth		48.34 171	52.27 168	43.76 173	54.35 149	56.46 166	47.24 183	40.92 151	43.45 149	21.04 186	41.16 177	43.22 175	28.90 190	61.57 165	60.65 167	70.12 160	59.34 187

UniMVSNet (Re)		55.15 121	60.39 84	49.03 133	55.31 133	64.59 98	55.77 139	50.63 46	48.66 101	20.95 187	51.47 81	50.40 122	34.41 172	67.81 94	67.89 69	77.11 67	71.88 107

NR-MVSNet		55.35 118	59.46 110	50.56 119	61.33 68	62.97 112	57.91 124	51.80 39	48.62 102	20.59 188	51.99 79	44.73 164	34.10 173	68.58 72	68.64 64	77.66 57	70.67 116

TranMVSNet+NR-MVSNet		55.87 112	60.14 93	50.88 117	59.46 78	63.82 100	57.93 123	52.98 33	48.94 93	20.52 189	52.87 75	47.33 137	36.81 163	69.12 67	69.03 60	77.56 61	69.89 118

TAMVS		44.02 195	49.18 191	37.99 202	47.03 188	45.97 207	45.04 196	28.47 213	39.11 191	20.23 190	43.22 156	48.52 126	28.49 191	58.15 183	57.95 182	58.71 199	51.36 206

SixPastTwentyTwo		47.55 182	50.25 187	44.41 172	47.30 187	54.31 177	47.81 178	40.36 159	33.76 207	19.93 191	43.75 148	32.77 212	42.07 135	59.82 172	60.94 165	68.98 167	66.37 145

PMVS	27.84 18	33.81 215	35.28 220	32.09 212	34.13 218	24.81 231	32.51 221	26.48 220	26.41 221	19.37 192	23.76 217	24.02 224	25.18 197	50.78 209	47.24 218	54.89 211	49.95 212
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
FPMVS		38.36 210	40.41 213	35.97 205	38.92 211	39.85 215	45.50 194	25.79 222	41.13 177	18.70 193	30.10 203	24.56 220	31.86 181	49.42 218	46.80 220	55.04 208	51.03 207

tfpn_n400		47.56 180	51.56 174	42.90 181	54.91 140	55.28 172	46.21 187	41.59 142	41.51 174	18.54 194	42.25 169	41.54 182	27.12 193	62.41 157	61.02 163	69.05 165	56.90 196

tfpnconf		47.56 180	51.56 174	42.90 181	54.91 140	55.28 172	46.21 187	41.59 142	41.51 174	18.54 194	42.25 169	41.54 182	27.12 193	62.41 157	61.02 163	69.05 165	56.90 196

tfpnview11		47.58 179	51.57 173	42.92 180	54.94 139	55.30 171	46.21 187	41.58 144	42.10 168	18.54 194	42.25 169	41.54 182	27.12 193	62.29 160	61.12 161	69.15 164	56.40 198

CHOSEN 280x420		40.80 202	45.05 204	35.84 207	32.95 221	29.57 228	44.98 197	23.71 224	37.54 201	18.42 197	31.36 201	47.07 141	46.41 119	56.71 191	54.65 196	48.55 221	58.47 191

MDA-MVSNet-bldmvs		41.36 200	43.15 208	39.27 198	28.74 226	52.68 182	44.95 198	40.84 154	32.89 209	18.13 198	31.61 200	22.09 227	38.97 148	50.45 212	56.11 184	64.01 188	56.23 199

Baseline_NR-MVSNet		53.50 129	57.89 123	48.37 144	54.60 145	59.25 150	56.10 135	51.84 38	49.32 89	17.92 199	45.38 137	47.68 132	36.93 162	68.11 83	65.95 95	72.84 127	69.57 123

pmmvs3		35.10 213	38.47 214	31.17 213	26.37 231	40.47 214	34.51 220	18.09 230	24.75 222	16.88 200	23.05 218	26.69 218	32.69 178	50.73 211	51.60 207	58.46 202	51.98 205

tfpn1000		46.75 186	51.24 179	41.51 188	54.39 148	55.60 170	43.85 201	40.90 152	41.82 170	16.71 201	41.26 176	41.58 181	23.96 200	60.76 168	60.27 171	69.26 163	57.42 193

testpf		34.85 214	36.16 219	33.31 210	47.49 185	35.56 224	36.85 215	32.31 206	23.08 225	15.63 202	29.39 207	29.48 215	19.62 209	41.38 226	41.07 225	47.95 222	53.18 202

test0.0.03 1		43.15 197	46.95 198	38.72 199	55.26 134	50.56 189	42.48 205	43.48 109	38.16 200	15.11 203	35.07 194	44.69 165	16.47 213	55.95 198	54.34 197	59.54 198	49.87 213

CVMVSNet		46.38 189	52.01 171	39.81 194	42.40 204	50.26 190	46.15 190	37.68 176	40.03 184	15.09 204	46.56 124	47.56 134	33.72 175	56.50 193	55.65 187	63.80 189	67.53 133

LTVRE_ROB	44.17 16	47.06 185	50.15 188	43.44 176	51.39 170	58.42 157	42.90 204	43.51 107	22.27 228	14.85 205	41.94 175	34.57 208	45.43 121	62.28 161	62.77 154	62.56 193	68.83 131
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		42.28 198	45.89 200	38.07 201	51.96 165	48.98 194	43.66 203	38.81 169	38.74 194	14.32 206	26.74 212	40.90 187	20.94 206	56.64 192	54.67 195	58.71 199	54.59 200

Vis-MVSNet (Re-imp)		50.37 156	57.73 125	41.80 187	57.53 94	54.35 176	45.70 193	45.24 72	49.80 83	13.43 207	58.23 57	56.42 84	20.11 208	62.96 153	63.36 145	68.76 169	58.96 189

tmp_tt				5.40 233	3.97 236	2.35 238	3.26 238	0.44 234	17.56 230	12.09 208	11.48 230	7.14 235	1.98 233	15.68 232	15.49 232	10.69 235

gg-mvs-nofinetune		49.07 166	52.56 165	45.00 167	61.99 66	59.78 142	53.55 159	41.63 141	31.62 213	12.08 209	29.56 206	53.28 95	29.57 186	66.27 128	64.49 133	71.19 154	62.92 172

gm-plane-assit		44.74 192	45.95 199	43.33 177	60.88 72	46.79 205	36.97 214	32.24 207	24.15 224	11.79 210	29.26 209	32.97 211	46.64 116	65.09 145	62.95 151	71.45 152	60.42 183

1111		31.35 219	33.52 223	28.83 216	44.28 200	32.44 225	31.71 222	33.25 200	27.87 217	10.92 211	22.18 220	24.05 222	15.89 215	49.03 221	44.09 222	36.94 229	34.96 225

.test1245		22.44 227	22.23 228	22.67 225	44.28 200	32.44 225	31.71 222	33.25 200	27.87 217	10.92 211	22.18 220	24.05 222	15.89 215	49.03 221	0.01 233	0.00 237	0.06 235

test2356		33.40 216	36.53 217	29.76 214	37.51 213	38.39 219	34.68 219	27.35 215	27.88 216	10.61 213	25.54 215	24.44 221	17.15 212	49.99 215	48.32 213	51.24 217	41.16 223

CP-MVSNet		48.37 169	53.53 158	42.34 184	51.35 171	58.01 161	46.56 185	50.54 47	41.62 173	10.61 213	46.53 126	40.68 190	23.18 202	58.71 179	61.83 158	71.81 149	67.36 136

PS-CasMVS		48.18 172	53.25 162	42.27 185	51.26 172	57.94 162	46.51 186	50.52 48	41.30 176	10.56 215	45.35 139	40.34 192	23.04 204	58.66 180	61.79 159	71.74 151	67.38 135

EU-MVSNet		40.63 206	45.65 202	34.78 209	39.11 210	46.94 203	40.02 211	34.03 192	33.50 208	10.37 216	35.57 193	37.80 201	23.65 201	51.90 207	50.21 210	61.49 195	63.62 171

PEN-MVS		49.21 164	54.32 150	43.24 179	54.33 150	59.26 149	47.04 184	51.37 43	41.67 172	9.97 217	46.22 128	41.80 180	22.97 205	60.52 169	64.03 137	73.73 114	66.75 139

test20.03		40.38 207	44.20 205	35.92 206	53.73 155	49.05 193	38.54 212	43.49 108	32.55 210	9.54 218	27.88 211	39.12 196	12.24 224	56.28 194	54.69 194	57.96 203	49.83 214

N_pmnet		32.67 218	36.85 216	27.79 219	40.55 208	32.13 227	35.80 216	26.79 219	37.24 202	9.10 219	32.02 199	30.94 213	16.30 214	47.22 224	41.21 224	38.21 227	37.21 224

Gipuma		25.87 224	26.91 227	24.66 224	28.98 225	20.17 232	20.46 231	34.62 190	29.55 215	9.10 219	4.91 235	5.31 237	15.76 217	49.37 219	49.10 212	39.03 226	29.95 228
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
testgi		38.71 209	43.64 206	32.95 211	52.30 164	48.63 196	35.59 218	35.05 188	31.58 214	9.03 221	30.29 202	40.75 189	11.19 229	55.30 200	53.47 203	54.53 212	45.48 217

WR-MVS		48.78 168	55.06 144	41.45 189	55.50 132	60.40 130	43.77 202	49.99 51	41.92 169	8.10 222	45.24 140	45.56 151	17.47 210	61.57 165	64.60 127	73.85 112	66.14 149

DTE-MVSNet		48.03 176	53.28 161	41.91 186	54.64 143	57.50 164	44.63 200	51.66 42	41.02 178	7.97 223	46.26 127	40.90 187	20.24 207	60.45 170	62.89 152	72.33 144	63.97 168

WR-MVS_H		47.65 177	53.67 156	40.63 192	51.45 169	59.74 144	44.71 199	49.37 53	40.69 180	7.61 224	46.04 131	44.34 170	17.32 211	57.79 185	61.18 160	73.30 122	65.86 151

Anonymous20231211		40.75 204	41.57 210	39.80 195	54.71 142	52.32 184	41.42 208	45.09 74	24.45 223	6.80 225	14.58 227	23.43 226	23.08 203	56.20 195	58.74 178	67.68 175	61.31 179

MIMVSNet1		35.51 212	41.41 211	28.63 217	27.53 228	43.36 212	38.09 213	33.82 194	32.01 211	6.77 226	21.63 222	35.43 207	11.97 226	55.05 202	53.99 200	53.59 214	48.36 216

new-patchmatchnet		33.24 217	37.20 215	28.62 218	44.32 199	38.26 221	29.68 227	36.05 184	31.97 212	6.33 227	26.59 213	27.33 217	11.12 230	50.08 214	41.05 226	44.23 225	45.15 218

testus		31.33 220	36.31 218	25.52 223	37.55 212	38.40 218	25.87 228	23.58 225	26.46 220	5.97 228	24.15 216	24.92 219	12.44 223	49.14 220	48.21 214	47.73 223	42.86 220

testmv		30.97 221	34.42 221	26.95 220	36.49 215	37.38 222	29.80 225	27.28 216	22.34 226	4.72 229	20.63 224	20.64 228	13.22 221	49.86 217	47.74 216	50.20 218	42.36 221

test1235678		30.97 221	34.42 221	26.95 220	36.49 215	37.38 222	29.79 226	27.28 216	22.33 227	4.72 229	20.62 225	20.64 228	13.22 221	49.87 216	47.74 216	50.20 218	42.36 221

new_pmnet		23.19 226	28.17 226	17.37 227	17.03 233	24.92 230	19.66 232	16.16 232	27.05 219	4.42 231	20.77 223	19.20 230	12.19 225	37.71 228	36.38 227	34.77 230	31.17 227

no-one		29.19 223	31.89 224	26.05 222	30.96 223	38.33 220	21.54 229	29.86 209	15.84 232	3.56 232	11.28 231	13.03 233	14.44 220	38.96 227	52.83 204	55.96 206	52.92 203

E-PMN		15.09 229	13.19 231	17.30 228	27.80 227	12.62 235	7.81 235	27.54 214	14.62 234	3.19 233	6.89 232	2.52 240	15.09 218	15.93 231	20.22 230	22.38 231	19.53 231

EMVS		14.49 230	12.45 232	16.87 230	27.02 229	12.56 236	8.13 234	27.19 218	15.05 233	3.14 234	6.69 233	2.67 239	15.08 219	14.60 233	18.05 231	20.67 232	17.56 233

FC-MVSNet-test		39.65 208	48.35 194	29.49 215	44.43 197	39.28 217	30.23 224	40.44 157	43.59 146	3.12 235	53.00 74	42.03 178	10.02 231	55.09 201	54.77 193	48.66 220	50.71 208

MVE	12.28 19	13.53 231	15.72 230	10.96 232	7.39 235	15.71 234	6.05 236	23.73 223	10.29 236	3.01 236	5.77 234	3.41 238	11.91 227	20.11 230	29.79 228	13.67 234	24.98 229
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
test12356		23.91 225	28.47 225	18.60 226	26.80 230	28.30 229	20.92 230	19.76 228	19.89 229	2.88 237	18.48 226	16.57 231	4.05 232	42.34 225	41.93 223	37.21 228	31.75 226

DeepMVS_CX						6.95 237	5.98 237	2.25 233	11.73 235	2.07 238	11.85 229	5.43 236	11.75 228	11.40 234		8.10 236	18.38 232

PMMVS2		15.84 228	19.68 229	11.35 231	15.74 234	16.95 233	13.31 233	17.64 231	16.08 231	0.36 239	13.12 228	11.47 234	1.69 234	28.82 229	27.24 229	19.38 233	24.09 230

GG-mvs-BLEND		36.62 211	53.39 160	17.06 229	0.01 237	58.61 155	48.63 174	0.01 235	47.13 112	0.02 240	43.98 145	60.64 70	0.03 235	54.92 203	51.47 208	53.64 213	56.99 195

sosnet-low-res		0.00 234	0.00 235	0.00 234	0.00 238	0.00 239	0.00 241	0.00 236	0.00 239	0.00 241	0.00 238	0.00 241	0.00 238	0.00 236	0.00 236	0.00 237	0.00 237

sosnet		0.00 234	0.00 235	0.00 234	0.00 238	0.00 239	0.00 241	0.00 236	0.00 239	0.00 241	0.00 238	0.00 241	0.00 238	0.00 236	0.00 236	0.00 237	0.00 237

testmvs		0.01 232	0.02 233	0.00 234	0.00 238	0.00 239	0.01 240	0.00 236	0.01 237	0.00 241	0.03 237	0.00 241	0.01 236	0.01 235	0.01 233	0.00 237	0.06 235

test123		0.01 232	0.02 233	0.00 234	0.00 238	0.00 239	0.00 241	0.00 236	0.01 237	0.00 241	0.04 236	0.00 241	0.01 236	0.00 236	0.01 233	0.00 237	0.07 234

Patchmatch-RL test							1.04 239

mPP-MVS					71.67 27							74.36 34

NP-MVS									72.00 34

High-resolution multi-view benchmark