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	botani.	boulde.	bridge	door	exhibi.	lectur.	living.	lounge	observ.	old co.	statue	terrac.

HSP-MVS		90.38 2	91.89 1	85.84 70	92.83 58	64.03 171	93.06 77	94.52 32	82.19 19	93.65 1	96.15 13	85.89 1	97.19 60	91.02 10	97.75 1	96.29 16

GG-mvs-BLEND				86.53 51	91.91 80	69.67 34	75.02 315	94.75 27		78.67 84	90.85 115	77.91 2	94.56 149	72.25 122	93.74 32	95.36 37

gg-mvs-nofinetune		77.18 170	74.31 187	85.80 72	91.42 99	68.36 55	71.78 318	94.72 28	49.61 316	77.12 98	45.92 339	77.41 3	93.98 185	67.62 162	93.16 40	95.05 53

test_part1								94.26 41				77.03 4			95.18 9	96.11 19

ESAPD		89.08 8	89.53 8	87.72 20	96.29 7	68.16 61	94.96 31	94.26 41	68.52 213	90.78 4	97.23 2	77.03 4	98.90 7	91.52 6	95.18 9	96.11 19

TSAR-MVS + GP.		87.96 14	88.37 12	86.70 43	93.51 44	65.32 141	95.15 25	93.84 46	78.17 55	85.93 25	94.80 51	75.80 6	98.21 24	89.38 13	88.78 80	96.59 10

DeepPCF-MVS	81.17 1	89.72 5	91.38 3	84.72 105	93.00 55	58.16 260	96.72 3	94.41 37	86.50 5	90.25 6	97.83 1	75.46 7	98.67 14	92.78 2	95.49 8	97.32 1

MVSTER		82.47 81	82.05 76	83.74 122	92.68 63	69.01 41	91.90 124	93.21 80	79.83 32	72.14 143	85.71 186	74.72 8	94.72 145	75.72 99	72.49 196	87.50 198

DELS-MVS		90.05 4	90.09 5	89.94 2	93.14 52	73.88 6	97.01 2	94.40 38	88.32 2	85.71 27	94.91 48	74.11 9	98.91 6	87.26 29	95.94 4	97.03 5
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
CSCG		86.87 29	86.26 32	88.72 9	95.05 20	70.79 17	93.83 59	95.33 15	68.48 216	77.63 91	94.35 62	73.04 10	98.45 18	84.92 44	93.71 33	96.92 6

MCST-MVS		91.08 1	91.46 2	89.94 2	97.66 2	73.37 7	97.13 1	95.58 13	89.33 1	85.77 26	96.26 10	72.84 11	99.38 1	92.64 4	95.93 5	97.08 4

TSAR-MVS + MP.		88.11 12	88.64 10	86.54 49	91.73 87	68.04 64	90.36 177	93.55 59	82.89 14	91.29 2	92.89 90	72.27 12	96.03 103	87.99 23	94.77 16	95.54 34

EPP-MVSNet		81.79 92	81.52 83	82.61 145	88.77 152	60.21 236	93.02 79	93.66 55	68.52 213	72.90 130	90.39 123	72.19 13	94.96 135	74.93 108	79.29 143	92.67 127

CostFormer		82.33 83	81.15 86	85.86 69	89.01 147	68.46 52	82.39 278	93.01 90	75.59 84	80.25 66	81.57 231	72.03 14	94.96 135	79.06 81	77.48 162	94.16 84

HPM-MVS++		89.37 7	89.95 7	87.64 21	95.10 19	68.23 60	95.24 22	94.49 34	82.43 17	88.90 11	96.35 8	71.89 15	98.63 15	88.76 21	96.40 2	96.06 21

CNVR-MVS		90.32 3	90.89 4	88.61 11	96.76 4	70.65 18	96.47 6	94.83 24	84.83 9	89.07 10	96.80 4	70.86 16	99.06 3	92.64 4	95.71 6	96.12 18

DWT-MVSNet_test		83.95 62	82.80 69	87.41 27	92.90 57	70.07 25	89.12 204	94.42 36	82.15 20	77.64 90	91.77 106	70.81 17	96.22 95	65.03 186	81.36 131	95.94 25

IB-MVS	77.80 4	82.18 85	80.46 96	87.35 29	89.14 144	70.28 22	95.59 17	95.17 17	78.85 47	70.19 163	85.82 184	70.66 18	97.67 36	72.19 125	66.52 238	94.09 89
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
agg_prior1		87.02 26	87.26 25	86.28 62	94.16 33	66.97 89	94.08 44	93.31 76	71.85 161	84.49 38	95.39 27	68.91 19	96.75 86	88.84 20	94.32 23	95.13 50

PatchFormer-LS_test		83.14 72	81.81 80	87.12 34	92.34 67	69.92 28	88.64 211	93.32 75	82.07 23	74.87 115	91.62 110	68.91 19	96.08 102	66.07 177	78.45 151	95.37 36

alignmvs		87.28 22	86.97 27	88.24 15	91.30 102	71.14 16	95.61 16	93.56 58	79.30 38	87.07 20	95.25 35	68.43 21	96.93 80	87.87 24	84.33 115	96.65 8

PAPM		85.89 41	85.46 43	87.18 32	88.20 164	72.42 9	92.41 100	92.77 98	82.11 21	80.34 65	93.07 84	68.27 22	95.02 133	78.39 86	93.59 35	94.09 89

train_agg		87.21 24	87.42 23	86.60 46	94.18 29	67.28 80	94.16 38	93.51 60	71.87 159	85.52 29	95.33 29	68.19 23	97.27 57	89.09 16	94.90 13	95.25 46

test_8					94.19 28	67.19 82	94.15 40	93.42 71	71.87 159	85.38 31	95.35 28	68.19 23	96.95 77

TEST9					94.18 29	67.28 80	94.16 38	93.51 60	71.75 167	85.52 29	95.33 29	68.01 25	97.27 57

test_prior3		87.38 21	87.70 18	86.42 55	94.71 23	67.35 78	95.10 27	93.10 88	75.40 89	85.25 33	95.61 24	67.94 26	96.84 82	87.47 26	94.77 16	95.05 53

test_prior2							95.10 27		75.40 89	85.25 33	95.61 24	67.94 26		87.47 26	94.77 16

WTY-MVS		86.32 35	85.81 38	87.85 17	92.82 60	69.37 37	95.20 23	95.25 16	82.71 15	81.91 54	94.73 52	67.93 28	97.63 41	79.55 76	82.25 127	96.54 12

Regformer-1		87.24 23	87.60 20	86.15 64	95.14 17	65.83 134	93.95 51	95.12 18	82.11 21	84.25 40	95.73 20	67.88 29	98.35 22	85.60 39	88.64 81	94.26 78

APDe-MVS		87.54 19	87.84 16	86.65 44	96.07 11	66.30 124	94.84 34	93.78 48	69.35 201	88.39 13	96.34 9	67.74 30	97.66 39	90.62 11	93.44 37	96.01 24

tpm2		79.80 121	77.95 130	85.34 89	88.28 162	68.26 59	81.56 286	91.42 149	70.11 194	77.59 93	80.50 247	67.40 31	94.26 169	67.34 164	77.35 163	93.51 105

tpmp4_e23		78.85 136	76.55 153	85.77 74	89.25 140	68.39 54	81.63 285	91.38 151	70.40 191	75.21 113	79.22 263	67.37 32	94.79 140	58.98 230	75.51 174	94.13 86

Regformer-2		87.00 27	87.43 22	85.71 78	95.14 17	64.73 153	93.95 51	94.95 21	81.69 26	84.03 44	95.73 20	67.35 33	98.19 26	85.40 41	88.64 81	94.20 80

tfpn1000		75.25 199	74.00 193	79.03 237	90.30 116	57.56 268	88.55 212	93.36 74	64.14 256	65.17 225	89.76 139	67.06 34	91.46 258	34.54 323	73.09 191	88.06 190

HY-MVS	76.49 5	84.28 57	83.36 63	87.02 38	92.22 72	67.74 69	84.65 260	94.50 33	79.15 42	82.23 52	87.93 157	66.88 35	96.94 78	80.53 72	82.20 128	96.39 14

EPNet		87.84 17	88.38 11	86.23 63	93.30 46	66.05 128	95.26 21	94.84 23	87.09 3	88.06 14	94.53 55	66.79 36	97.34 53	83.89 51	91.68 57	95.29 41
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
tfpn_ndepth		76.45 183	75.22 177	80.14 206	90.97 106	58.92 254	90.11 182	93.24 79	65.96 236	67.37 209	90.52 121	66.67 37	92.29 232	37.71 308	74.44 181	89.21 170

FIs		79.47 127	79.41 110	79.67 218	85.95 195	59.40 248	91.68 136	93.94 44	78.06 56	68.96 184	88.28 150	66.61 38	91.77 242	66.20 176	74.99 179	87.82 195

NCCC		89.07 9	89.46 9	87.91 16	96.60 5	69.05 40	96.38 7	94.64 31	84.42 10	86.74 21	96.20 11	66.56 39	98.76 13	89.03 19	94.56 21	95.92 27

SD-MVS		87.49 20	87.49 21	87.50 26	93.60 41	68.82 46	93.90 55	92.63 105	76.86 71	87.90 15	95.76 19	66.17 40	97.63 41	89.06 18	91.48 61	96.05 22

agg_prior3		86.93 28	87.08 26	86.48 52	94.21 27	66.95 91	94.14 41	93.40 72	71.80 164	84.86 35	95.13 39	66.16 41	97.25 59	89.09 16	94.90 13	95.25 46

UniMVSNet_NR-MVSNet		78.15 152	77.55 136	79.98 211	84.46 212	60.26 234	92.25 102	93.20 82	77.50 64	68.88 185	86.61 175	66.10 42	92.13 235	66.38 173	62.55 264	87.54 197

CHOSEN 280x420		77.35 165	76.95 149	78.55 246	87.07 180	62.68 201	69.71 324	82.95 305	68.80 207	71.48 151	87.27 172	66.03 43	84.00 313	76.47 97	82.81 125	88.95 171

CANet		89.61 6	89.99 6	88.46 13	94.39 26	69.71 32	96.53 5	93.78 48	86.89 4	89.68 7	95.78 18	65.94 44	99.10 2	92.99 1	93.91 28	96.58 11

segment_acmp												65.94 44

Regformer-3		85.80 42	85.92 36	85.46 82	94.17 31	65.09 149	92.95 81	95.11 19	81.13 27	81.68 56	95.04 40	65.82 46	98.32 23	83.02 55	84.36 112	92.97 121

Vis-MVSNet (Re-imp)		79.24 130	79.57 105	78.24 252	88.46 158	52.29 298	90.41 176	89.12 227	74.24 106	69.13 180	91.91 104	65.77 47	90.09 274	59.00 229	88.09 85	92.33 134

FC-MVSNet-test		77.99 155	78.08 128	77.70 257	84.89 206	55.51 285	90.27 179	93.75 52	76.87 70	66.80 215	87.59 163	65.71 48	90.23 268	62.89 206	73.94 185	87.37 206

test12				87.09 36	94.60 25	68.86 44		92.91 94		82.67 51		65.44 49	97.55 44		93.69 34	94.84 61

旧先验1					91.94 77	60.74 226		91.50 146			94.36 58	65.23 50			91.84 54	94.55 69

SMA-MVS		87.99 13	88.11 14	87.62 24	93.21 49	68.55 50	93.85 57	93.82 47	74.24 106	90.84 3	96.67 5	65.20 51	98.42 21	89.24 15	95.96 3	95.88 28

Regformer-4		85.45 45	85.69 41	84.73 103	94.17 31	63.23 186	92.95 81	94.83 24	80.66 29	81.29 58	95.04 40	65.12 52	98.08 29	82.74 56	84.36 112	92.88 125

conf0.01		74.95 205	73.61 198	78.96 238	89.65 127	56.94 274	87.72 228	93.45 63	65.14 244	65.68 217	89.99 130	65.09 53	91.67 244	35.16 315	70.61 207	88.27 184

conf0.002		74.95 205	73.61 198	78.96 238	89.65 127	56.94 274	87.72 228	93.45 63	65.14 244	65.68 217	89.99 130	65.09 53	91.67 244	35.16 315	70.61 207	88.27 184

thresconf0.02		74.92 208	73.61 198	78.85 241	89.65 127	56.94 274	87.72 228	93.45 63	65.14 244	65.68 217	89.99 130	65.09 53	91.67 244	35.16 315	70.61 207	87.94 191

tfpn_n400		74.92 208	73.61 198	78.85 241	89.65 127	56.94 274	87.72 228	93.45 63	65.14 244	65.68 217	89.99 130	65.09 53	91.67 244	35.16 315	70.61 207	87.94 191

tfpnconf		74.92 208	73.61 198	78.85 241	89.65 127	56.94 274	87.72 228	93.45 63	65.14 244	65.68 217	89.99 130	65.09 53	91.67 244	35.16 315	70.61 207	87.94 191

tfpnview11		74.92 208	73.61 198	78.85 241	89.65 127	56.94 274	87.72 228	93.45 63	65.14 244	65.68 217	89.99 130	65.09 53	91.67 244	35.16 315	70.61 207	87.94 191

1112_ss		80.56 105	79.83 102	82.77 139	88.65 153	60.78 223	92.29 101	88.36 245	72.58 140	72.46 139	94.95 43	65.09 53	93.42 201	66.38 173	77.71 154	94.10 88

MVSFormer		83.75 67	82.88 67	86.37 58	89.24 142	71.18 14	89.07 205	90.69 170	65.80 237	87.13 18	94.34 63	64.99 60	92.67 220	72.83 116	91.80 55	95.27 43

lupinMVS		87.74 18	87.77 17	87.63 23	89.24 142	71.18 14	96.57 4	92.90 95	82.70 16	87.13 18	95.27 33	64.99 60	95.80 109	89.34 14	91.80 55	95.93 26

tpmrst		80.57 104	79.14 117	84.84 101	90.10 119	68.28 58	81.70 282	89.72 209	77.63 62	75.96 105	79.54 261	64.94 62	92.71 218	75.43 101	77.28 165	93.55 104

Test_1112_low_res		79.56 126	78.60 121	82.43 151	88.24 163	60.39 232	92.09 108	87.99 252	72.10 155	71.84 146	87.42 166	64.62 63	93.04 205	65.80 181	77.30 164	93.85 100

DeepC-MVS_fast	79.48 2	87.95 15	88.00 15	87.79 19	95.86 14	68.32 56	95.74 12	94.11 43	83.82 12	83.49 47	96.19 12	64.53 64	98.44 19	83.42 54	94.88 15	96.61 9
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
MG-MVS		87.11 25	86.27 31	89.62 5	97.79 1	76.27 3	94.96 31	94.49 34	78.74 51	83.87 46	92.94 87	64.34 65	96.94 78	75.19 103	94.09 25	95.66 30

tpm		78.58 145	77.03 145	83.22 133	85.94 197	64.56 154	83.21 273	91.14 159	78.31 54	73.67 125	79.68 258	64.01 66	92.09 237	66.07 177	71.26 205	93.03 119

CDS-MVSNet		81.43 95	80.74 91	83.52 129	86.26 190	64.45 158	92.09 108	90.65 173	75.83 83	73.95 124	89.81 137	63.97 67	92.91 212	71.27 132	82.82 124	93.20 114
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
MVS_Test		84.16 59	83.20 64	87.05 37	91.56 91	69.82 29	89.99 186	92.05 125	77.77 59	82.84 50	86.57 176	63.93 68	96.09 100	74.91 109	89.18 78	95.25 46

APD-MVS		85.93 40	85.99 35	85.76 75	95.98 13	65.21 143	93.59 64	92.58 107	66.54 231	86.17 22	95.88 17	63.83 69	97.00 70	86.39 35	92.94 41	95.06 52
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
mvs_anonymous		81.36 96	79.99 99	85.46 82	90.39 115	68.40 53	86.88 248	90.61 174	74.41 100	70.31 162	84.67 195	63.79 70	92.32 231	73.13 113	85.70 103	95.67 29

PVSNet_Blended_VisFu		83.97 61	83.50 56	85.39 87	90.02 120	66.59 114	93.77 60	91.73 136	77.43 66	77.08 100	89.81 137	63.77 71	96.97 75	79.67 75	88.21 84	92.60 129

CDPH-MVS		85.71 43	85.46 43	86.46 53	94.75 22	67.19 82	93.89 56	92.83 97	70.90 180	83.09 49	95.28 32	63.62 72	97.36 51	80.63 71	94.18 24	94.84 61

HyFIR lowres test		81.03 100	79.56 106	85.43 85	87.81 171	68.11 63	90.18 181	90.01 198	70.65 188	72.95 129	86.06 182	63.61 73	94.50 153	75.01 107	79.75 139	93.67 101

canonicalmvs		86.85 30	86.25 33	88.66 10	91.80 86	71.92 10	93.54 66	91.71 138	80.26 31	87.55 16	95.25 35	63.59 74	96.93 80	88.18 22	84.34 114	97.11 3

SteuartSystems-ACMMP		86.82 32	86.90 28	86.58 48	90.42 113	66.38 121	96.09 9	93.87 45	77.73 60	84.01 45	95.66 22	63.39 75	97.94 30	87.40 28	93.55 36	95.42 35
Skip Steuart: Steuart Systems R&D Blog.
EI-MVSNet-Vis-set		83.77 66	83.67 54	84.06 117	92.79 62	63.56 183	91.76 132	94.81 26	79.65 36	77.87 87	94.09 67	63.35 76	97.90 32	79.35 77	79.36 141	90.74 155

UniMVSNet (Re)		77.58 159	76.78 150	79.98 211	84.11 218	60.80 222	91.76 132	93.17 85	76.56 77	69.93 169	84.78 194	63.32 77	92.36 230	64.89 187	62.51 266	86.78 218

PVSNet_BlendedMVS		83.38 69	83.43 59	83.22 133	93.76 37	67.53 75	94.06 45	93.61 56	79.13 43	81.00 60	85.14 190	63.19 78	97.29 55	87.08 30	73.91 186	84.83 253

PVSNet_Blended		86.73 33	86.86 29	86.31 61	93.76 37	67.53 75	96.33 8	93.61 56	82.34 18	81.00 60	93.08 82	63.19 78	97.29 55	87.08 30	91.38 62	94.13 86

PAPM_NR		82.97 75	81.84 79	86.37 58	94.10 35	66.76 103	87.66 235	92.84 96	69.96 196	74.07 122	93.57 75	63.10 80	97.50 46	70.66 140	90.58 71	94.85 60

nrg030		80.93 101	79.86 101	84.13 116	83.69 222	68.83 45	93.23 73	91.20 156	75.55 85	75.06 114	88.22 155	63.04 81	94.74 144	81.88 62	66.88 235	88.82 174

EI-MVSNet-UG-set		83.14 72	82.96 65	83.67 127	92.28 70	63.19 190	91.38 146	94.68 29	79.22 40	76.60 102	93.75 72	62.64 82	97.76 34	78.07 88	78.01 152	90.05 162

DeepC-MVS	77.85 3	85.52 44	85.24 45	86.37 58	88.80 151	66.64 111	92.15 104	93.68 54	81.07 28	76.91 101	93.64 74	62.59 83	98.44 19	85.50 40	92.84 43	94.03 93
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
MVS_111021_HR		86.19 38	85.80 39	87.37 28	93.17 51	69.79 30	93.99 49	93.76 51	79.08 45	78.88 79	93.99 69	62.25 84	98.15 27	85.93 38	91.15 65	94.15 85

PHI-MVS		86.83 31	86.85 30	86.78 42	93.47 45	65.55 138	95.39 20	95.10 20	71.77 166	85.69 28	96.52 6	62.07 85	98.77 12	86.06 37	95.60 7	96.03 23

MP-MVS		85.02 48	84.97 47	85.17 94	92.60 64	64.27 168	93.24 72	92.27 114	73.13 131	79.63 72	94.43 56	61.90 86	97.17 61	85.00 43	92.56 46	94.06 92

jason		86.40 34	86.17 34	87.11 35	86.16 192	70.54 20	95.71 15	92.19 122	82.00 24	84.58 37	94.34 63	61.86 87	95.53 125	87.76 25	90.89 67	95.27 43
jason: jason.
PAPR		85.15 47	84.47 49	87.18 32	96.02 12	68.29 57	91.85 127	93.00 92	76.59 76	79.03 78	95.00 42	61.59 88	97.61 43	78.16 87	89.00 79	95.63 31

IS-MVSNet		80.14 113	79.41 110	82.33 158	87.91 169	60.08 240	91.97 116	88.27 248	72.90 136	71.44 152	91.73 109	61.44 89	93.66 196	62.47 211	86.53 98	93.24 112

EI-MVSNet		78.97 134	78.22 126	81.25 185	85.33 200	62.73 200	89.53 197	93.21 80	72.39 144	72.14 143	90.13 127	60.99 90	94.72 145	67.73 161	72.49 196	86.29 224

IterMVS-LS		76.49 181	75.18 178	80.43 202	84.49 211	62.74 199	90.64 171	88.80 236	72.40 143	65.16 226	81.72 230	60.98 91	92.27 233	67.74 160	64.65 256	86.29 224
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
view600		76.93 172	75.50 172	81.23 186	91.44 95	62.00 208	89.94 187	96.56 1	70.68 184	68.54 191	87.31 167	60.79 92	94.19 170	38.90 302	75.31 175	87.48 199

view800		76.93 172	75.50 172	81.23 186	91.44 95	62.00 208	89.94 187	96.56 1	70.68 184	68.54 191	87.31 167	60.79 92	94.19 170	38.90 302	75.31 175	87.48 199

conf0.05thres1000		76.93 172	75.50 172	81.23 186	91.44 95	62.00 208	89.94 187	96.56 1	70.68 184	68.54 191	87.31 167	60.79 92	94.19 170	38.90 302	75.31 175	87.48 199

tfpn		76.93 172	75.50 172	81.23 186	91.44 95	62.00 208	89.94 187	96.56 1	70.68 184	68.54 191	87.31 167	60.79 92	94.19 170	38.90 302	75.31 175	87.48 199

tpm cat1		75.30 198	72.21 219	84.58 108	88.52 155	67.77 68	78.16 310	88.02 251	61.88 273	68.45 196	76.37 281	60.65 96	94.03 183	53.77 247	74.11 183	91.93 141

TAMVS		80.37 108	79.45 109	83.13 135	85.14 203	63.37 184	91.23 153	90.76 169	74.81 98	72.65 133	88.49 146	60.63 97	92.95 208	69.41 149	81.95 130	93.08 118

tfpn111		78.00 153	76.62 152	82.13 168	91.89 81	63.21 187	91.19 157	96.33 5	72.28 146	70.45 157	87.89 158	60.31 98	94.91 139	42.61 289	76.64 167	88.27 184

conf200view11		78.32 150	77.01 146	82.27 161	91.89 81	63.21 187	91.19 157	96.33 5	72.28 146	70.45 157	87.89 158	60.31 98	95.32 128	45.16 276	77.58 157	88.27 184

thres100view900		78.37 148	77.01 146	82.46 147	91.89 81	63.21 187	91.19 157	96.33 5	72.28 146	70.45 157	87.89 158	60.31 98	95.32 128	45.16 276	77.58 157	88.83 172

thres600view7		78.00 153	76.66 151	82.03 174	91.93 78	63.69 179	91.30 151	96.33 5	72.43 142	70.46 156	87.89 158	60.31 98	94.92 138	42.64 288	76.64 167	87.48 199

CHOSEN 1792x2688		84.98 49	83.45 58	89.57 6	89.94 122	75.14 4	92.07 110	92.32 112	81.87 25	75.68 106	88.27 151	60.18 102	98.60 16	80.46 73	90.27 74	94.96 58

tfpn200view9		78.79 139	77.43 139	82.88 137	92.21 73	64.49 155	92.05 111	96.28 10	73.48 126	71.75 148	88.26 152	60.07 103	95.32 128	45.16 276	77.58 157	88.83 172

thres400		78.68 142	77.43 139	82.43 151	92.21 73	64.49 155	92.05 111	96.28 10	73.48 126	71.75 148	88.26 152	60.07 103	95.32 128	45.16 276	77.58 157	87.48 199

MVS		84.66 53	82.86 68	90.06 1	90.93 107	74.56 5	87.91 224	95.54 14	68.55 212	72.35 142	94.71 53	59.78 105	98.90 7	81.29 70	94.69 20	96.74 7

thres200		79.66 123	78.33 123	83.66 128	92.54 65	65.82 135	93.06 77	96.31 9	74.90 97	73.30 127	88.66 144	59.67 106	95.61 119	47.84 267	78.67 148	89.56 168

Effi-MVS+		83.82 65	82.76 70	86.99 39	89.56 135	69.40 36	91.35 148	86.12 279	72.59 139	83.22 48	92.81 92	59.60 107	96.01 105	81.76 63	87.80 87	95.56 33

ACMMP_Plus		86.05 39	85.80 39	86.80 41	91.58 90	67.53 75	91.79 129	93.49 62	74.93 96	84.61 36	95.30 31	59.42 108	97.92 31	86.13 36	94.92 12	94.94 59

UA-Net		80.02 116	79.65 104	81.11 192	89.33 138	57.72 264	86.33 253	89.00 233	77.44 65	81.01 59	89.15 141	59.33 109	95.90 106	61.01 218	84.28 117	89.73 166

NR-MVSNet		76.05 188	74.59 182	80.44 201	82.96 231	62.18 207	90.83 165	91.73 136	77.12 67	60.96 252	86.35 177	59.28 110	91.80 241	60.74 219	61.34 277	87.35 208

MP-MVS-pluss		85.24 46	85.13 46	85.56 79	91.42 99	65.59 137	91.54 141	92.51 109	74.56 99	80.62 62	95.64 23	59.15 111	97.00 70	86.94 32	93.80 29	94.07 91
MP-MVS-pluss: MP-MVS-pluss. MP-MVS-pluss
HFP-MVS		84.73 51	84.40 51	85.72 76	93.75 39	65.01 150	93.50 67	93.19 83	72.19 151	79.22 75	94.93 45	59.04 112	97.67 36	81.55 65	92.21 49	94.49 75

#test#		84.98 49	84.74 48	85.72 76	93.75 39	65.01 150	94.09 43	93.19 83	73.55 125	79.22 75	94.93 45	59.04 112	97.67 36	82.66 57	92.21 49	94.49 75

MSLP-MVS++		86.27 36	85.91 37	87.35 29	92.01 76	68.97 43	95.04 29	92.70 100	79.04 46	81.50 57	96.50 7	58.98 114	96.78 84	83.49 53	93.93 27	96.29 16

Patchmatch-test		65.86 278	60.94 289	80.62 200	83.75 221	58.83 255	58.91 341	75.26 327	44.50 331	50.95 305	77.09 279	58.81 115	87.90 295	35.13 321	64.03 259	95.12 51

EPNet_dtu		78.80 138	79.26 114	77.43 262	88.06 166	49.71 311	91.96 117	91.95 130	77.67 61	76.56 103	91.28 112	58.51 116	90.20 269	56.37 237	80.95 134	92.39 133
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
VNet		86.20 37	85.65 42	87.84 18	93.92 36	69.99 26	95.73 14	95.94 12	78.43 53	86.00 24	93.07 84	58.22 117	97.00 70	85.22 42	84.33 115	96.52 13

TESTMET0.1,1		82.41 82	81.98 78	83.72 125	88.08 165	63.74 176	92.70 89	93.77 50	79.30 38	77.61 92	87.57 164	58.19 118	94.08 178	73.91 112	86.68 95	93.33 110

原ACMM1				84.42 112	93.21 49	64.27 168		93.40 72	65.39 240	79.51 73	92.50 94	58.11 119	96.69 88	65.27 185	93.96 26	92.32 135

sam_mvs1												57.85 120				94.68 65

CR-MVSNet		73.79 220	70.82 229	82.70 141	83.15 229	67.96 65	70.25 321	84.00 295	73.67 123	69.97 167	72.41 300	57.82 121	89.48 283	52.99 251	73.13 189	90.64 157

Patchmtry		67.53 270	63.93 274	78.34 247	82.12 236	64.38 162	68.72 325	84.00 295	48.23 321	59.24 259	72.41 300	57.82 121	89.27 285	46.10 273	56.68 296	81.36 290

patchmatchnet-post											67.62 318	57.62 123	90.25 265

PCF-MVS	73.15 9	79.29 129	77.63 135	84.29 114	86.06 193	65.96 130	87.03 243	91.10 160	69.86 197	69.79 170	90.64 117	57.54 124	96.59 89	64.37 193	82.29 126	90.32 159
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
Fast-Effi-MVS+		81.14 98	80.01 98	84.51 111	90.24 118	65.86 132	94.12 42	89.15 226	73.81 119	75.37 112	88.26 152	57.26 125	94.53 152	66.97 168	84.92 107	93.15 115

1121		81.25 97	80.05 97	84.87 100	92.30 69	64.31 165	87.91 224	91.39 150	59.44 286	79.94 68	92.91 88	57.09 126	97.01 68	66.63 169	92.81 44	93.29 111

PatchT		69.11 260	65.37 265	80.32 203	82.07 237	63.68 180	67.96 330	87.62 256	50.86 314	69.37 177	65.18 324	57.09 126	88.53 291	41.59 292	66.60 237	88.74 175

testdata				81.34 184	89.02 146	57.72 264		89.84 202	58.65 290	85.32 32	94.09 67	57.03 128	93.28 202	69.34 150	90.56 72	93.03 119

PatchmatchNet		77.46 160	74.63 181	85.96 67	89.55 136	70.35 21	79.97 299	89.55 212	72.23 149	70.94 153	76.91 280	57.03 128	92.79 216	54.27 244	81.17 132	94.74 63
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
region2R		84.36 55	84.03 53	85.36 88	93.54 43	64.31 165	93.43 70	92.95 93	72.16 154	78.86 80	94.84 50	56.97 130	97.53 45	81.38 68	92.11 53	94.24 79

新几何1				84.73 103	92.32 68	64.28 167		91.46 148	59.56 285	79.77 70	92.90 89	56.95 131	96.57 91	63.40 198	92.91 42	93.34 108

diffmvs		80.18 111	78.55 122	85.07 95	88.56 154	66.93 92	86.70 251	88.62 240	70.42 190	78.69 83	85.26 188	56.93 132	94.77 141	68.68 156	83.09 121	93.51 105

MVS_0304		88.39 10	88.35 13	88.50 12	93.01 54	70.11 23	95.90 10	92.20 120	86.27 6	88.70 12	95.92 16	56.76 133	99.02 4	92.68 3	93.76 31	96.37 15

WR-MVS		76.76 179	75.74 164	79.82 216	84.60 208	62.27 206	92.60 94	92.51 109	76.06 80	67.87 202	85.34 187	56.76 133	90.24 267	62.20 212	63.69 263	86.94 216

HPM-MVS		83.25 70	82.95 66	84.17 115	92.25 71	62.88 197	90.91 163	91.86 132	70.30 193	77.12 98	93.96 70	56.75 135	96.28 94	82.04 61	91.34 64	93.34 108
Chunlin Ren, Qingshan Xu, Shikun Zhang, Jiaqi Yang: Hierarchical Prior Mining for Non-local Multi-View Stereo. ICCV 2023
sss		82.71 79	82.38 74	83.73 124	89.25 140	59.58 246	92.24 103	94.89 22	77.96 57	79.86 69	92.38 98	56.70 136	97.05 65	77.26 93	80.86 135	94.55 69

ACMMPR		84.37 54	84.06 52	85.28 90	93.56 42	64.37 163	93.50 67	93.15 86	72.19 151	78.85 81	94.86 49	56.69 137	97.45 47	81.55 65	92.20 51	94.02 94

FMVSNet3		77.73 158	76.04 159	82.80 138	91.20 104	68.99 42	91.87 125	91.99 127	73.35 129	67.04 211	83.19 208	56.62 138	92.14 234	59.80 225	69.34 218	87.28 210

Patchmatch-RL test		68.17 265	64.49 271	79.19 233	71.22 319	53.93 292	70.07 323	71.54 336	69.22 202	56.79 275	62.89 328	56.58 139	88.61 288	69.53 148	52.61 305	95.03 56

test_post											23.01 348	56.49 140	92.67 220

RPMNet		69.58 256	65.21 266	82.70 141	83.15 229	67.96 65	70.25 321	86.15 278	46.83 324	69.97 167	65.10 325	56.48 141	89.48 283	35.79 314	73.13 189	90.64 157

DU-MVS		76.86 176	75.84 162	79.91 213	82.96 231	60.26 234	91.26 152	91.54 143	76.46 78	68.88 185	86.35 177	56.16 142	92.13 235	66.38 173	62.55 264	87.35 208

Baseline_NR-MVSNet		73.99 218	72.83 211	77.48 261	80.78 247	59.29 251	91.79 129	84.55 290	68.85 206	68.99 183	80.70 244	56.16 142	92.04 238	62.67 209	60.98 279	81.11 296

API-MVS		82.28 84	80.53 95	87.54 25	96.13 10	70.59 19	93.63 62	91.04 163	65.72 239	75.45 111	92.83 91	56.11 144	98.89 10	64.10 194	89.75 77	93.15 115

zzz-MVS		84.73 51	84.47 49	85.50 80	91.89 81	65.16 144	91.55 140	92.23 115	75.32 91	80.53 63	95.21 37	56.06 145	97.16 62	84.86 45	92.55 47	94.18 81

MTAPA		83.91 63	83.38 62	85.50 80	91.89 81	65.16 144	81.75 281	92.23 115	75.32 91	80.53 63	95.21 37	56.06 145	97.16 62	84.86 45	92.55 47	94.18 81

JIA-IIPM		66.06 277	62.45 283	76.88 268	81.42 243	54.45 291	57.49 342	88.67 238	49.36 317	63.86 236	46.86 338	56.06 145	90.25 265	49.53 261	68.83 222	85.95 235

v148		76.19 185	74.47 186	81.36 183	80.05 275	64.44 159	91.75 134	90.23 187	73.68 122	67.13 210	80.84 243	55.92 148	93.86 192	68.95 153	61.73 273	85.76 241

WR-MVS_H		70.59 247	69.94 231	72.53 294	81.03 244	51.43 302	87.35 240	92.03 126	67.38 227	60.23 255	80.70 244	55.84 149	83.45 317	46.33 272	58.58 290	82.72 277

XVS		83.87 64	83.47 57	85.05 96	93.22 47	63.78 174	92.92 83	92.66 103	73.99 112	78.18 85	94.31 65	55.25 150	97.41 48	79.16 79	91.58 59	93.95 96

X-MVStestdata		76.86 176	74.13 190	85.05 96	93.22 47	63.78 174	92.92 83	92.66 103	73.99 112	78.18 85	10.19 354	55.25 150	97.41 48	79.16 79	91.58 59	93.95 96

BH-w/o		80.49 107	79.30 113	84.05 118	90.83 111	64.36 164	93.60 63	89.42 216	74.35 105	69.09 181	90.15 126	55.23 152	95.61 119	64.61 189	86.43 100	92.17 139

CP-MVS		83.71 68	83.40 61	84.65 106	93.14 52	63.84 172	94.59 35	92.28 113	71.03 178	77.41 94	94.92 47	55.21 153	96.19 96	81.32 69	90.70 69	93.91 98

PGM-MVS		83.25 70	82.70 72	84.92 98	92.81 61	64.07 170	90.44 174	92.20 120	71.28 176	77.23 97	94.43 56	55.17 154	97.31 54	79.33 78	91.38 62	93.37 107

v1		77.29 166	75.57 169	82.42 154	80.61 261	66.73 104	91.96 117	90.42 178	74.41 100	69.46 174	82.12 222	55.14 155	94.40 159	71.00 133	65.04 249	86.13 227

divwei89l23v2f112		77.28 167	75.57 169	82.42 154	80.62 258	66.72 106	91.96 117	90.42 178	74.41 100	69.46 174	82.12 222	55.11 156	94.40 159	71.00 133	65.04 249	86.12 228

v1141		77.28 167	75.57 169	82.42 154	80.63 257	66.73 104	91.96 117	90.42 178	74.41 100	69.46 174	82.12 222	55.09 157	94.40 159	70.99 135	65.05 248	86.12 228

tpmvs		72.88 226	69.76 234	82.22 165	90.98 105	67.05 87	78.22 309	88.30 246	63.10 264	64.35 234	74.98 288	55.09 157	94.27 167	43.25 282	69.57 217	85.34 249

v18		71.94 232	69.43 235	79.50 224	80.74 248	66.82 96	88.16 218	86.66 265	68.95 205	55.55 278	72.66 295	55.03 159	90.15 270	64.78 188	52.30 306	81.54 284

v6		77.39 162	75.71 165	82.44 148	80.67 253	66.82 96	91.94 121	90.18 190	74.19 108	69.60 171	82.50 215	55.00 160	94.44 154	71.68 128	65.60 241	86.05 231

v1neww		77.39 162	75.71 165	82.44 148	80.69 251	66.83 94	91.94 121	90.18 190	74.19 108	69.60 171	82.51 212	54.99 161	94.44 154	71.68 128	65.60 241	86.05 231

v7new		77.39 162	75.71 165	82.44 148	80.69 251	66.83 94	91.94 121	90.18 190	74.19 108	69.60 171	82.51 212	54.99 161	94.44 154	71.68 128	65.60 241	86.05 231

v8		75.35 197	73.26 208	81.61 181	80.67 253	66.82 96	89.54 196	89.27 220	71.65 168	63.30 242	80.30 251	54.99 161	94.06 179	67.33 165	62.33 267	83.94 259

sam_mvs												54.91 164

v16		71.81 233	69.26 237	79.47 225	80.66 255	66.81 100	87.93 222	86.63 267	68.70 210	55.35 279	72.51 296	54.75 165	90.12 272	64.51 190	52.28 307	81.47 285

v17		71.77 235	69.20 238	79.46 226	80.62 258	66.81 100	87.93 222	86.63 267	68.71 209	55.25 280	72.49 297	54.72 166	90.11 273	64.50 191	51.97 308	81.47 285

EPMVS		78.49 147	75.98 160	86.02 65	91.21 103	69.68 33	80.23 295	91.20 156	75.25 93	72.48 138	78.11 268	54.65 167	93.69 195	57.66 235	83.04 122	94.69 64

v15		71.40 237	68.75 240	79.35 227	80.39 262	66.70 108	87.57 237	86.64 266	68.66 211	54.68 282	72.00 304	54.50 168	89.98 275	63.69 196	50.66 313	81.38 289

ab-mvs		80.18 111	78.31 124	85.80 72	88.44 159	65.49 140	83.00 275	92.67 102	71.82 163	77.36 95	85.01 191	54.50 168	96.59 89	76.35 98	75.63 173	95.32 40

DP-MVS Recon		82.73 77	81.65 82	85.98 66	97.31 3	67.06 86	95.15 25	91.99 127	69.08 204	76.50 104	93.89 71	54.48 170	98.20 25	70.76 139	85.66 104	92.69 126

XXY-MVS		77.94 156	76.44 155	82.43 151	82.60 233	64.44 159	92.01 113	91.83 134	73.59 124	70.00 166	85.82 184	54.43 171	94.76 142	69.63 146	68.02 229	88.10 189

MDTV_nov1_ep13_2view						59.90 242	80.13 297		67.65 224	72.79 131		54.33 172		59.83 224		92.58 130

V14		71.29 239	68.61 242	79.31 228	80.34 266	66.65 110	87.39 239	86.61 269	68.41 217	54.49 284	71.91 305	54.25 173	89.96 276	63.50 197	50.62 314	81.33 291

Test By Simon												54.21 174

V9		71.16 240	68.46 244	79.27 230	80.26 269	66.60 112	87.21 242	86.56 270	68.17 218	54.26 287	71.81 307	54.00 175	89.93 277	63.28 200	50.57 315	81.27 292

MAR-MVS		84.18 58	83.43 59	86.44 54	96.25 9	65.93 131	94.28 37	94.27 40	74.41 100	79.16 77	95.61 24	53.99 176	98.88 11	69.62 147	93.26 39	94.50 74
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
test-LLR		80.10 114	79.56 106	81.72 178	86.93 183	61.17 218	92.70 89	91.54 143	71.51 174	75.62 107	86.94 173	53.83 177	92.38 228	72.21 123	84.76 110	91.60 144

test0.0.03 1		72.76 227	72.71 213	72.88 292	80.25 270	47.99 316	91.22 154	89.45 214	71.51 174	62.51 249	87.66 162	53.83 177	85.06 306	50.16 258	67.84 232	85.58 243

v2v482		77.42 161	75.65 168	82.73 140	80.38 263	67.13 85	91.85 127	90.23 187	75.09 94	69.37 177	83.39 206	53.79 179	94.44 154	71.77 126	65.00 252	86.63 222

v12		71.02 244	68.29 249	79.22 232	80.18 272	66.53 117	87.01 245	86.54 272	67.90 220	54.00 290	71.70 308	53.66 180	89.91 278	63.09 202	50.51 316	81.21 293

pcd_1.5k_mvsjas		4.46 335	5.95 336	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	0.00 358	53.55 181	0.00 359	0.00 356	0.00 356	0.00 357

PS-MVSNAJss		77.26 169	76.31 156	80.13 208	80.64 256	59.16 252	90.63 173	91.06 161	72.80 137	68.58 190	84.57 197	53.55 181	93.96 186	72.97 114	71.96 199	87.27 211

PS-MVSNAJ		88.14 11	87.61 19	89.71 4	92.06 75	76.72 1	95.75 11	93.26 78	83.86 11	89.55 8	96.06 14	53.55 181	97.89 33	91.10 8	93.31 38	94.54 71

v13		70.90 245	68.15 250	79.15 236	80.08 273	66.45 119	86.83 249	86.50 273	67.62 226	53.78 292	71.61 310	53.51 184	89.87 280	62.89 206	50.50 317	81.14 295

mPP-MVS		82.96 76	82.44 73	84.52 110	92.83 58	62.92 195	92.76 86	91.85 133	71.52 173	75.61 109	94.24 66	53.48 185	96.99 73	78.97 82	90.73 68	93.64 103

xiu_mvs_v2_base		87.92 16	87.38 24	89.55 7	91.41 101	76.43 2	95.74 12	93.12 87	83.53 13	89.55 8	95.95 15	53.45 186	97.68 35	91.07 9	92.62 45	94.54 71

test_post1							78.95 303				20.70 352	53.05 187	91.50 257	60.43 221

MDTV_nov1_ep13			72.61 214		89.06 145	68.48 51	80.33 293	90.11 194	71.84 162	71.81 147	75.92 285	53.01 188	93.92 188	48.04 266	73.38 187

v11		71.05 243	68.32 247	79.23 231	80.34 266	66.57 115	87.01 245	86.55 271	68.11 219	54.40 285	71.66 309	52.94 189	89.91 278	62.71 208	51.12 311	81.21 293

test222					89.77 125	61.60 215	89.55 195	89.42 216	56.83 299	77.28 96	92.43 97	52.76 190			91.14 66	93.09 117

v1144		76.73 180	74.88 180	82.27 161	80.23 271	66.60 112	91.68 136	90.21 189	73.69 121	69.06 182	81.89 227	52.73 191	94.40 159	69.21 151	65.23 245	85.80 238

v7		76.83 178	75.01 179	82.29 160	80.35 264	66.70 108	91.68 136	89.97 199	73.47 128	69.22 179	82.22 219	52.52 192	94.43 158	69.73 145	65.96 240	85.74 242

v10		74.77 212	72.54 216	81.46 182	80.33 268	66.71 107	89.15 203	89.08 229	70.94 179	63.08 243	79.86 256	52.52 192	94.04 182	65.70 182	62.17 268	83.64 261

CLD-MVS		82.73 77	82.35 75	83.86 120	87.90 170	67.65 72	95.45 18	92.18 123	85.06 8	72.58 135	92.27 101	52.46 194	95.78 110	84.18 47	79.06 144	88.16 188
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
TranMVSNet+NR-MVSNet		75.86 191	74.52 185	79.89 214	82.44 234	60.64 229	91.37 147	91.37 152	76.63 75	67.65 204	86.21 180	52.37 195	91.55 252	61.84 214	60.81 280	87.48 199

VPA-MVSNet		79.03 132	78.00 129	82.11 172	85.95 195	64.48 157	93.22 74	94.66 30	75.05 95	74.04 123	84.95 192	52.17 196	93.52 198	74.90 110	67.04 234	88.32 183

APD-MVS_3200maxsize		81.64 93	81.32 85	82.59 146	92.36 66	58.74 257	91.39 144	91.01 164	63.35 260	79.72 71	94.62 54	51.82 197	96.14 98	79.71 74	87.93 86	92.89 124

dp		75.01 203	72.09 220	83.76 121	89.28 139	66.22 127	79.96 300	89.75 204	71.16 177	67.80 203	77.19 276	51.81 198	92.54 223	50.39 257	71.44 204	92.51 132

v144192		76.05 188	74.03 191	82.12 169	79.50 280	66.55 116	91.39 144	89.71 210	72.30 145	68.17 197	81.33 235	51.75 199	94.03 183	67.94 158	64.19 258	85.77 239

BH-untuned		78.68 142	77.08 144	83.48 131	89.84 124	63.74 176	92.70 89	88.59 241	71.57 171	66.83 214	88.65 145	51.75 199	95.39 126	59.03 228	84.77 109	91.32 150

HQP2-MVS												51.63 201

HQP-MVS		81.14 98	80.64 93	82.64 144	87.54 173	63.66 181	94.06 45	91.70 139	79.80 33	74.18 118	90.30 124	51.63 201	95.61 119	77.63 91	78.90 145	88.63 176

V42		76.46 182	74.55 184	82.19 166	79.14 285	67.82 67	90.26 180	89.42 216	73.75 120	68.63 189	81.89 227	51.31 203	94.09 177	71.69 127	64.84 253	84.66 254

TransMVSNet (Re)		70.07 251	67.66 252	77.31 265	80.62 258	59.13 253	91.78 131	84.94 288	65.97 235	60.08 256	80.44 248	50.78 204	91.87 240	48.84 263	45.46 326	80.94 298

HQP_MVS		80.34 109	79.75 103	82.12 169	86.94 181	62.42 202	93.13 75	91.31 153	78.81 49	72.53 136	89.14 142	50.66 205	95.55 123	76.74 94	78.53 149	88.39 181

plane_prior6					87.23 177	62.32 204						50.66 205

ACMMP		81.49 94	80.67 92	83.93 119	91.71 88	62.90 196	92.13 105	92.22 119	71.79 165	71.68 150	93.49 77	50.32 207	96.96 76	78.47 84	84.22 119	91.93 141
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
MVS_111021_LR		82.02 90	81.52 83	83.51 130	88.42 160	62.88 197	89.77 193	88.93 234	76.78 73	75.55 110	93.10 80	50.31 208	95.38 127	83.82 52	87.02 91	92.26 138

1314		80.70 103	78.95 118	85.94 68	87.77 172	67.56 74	87.91 224	92.55 108	72.17 153	67.44 205	93.09 81	50.27 209	97.04 67	71.68 128	87.64 88	93.23 113

CP-MVSNet		70.50 249	69.91 232	72.26 297	80.71 250	51.00 305	87.23 241	90.30 185	67.84 221	59.64 257	82.69 211	50.23 210	82.30 324	51.28 254	59.28 284	83.46 266

LCM-MVSNet-Re		72.93 224	71.84 221	76.18 273	88.49 156	48.02 315	80.07 298	70.17 337	73.96 115	52.25 298	80.09 255	49.98 211	88.24 293	67.35 163	84.23 118	92.28 137

Vis-MVSNet		80.92 102	79.98 100	83.74 122	88.48 157	61.80 213	93.44 69	88.26 249	73.96 115	77.73 88	91.76 107	49.94 212	94.76 142	65.84 180	90.37 73	94.65 67
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
v1192		75.98 190	73.92 194	82.15 167	79.73 276	66.24 126	91.22 154	89.75 204	72.67 138	68.49 195	81.42 233	49.86 213	94.27 167	67.08 166	65.02 251	85.95 235

test-mter		79.96 117	79.38 112	81.72 178	86.93 183	61.17 218	92.70 89	91.54 143	73.85 117	75.62 107	86.94 173	49.84 214	92.38 228	72.21 123	84.76 110	91.60 144

cdsmvs_eth3d_5k		19.86 330	26.47 326	0.00 346	0.00 360	0.00 361	0.00 352	93.45 63	0.00 356	0.00 357	95.27 33	49.56 215	0.00 359	0.00 356	0.00 356	0.00 357

3Dnovator+	73.60 7	82.10 89	80.60 94	86.60 46	90.89 109	66.80 102	95.20 23	93.44 70	74.05 111	67.42 206	92.49 95	49.46 216	97.65 40	70.80 138	91.68 57	95.33 38

MVP-Stereo		77.12 171	76.23 157	79.79 217	81.72 239	66.34 123	89.29 199	90.88 165	70.56 189	62.01 251	82.88 209	49.34 217	94.13 175	65.55 183	93.80 29	78.88 315
Qingsong Yan: MVP-Stereo: A Parallel Multi-View Patchmatch Stereo Method with Dilation Matching for Photogrammetric Application.
OMC-MVS		78.67 144	77.91 132	80.95 198	85.76 199	57.40 269	88.49 213	88.67 238	73.85 117	72.43 140	92.10 102	49.29 218	94.55 150	72.73 118	77.89 153	90.91 154

VPNet		78.82 137	77.53 137	82.70 141	84.52 210	66.44 120	93.93 53	92.23 115	80.46 30	72.60 134	88.38 149	49.18 219	93.13 204	72.47 121	63.97 261	88.55 178

CVMVSNet		74.04 217	74.27 188	73.33 288	85.33 200	43.94 325	89.53 197	88.39 244	54.33 306	70.37 160	90.13 127	49.17 220	84.05 310	61.83 215	79.36 141	91.99 140

v1921920		75.63 195	73.49 205	82.06 173	79.38 281	66.35 122	91.07 162	89.48 213	71.98 156	67.99 198	81.22 238	49.16 221	93.90 189	66.56 171	64.56 257	85.92 237

pm-mvs1		72.89 225	71.09 227	78.26 251	79.10 287	57.62 266	90.80 166	89.30 219	67.66 223	62.91 245	81.78 229	49.11 222	92.95 208	60.29 222	58.89 289	84.22 257

pmmvs4		73.92 219	71.81 222	80.25 205	79.17 284	65.24 142	87.43 238	87.26 262	67.64 225	63.46 240	83.91 201	48.96 223	91.53 256	62.94 205	65.49 244	83.96 258

TAPA-MVS	70.22 12	74.94 207	73.53 204	79.17 234	90.40 114	52.07 299	89.19 202	89.61 211	62.69 266	70.07 164	92.67 93	48.89 224	94.32 164	38.26 307	79.97 137	91.12 152
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
3Dnovator	73.91 6	82.69 80	80.82 90	88.31 14	89.57 134	71.26 13	92.60 94	94.39 39	78.84 48	67.89 201	92.48 96	48.42 225	98.52 17	68.80 155	94.40 22	95.15 49

CPTT-MVS		79.59 125	79.16 116	80.89 199	91.54 93	59.80 243	92.10 107	88.54 243	60.42 280	72.96 128	93.28 79	48.27 226	92.80 215	78.89 83	86.50 99	90.06 161

GBi-Net		75.65 193	73.83 195	81.10 193	88.85 148	65.11 146	90.01 183	90.32 181	70.84 181	67.04 211	80.25 252	48.03 227	91.54 253	59.80 225	69.34 218	86.64 219

test1		75.65 193	73.83 195	81.10 193	88.85 148	65.11 146	90.01 183	90.32 181	70.84 181	67.04 211	80.25 252	48.03 227	91.54 253	59.80 225	69.34 218	86.64 219

FMVSNet2		76.07 187	74.01 192	82.26 164	88.85 148	67.66 71	91.33 149	91.61 141	70.84 181	65.98 216	82.25 218	48.03 227	92.00 239	58.46 231	68.73 224	87.10 212

LFMVS		84.34 56	82.73 71	89.18 8	94.76 21	73.25 8	94.99 30	91.89 131	71.90 157	82.16 53	93.49 77	47.98 230	97.05 65	82.55 58	84.82 108	97.25 2

QAPM		79.95 118	77.39 141	87.64 21	89.63 133	71.41 12	93.30 71	93.70 53	65.34 242	67.39 208	91.75 108	47.83 231	98.96 5	57.71 234	89.81 75	92.54 131

HPM-MVS_fast		80.25 110	79.55 108	82.33 158	91.55 92	59.95 241	91.32 150	89.16 225	65.23 243	74.71 116	93.07 84	47.81 232	95.74 112	74.87 111	88.23 83	91.31 151

CANet_DTU		84.09 60	83.52 55	85.81 71	90.30 116	66.82 96	91.87 125	89.01 232	85.27 7	86.09 23	93.74 73	47.71 233	96.98 74	77.90 90	89.78 76	93.65 102

abl_6		79.82 120	79.20 115	81.70 180	89.85 123	58.34 259	88.47 214	90.07 195	62.56 267	77.71 89	93.08 82	47.65 234	96.78 84	77.94 89	85.45 106	89.99 163

v1240		75.21 201	72.98 210	81.88 175	79.20 283	66.00 129	90.75 168	89.11 228	71.63 169	67.41 207	81.22 238	47.36 235	93.87 190	65.46 184	64.72 255	85.77 239

PEN-MVS		69.46 258	68.56 243	72.17 299	79.27 282	49.71 311	86.90 247	89.24 222	67.24 229	59.08 261	82.51 212	47.23 236	83.54 316	48.42 265	57.12 292	83.25 269

DI_MVS_plusplus_test		79.78 122	77.50 138	86.62 45	80.90 245	69.46 35	90.69 169	91.97 129	77.00 68	59.07 262	82.34 216	46.82 237	95.88 107	82.14 60	86.59 97	94.53 73

CNLPA		74.31 215	72.30 218	80.32 203	91.49 94	61.66 214	90.85 164	80.72 312	56.67 300	63.85 237	90.64 117	46.75 238	90.84 261	53.79 246	75.99 172	88.47 180

test_normal		79.66 123	77.36 143	86.54 49	80.72 249	69.21 38	90.68 170	92.16 124	76.99 69	58.63 266	82.03 225	46.70 239	95.86 108	81.74 64	86.63 96	94.56 68

114514_t		79.17 131	77.67 133	83.68 126	95.32 16	65.53 139	92.85 85	91.60 142	63.49 259	67.92 200	90.63 119	46.65 240	95.72 117	67.01 167	83.54 120	89.79 164

PS-CasMVS		69.86 253	69.13 239	72.07 300	80.35 264	50.57 307	87.02 244	89.75 204	67.27 228	59.19 260	82.28 217	46.58 241	82.24 325	50.69 256	59.02 287	83.39 268

DTE-MVSNet		68.46 264	67.33 254	71.87 303	77.94 296	49.00 314	86.16 255	88.58 242	66.36 233	58.19 267	82.21 220	46.36 242	83.87 314	44.97 280	55.17 299	82.73 276

PMMVS		81.98 91	82.04 77	81.78 176	89.76 126	56.17 281	91.13 160	90.69 170	77.96 57	80.09 67	93.57 75	46.33 243	94.99 134	81.41 67	87.46 89	94.17 83

OPM-MVS		79.00 133	78.09 127	81.73 177	83.52 225	63.83 173	91.64 139	90.30 185	76.36 79	71.97 145	89.93 136	46.30 244	95.17 132	75.10 104	77.70 155	86.19 226

BH-RMVSNet		79.46 128	77.65 134	84.89 99	91.68 89	65.66 136	93.55 65	88.09 250	72.93 135	73.37 126	91.12 113	46.20 245	96.12 99	56.28 238	85.61 105	92.91 123

TR-MVS		78.77 140	77.37 142	82.95 136	90.49 112	60.88 221	93.67 61	90.07 195	70.08 195	74.51 117	91.37 111	45.69 246	95.70 118	60.12 223	80.32 136	92.29 136

Patchmatch-test1		75.00 204	71.80 223	84.58 108	86.63 185	70.08 24	81.06 288	89.19 223	71.60 170	70.01 165	77.16 278	45.53 247	88.63 287	51.79 253	73.27 188	95.02 57

semantic-postprocess				76.32 271	81.48 240	60.67 228		85.99 281	66.17 234	59.50 258	78.88 264	45.51 248	83.65 315	62.58 210	61.93 269	84.63 256

IterMVS		72.65 230	70.83 228	78.09 255	82.17 235	62.96 192	87.64 236	86.28 275	71.56 172	60.44 254	78.85 265	45.42 249	86.66 301	63.30 199	61.83 270	84.65 255
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
Effi-MVS+-dtu		76.14 186	75.28 176	78.72 245	83.22 227	55.17 287	89.87 191	87.78 254	75.42 87	67.98 199	81.43 232	45.08 250	92.52 224	75.08 105	71.63 200	88.48 179

mvs-test1		78.74 141	77.95 130	81.14 191	83.22 227	57.13 271	93.96 50	87.78 254	75.42 87	72.68 132	90.80 116	45.08 250	94.54 151	75.08 105	77.49 161	91.74 143

v748		70.55 248	67.97 251	78.27 250	75.75 308	58.78 256	86.29 254	89.25 221	65.12 250	56.66 276	77.17 277	45.05 252	92.95 208	58.13 232	58.33 291	83.10 273

XVG-OURS-SEG-HR		74.70 213	73.08 209	79.57 221	78.25 293	57.33 270	80.49 291	87.32 260	63.22 262	68.76 187	90.12 129	44.89 253	91.59 251	70.55 141	74.09 184	89.79 164

v7n		71.31 238	68.65 241	79.28 229	76.40 303	60.77 224	86.71 250	89.45 214	64.17 255	58.77 265	78.24 267	44.59 254	93.54 197	57.76 233	61.75 272	83.52 264

pmmvs5		73.35 222	71.52 224	78.86 240	78.64 291	60.61 230	91.08 161	86.90 263	67.69 222	63.32 241	83.64 202	44.33 255	90.53 262	62.04 213	66.02 239	85.46 246

OpenMVS	70.45 11	78.54 146	75.92 161	86.41 57	85.93 198	71.68 11	92.74 87	92.51 109	66.49 232	64.56 230	91.96 103	43.88 256	98.10 28	54.61 242	90.65 70	89.44 169

AdaColmap		78.94 135	77.00 148	84.76 102	96.34 6	65.86 132	92.66 93	87.97 253	62.18 269	70.56 154	92.37 99	43.53 257	97.35 52	64.50 191	82.86 123	91.05 153

tfpnnormal		70.10 250	67.36 253	78.32 248	83.45 226	60.97 220	88.85 207	92.77 98	64.85 251	60.83 253	78.53 266	43.52 258	93.48 199	31.73 331	61.70 274	80.52 303

test_djsdf		73.76 221	72.56 215	77.39 263	77.00 301	53.93 292	89.07 205	90.69 170	65.80 237	63.92 235	82.03 225	43.14 259	92.67 220	72.83 116	68.53 225	85.57 244

GA-MVS		78.33 149	76.23 157	84.65 106	83.65 223	66.30 124	91.44 142	90.14 193	76.01 81	70.32 161	84.02 200	42.50 260	94.72 145	70.98 136	77.00 166	92.94 122

PLC	68.80 14	75.23 200	73.68 197	79.86 215	92.93 56	58.68 258	90.64 171	88.30 246	60.90 277	64.43 233	90.53 120	42.38 261	94.57 148	56.52 236	76.54 169	86.33 223
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
1111		56.66 305	54.98 304	61.69 320	61.99 336	31.38 342	79.81 301	83.17 303	45.66 325	41.94 329	65.44 322	41.50 262	79.56 333	27.64 335	47.68 322	74.14 328

.test1245		46.52 315	49.68 311	37.02 337	61.99 336	31.38 342	79.81 301	83.17 303	45.66 325	41.94 329	65.44 322	41.50 262	79.56 333	27.64 335	0.01 354	0.13 355

V4		69.80 254	67.02 256	78.15 253	71.86 316	60.10 238	82.02 279	87.39 257	64.48 252	57.78 271	75.98 283	41.49 264	92.90 213	63.00 203	59.16 285	81.44 288

v52		69.80 254	67.01 257	78.15 253	71.84 317	60.10 238	82.02 279	87.39 257	64.48 252	57.80 270	75.97 284	41.47 265	92.90 213	63.00 203	59.13 286	81.45 287

Fast-Effi-MVS+-dtu		75.04 202	73.37 207	80.07 209	80.86 246	59.52 247	91.20 156	85.38 286	71.90 157	65.20 224	84.84 193	41.46 266	92.97 207	66.50 172	72.96 192	87.73 196

MS-PatchMatch		77.90 157	76.50 154	82.12 169	85.99 194	69.95 27	91.75 134	92.70 100	73.97 114	62.58 248	84.44 198	41.11 267	95.78 110	63.76 195	92.17 52	80.62 302

pcd1.5k->3k		31.17 324	31.85 322	29.12 339	81.48 240	0.00 361	0.00 352	91.79 135	0.00 356	0.00 357	0.00 358	41.05 268	0.00 359	0.00 356	72.34 198	87.36 207

XVG-OURS		74.25 216	72.46 217	79.63 219	78.45 292	57.59 267	80.33 293	87.39 257	63.86 258	68.76 187	89.62 140	40.50 269	91.72 243	69.00 152	74.25 182	89.58 167

VDD-MVS		83.06 74	81.81 80	86.81 40	90.86 110	67.70 70	95.40 19	91.50 146	75.46 86	81.78 55	92.34 100	40.09 270	97.13 64	86.85 33	82.04 129	95.60 32

DP-MVS		69.90 252	66.48 258	80.14 206	95.36 15	62.93 193	89.56 194	76.11 320	50.27 315	57.69 272	85.23 189	39.68 271	95.73 113	33.35 325	71.05 206	81.78 283

ADS-MVSNet2		66.90 273	63.44 276	77.26 266	88.06 166	60.70 227	68.01 328	75.56 325	57.57 292	64.48 231	69.87 315	38.68 272	84.10 309	40.87 294	67.89 230	86.97 213

ADS-MVSNet		68.54 263	64.38 273	81.03 196	88.06 166	66.90 93	68.01 328	84.02 294	57.57 292	64.48 231	69.87 315	38.68 272	89.21 286	40.87 294	67.89 230	86.97 213

LPG-MVS_test		75.82 192	74.58 183	79.56 222	84.31 215	59.37 249	90.44 174	89.73 207	69.49 199	64.86 227	88.42 147	38.65 274	94.30 165	72.56 119	72.76 193	85.01 251

LGP-MVS_train				79.56 222	84.31 215	59.37 249		89.73 207	69.49 199	64.86 227	88.42 147	38.65 274	94.30 165	72.56 119	72.76 193	85.01 251

VDDNet		80.50 106	78.26 125	87.21 31	86.19 191	69.79 30	94.48 36	91.31 153	60.42 280	79.34 74	90.91 114	38.48 276	96.56 92	82.16 59	81.05 133	95.27 43

ACMP	71.68 10	75.58 196	74.23 189	79.62 220	84.97 205	59.64 244	90.80 166	89.07 230	70.39 192	62.95 244	87.30 171	38.28 277	93.87 190	72.89 115	71.45 203	85.36 248
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
UGNet		79.87 119	78.68 119	83.45 132	89.96 121	61.51 216	92.13 105	90.79 167	76.83 72	78.85 81	86.33 179	38.16 278	96.17 97	67.93 159	87.17 90	92.67 127
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
anonymousdsp		71.14 241	69.37 236	76.45 270	72.95 313	54.71 289	84.19 262	88.88 235	61.92 272	62.15 250	79.77 257	38.14 279	91.44 259	68.90 154	67.45 233	83.21 270

xiu_mvs_v1_base_debu		82.16 86	81.12 87	85.26 91	86.42 186	68.72 47	92.59 96	90.44 175	73.12 132	84.20 41	94.36 58	38.04 280	95.73 113	84.12 48	86.81 92	91.33 147

xiu_mvs_v1_base		82.16 86	81.12 87	85.26 91	86.42 186	68.72 47	92.59 96	90.44 175	73.12 132	84.20 41	94.36 58	38.04 280	95.73 113	84.12 48	86.81 92	91.33 147

xiu_mvs_v1_base_debi		82.16 86	81.12 87	85.26 91	86.42 186	68.72 47	92.59 96	90.44 175	73.12 132	84.20 41	94.36 58	38.04 280	95.73 113	84.12 48	86.81 92	91.33 147

PVSNet_0	68.08 15	71.81 233	68.32 247	82.27 161	84.68 207	62.31 205	88.68 210	90.31 184	75.84 82	57.93 269	80.65 246	37.85 283	94.19 170	69.94 144	29.05 342	90.31 160

Anonymous20231206		67.53 270	65.78 260	72.79 293	74.95 309	47.59 318	88.23 217	87.32 260	61.75 275	58.07 268	77.29 274	37.79 284	87.29 299	42.91 284	63.71 262	83.48 265

ACMM	69.62 13	74.34 214	72.73 212	79.17 234	84.25 217	57.87 262	90.36 177	89.93 200	63.17 263	65.64 223	86.04 183	37.79 284	94.10 176	65.89 179	71.52 202	85.55 245
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
cascas		78.18 151	75.77 163	85.41 86	87.14 179	69.11 39	92.96 80	91.15 158	66.71 230	70.47 155	86.07 181	37.49 286	96.48 93	70.15 143	79.80 138	90.65 156

LS3D		69.17 259	66.40 259	77.50 260	91.92 79	56.12 282	85.12 257	80.37 313	46.96 322	56.50 277	87.51 165	37.25 287	93.71 194	32.52 330	79.40 140	82.68 278

MDA-MVSNet_test_wron		63.78 289	60.16 290	74.64 279	78.15 294	60.41 231	83.49 267	84.03 293	56.17 303	39.17 334	71.59 312	37.22 288	83.24 320	42.87 286	48.73 320	80.26 306

YYNet1		63.76 290	60.14 291	74.62 280	78.06 295	60.19 237	83.46 269	83.99 297	56.18 302	39.25 333	71.56 313	37.18 289	83.34 318	42.90 285	48.70 321	80.32 305

FMVSNet5		68.04 266	65.66 262	75.18 277	84.43 213	57.89 261	83.54 266	86.26 276	61.83 274	53.64 293	73.30 292	37.15 290	85.08 305	48.99 262	61.77 271	82.56 279

test20.03		63.83 288	62.65 282	67.38 312	70.58 323	39.94 333	86.57 252	84.17 292	63.29 261	51.86 299	77.30 273	37.09 291	82.47 322	38.87 306	54.13 303	79.73 310

PVSNet	73.49 8	80.05 115	78.63 120	84.31 113	90.92 108	64.97 152	92.47 99	91.05 162	79.18 41	72.43 140	90.51 122	37.05 292	94.06 179	68.06 157	86.00 102	93.90 99

EU-MVSNet		64.01 287	63.01 279	67.02 313	74.40 311	38.86 337	83.27 271	86.19 277	45.11 328	54.27 286	81.15 241	36.91 293	80.01 330	48.79 264	57.02 293	82.19 281

FMVSNet1		72.71 228	69.91 232	81.10 193	83.60 224	65.11 146	90.01 183	90.32 181	63.92 257	63.56 239	80.25 252	36.35 294	91.54 253	54.46 243	66.75 236	86.64 219

testpf		57.17 302	56.93 299	57.88 324	79.13 286	42.40 326	34.23 348	85.97 282	52.64 308	47.66 314	66.50 319	36.33 295	79.65 332	53.60 248	56.31 297	51.60 343

CMPMVS	48.56 21	66.77 274	64.41 272	73.84 285	70.65 322	50.31 308	77.79 311	85.73 285	45.54 327	44.76 322	82.14 221	35.40 296	90.14 271	63.18 201	74.54 180	81.07 297
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
pmmvs6		67.57 269	64.76 268	76.00 274	72.82 315	53.37 294	88.71 209	86.78 264	53.19 307	57.58 273	78.03 269	35.33 297	92.41 227	55.56 240	54.88 301	82.21 280

test2356		64.16 286	63.28 277	66.81 314	69.37 327	39.86 335	87.76 227	86.02 280	59.83 284	53.54 294	73.23 293	34.94 298	80.67 329	39.66 298	65.20 246	79.89 308

PatchMatch-RL		72.06 231	69.98 230	78.28 249	89.51 137	55.70 284	83.49 267	83.39 301	61.24 276	63.72 238	82.76 210	34.77 299	93.03 206	53.37 250	77.59 156	86.12 228

LTVRE_ROB	59.60 19	66.27 276	63.54 275	74.45 281	84.00 220	51.55 301	67.08 331	83.53 298	58.78 289	54.94 281	80.31 250	34.54 300	93.23 203	40.64 296	68.03 228	78.58 318
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
F-COLMAP		70.66 246	68.44 245	77.32 264	86.37 189	55.91 283	88.00 220	86.32 274	56.94 298	57.28 274	88.07 156	33.58 301	92.49 225	51.02 255	68.37 226	83.55 262

pmmvs-eth3d		65.53 280	62.32 284	75.19 276	69.39 326	59.59 245	82.80 276	83.43 299	62.52 268	51.30 303	72.49 297	32.86 302	87.16 300	55.32 241	50.73 312	78.83 316

MDA-MVSNet-bldmvs		61.54 295	57.70 296	73.05 290	79.53 279	57.00 273	83.08 274	81.23 308	57.57 292	34.91 337	72.45 299	32.79 303	86.26 304	35.81 313	41.95 330	75.89 325

MIMVSNet		71.64 236	68.44 245	81.23 186	81.97 238	64.44 159	73.05 317	88.80 236	69.67 198	64.59 229	74.79 289	32.79 303	87.82 296	53.99 245	76.35 170	91.42 146

UnsupCasMVSNet_eth		65.79 279	63.10 278	73.88 284	70.71 321	50.29 309	81.09 287	89.88 201	72.58 140	49.25 309	74.77 290	32.57 305	87.43 298	55.96 239	41.04 332	83.90 260

N_pmnet		50.55 311	49.11 313	54.88 328	77.17 300	4.02 358	84.36 261	2.00 358	48.59 318	45.86 318	68.82 317	32.22 306	82.80 321	31.58 332	51.38 310	77.81 321

test_0402		64.54 283	61.09 288	74.92 278	84.10 219	60.75 225	87.95 221	79.71 315	52.03 310	52.41 297	77.20 275	32.21 307	91.64 250	23.14 341	61.03 278	72.36 330

DSMNet-mixed		56.78 303	54.44 305	63.79 319	63.21 333	29.44 346	64.43 334	64.10 344	42.12 336	51.32 302	71.60 311	31.76 308	75.04 339	36.23 311	65.20 246	86.87 217

Test4		76.45 183	73.45 206	85.45 84	76.07 307	67.61 73	88.38 216	90.83 166	76.71 74	53.06 295	79.65 260	31.61 309	94.35 163	78.47 84	86.22 101	94.40 77

MSDG		69.54 257	65.73 261	80.96 197	85.11 204	63.71 178	84.19 262	83.28 302	56.95 297	54.50 283	84.03 199	31.50 310	96.03 103	42.87 286	69.13 221	83.14 272

RPSCF		64.24 285	61.98 286	71.01 304	76.10 305	45.00 322	75.83 314	75.94 322	46.94 323	58.96 263	84.59 196	31.40 311	82.00 326	47.76 268	60.33 283	86.04 234

jajsoiax		73.05 223	71.51 225	77.67 258	77.46 298	54.83 288	88.81 208	90.04 197	69.13 203	62.85 246	83.51 204	31.16 312	92.75 217	70.83 137	69.80 214	85.43 247

MVS-HIRNet		60.25 297	55.55 303	74.35 282	84.37 214	56.57 280	71.64 319	74.11 329	34.44 340	45.54 320	42.24 342	31.11 313	89.81 281	40.36 297	76.10 171	76.67 324

SixPastTwentyTwo		64.92 281	61.78 287	74.34 283	78.74 289	49.76 310	83.42 270	79.51 316	62.86 265	50.27 306	77.35 272	30.92 314	90.49 263	45.89 274	47.06 323	82.78 274

LP		56.71 304	51.64 308	71.91 302	80.08 273	60.33 233	61.72 336	75.61 324	43.87 333	43.76 326	60.30 332	30.46 315	84.05 310	22.94 342	46.06 325	71.34 332

mvs_tets		72.71 228	71.11 226	77.52 259	77.41 299	54.52 290	88.45 215	89.76 203	68.76 208	62.70 247	83.26 207	29.49 316	92.71 218	70.51 142	69.62 216	85.34 249

K. test v3		63.09 291	59.61 293	73.53 287	76.26 304	49.38 313	83.27 271	77.15 319	64.35 254	47.77 312	72.32 302	28.73 317	87.79 297	49.93 260	36.69 337	83.41 267

UnsupCasMVSNet_bld		61.60 294	57.71 295	73.29 289	68.73 328	51.64 300	78.61 305	89.05 231	57.20 296	46.11 315	61.96 330	28.70 318	88.60 289	50.08 259	38.90 335	79.63 311

lessismore_v0				73.72 286	72.93 314	47.83 317		61.72 347		45.86 318	73.76 291	28.63 319	89.81 281	47.75 269	31.37 341	83.53 263

new-patchmatchnet		59.30 301	56.48 301	67.79 310	65.86 330	44.19 323	82.47 277	81.77 306	59.94 283	43.65 327	66.20 321	27.67 320	81.68 327	39.34 299	41.40 331	77.50 322

testing_2		71.09 242	67.32 255	82.40 157	69.82 324	66.52 118	83.64 265	90.77 168	72.21 150	45.12 321	71.07 314	27.60 321	93.74 193	75.71 100	69.96 213	86.95 215

ACMH	63.93 17	68.62 261	64.81 267	80.03 210	85.22 202	63.25 185	87.72 228	84.66 289	60.83 278	51.57 301	79.43 262	27.29 322	94.96 135	41.76 290	64.84 253	81.88 282
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
OurMVSNet-221017-0		64.68 282	62.17 285	72.21 298	76.08 306	47.35 319	80.67 290	81.02 310	56.19 301	51.60 300	79.66 259	27.05 323	88.56 290	53.60 248	53.63 304	80.71 301

ACMH+	65.35 16	67.65 268	64.55 269	76.96 267	84.59 209	57.10 272	88.08 219	80.79 311	58.59 291	53.00 296	81.09 242	26.63 324	92.95 208	46.51 271	61.69 275	80.82 299

OpenMVS_ROB	61.12 18	66.39 275	62.92 280	76.80 269	76.51 302	57.77 263	89.22 200	83.41 300	55.48 304	53.86 291	77.84 270	26.28 325	93.95 187	34.90 322	68.76 223	78.68 317

COLMAP_ROB	57.96 20	62.98 292	59.65 292	72.98 291	81.44 242	53.00 296	83.75 264	75.53 326	48.34 320	48.81 310	81.40 234	24.14 326	90.30 264	32.95 327	60.52 282	75.65 326
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
MIMVSNet1		60.16 298	57.33 298	68.67 308	69.71 325	44.13 324	78.92 304	84.21 291	55.05 305	44.63 323	71.85 306	23.91 327	81.54 328	32.63 329	55.03 300	80.35 304

testgi		64.48 284	62.87 281	69.31 307	71.24 318	40.62 332	85.49 256	79.92 314	65.36 241	54.18 288	83.49 205	23.74 328	84.55 307	41.60 291	60.79 281	82.77 275

ITE_SJBPF				70.43 305	74.44 310	47.06 320		77.32 318	60.16 282	54.04 289	83.53 203	23.30 329	84.01 312	43.07 283	61.58 276	80.21 307

EG-PatchMatch MVS		68.55 262	65.41 264	77.96 256	78.69 290	62.93 193	89.86 192	89.17 224	60.55 279	50.27 306	77.73 271	22.60 330	94.06 179	47.18 270	72.65 195	76.88 323

tmp_tt		22.26 329	23.75 328	17.80 342	5.23 357	12.06 357	35.26 347	39.48 353	2.82 353	18.94 344	44.20 341	22.23 331	24.64 355	36.30 310	9.31 351	16.69 350

USDC		67.43 272	64.51 270	76.19 272	77.94 296	55.29 286	78.38 307	85.00 287	73.17 130	48.36 311	80.37 249	21.23 332	92.48 226	52.15 252	64.02 260	80.81 300

XVG-ACMP-BASELINE		68.04 266	65.53 263	75.56 275	74.06 312	52.37 297	78.43 306	85.88 283	62.03 270	58.91 264	81.21 240	20.38 333	91.15 260	60.69 220	68.18 227	83.16 271

AllTest		61.66 293	58.06 294	72.46 295	79.57 277	51.42 303	80.17 296	68.61 339	51.25 312	45.88 316	81.23 236	19.86 334	86.58 302	38.98 300	57.01 294	79.39 312

TestCases				72.46 295	79.57 277	51.42 303		68.61 339	51.25 312	45.88 316	81.23 236	19.86 334	86.58 302	38.98 300	57.01 294	79.39 312

test1235678		55.73 306	52.74 306	64.68 318	60.16 339	35.56 340	81.65 283	81.46 307	51.27 311	38.93 335	62.82 329	17.44 336	78.58 337	30.87 333	50.09 319	79.89 308

pmmvs3		55.51 307	51.50 310	67.53 311	57.90 341	50.93 306	80.37 292	73.66 330	40.63 337	44.15 325	64.75 326	16.30 337	78.97 335	44.77 281	40.98 333	72.69 329

Anonymous20231211		53.57 310	49.43 312	66.00 316	65.01 331	42.08 327	80.95 289	72.60 332	38.46 338	41.65 331	64.48 327	15.72 338	84.23 308	25.78 338	40.24 334	71.68 331

TDRefinement		55.28 308	51.58 309	66.39 315	59.53 340	46.15 321	76.23 313	72.80 331	44.60 330	42.49 328	76.28 282	15.29 339	82.39 323	33.20 326	43.75 328	70.62 334

new_pmnet		49.31 312	46.44 314	57.93 323	62.84 334	40.74 331	68.47 327	62.96 346	36.48 339	35.09 336	57.81 334	14.97 340	72.18 340	32.86 328	46.44 324	60.88 341

TinyColmap		60.32 296	56.42 302	72.00 301	78.78 288	53.18 295	78.36 308	75.64 323	52.30 309	41.59 332	75.82 286	14.76 341	88.35 292	35.84 312	54.71 302	74.46 327

LF4IMVS		54.01 309	52.12 307	59.69 322	62.41 335	39.91 334	68.59 326	68.28 341	42.96 334	44.55 324	75.18 287	14.09 342	68.39 343	41.36 293	51.68 309	70.78 333

PM-MVS		59.40 299	56.59 300	67.84 309	63.63 332	41.86 329	76.76 312	63.22 345	59.01 288	51.07 304	72.27 303	11.72 343	83.25 319	61.34 216	50.28 318	78.39 319

no-one		44.13 317	38.39 318	61.34 321	45.91 349	41.94 328	61.67 337	75.07 328	45.05 329	20.07 343	40.68 345	11.58 344	79.82 331	30.18 334	15.30 345	62.26 340

testus		59.36 300	57.51 297	64.90 317	66.72 329	37.56 338	84.98 258	81.09 309	57.46 295	47.72 313	72.76 294	11.43 345	78.78 336	36.56 309	58.91 288	78.36 320

ambc				69.61 306	61.38 338	41.35 330	49.07 345	85.86 284		50.18 308	66.40 320	10.16 346	88.14 294	45.73 275	44.20 327	79.32 314

FPMVS		45.64 316	43.10 317	53.23 330	51.42 344	36.46 339	64.97 333	71.91 334	29.13 342	27.53 340	61.55 331	9.83 347	65.01 347	16.00 347	55.58 298	58.22 342

ANet_high		40.27 319	35.20 320	55.47 327	34.74 353	34.47 341	63.84 335	71.56 335	48.42 319	18.80 345	41.08 343	9.52 348	64.45 348	20.18 344	8.66 352	67.49 338

test12356		47.51 313	44.82 315	55.56 326	52.53 342	21.09 353	71.45 320	76.03 321	44.14 332	30.69 338	58.18 333	9.01 349	76.14 338	26.95 337	34.43 340	69.46 336

testmv		46.98 314	43.53 316	57.35 325	47.75 347	30.41 345	74.99 316	77.69 317	42.84 335	28.03 339	53.36 335	8.18 350	71.18 341	24.36 340	34.55 338	70.46 335

EMVS		23.76 328	23.20 329	25.46 341	41.52 351	16.90 356	60.56 339	38.79 355	14.62 349	8.99 352	20.24 353	7.35 351	45.82 353	7.25 352	9.46 350	13.64 352

E-PMN		24.61 326	24.00 327	26.45 340	43.74 350	18.44 355	60.86 338	39.66 352	15.11 348	9.53 351	22.10 349	6.52 352	46.94 352	8.31 351	10.14 348	13.98 351

DeepMVS_CX				34.71 338	51.45 343	24.73 352		28.48 357	31.46 341	17.49 346	52.75 336	5.80 353	42.60 354	18.18 346	19.42 343	36.81 347

Gipuma		34.91 321	31.44 323	45.30 333	70.99 320	39.64 336	19.85 351	72.56 333	20.10 347	16.16 347	21.47 350	5.08 354	71.16 342	13.07 348	43.70 329	25.08 349
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
LCM-MVSNet		40.54 318	35.79 319	54.76 329	36.92 352	30.81 344	51.41 343	69.02 338	22.07 344	24.63 341	45.37 340	4.56 355	65.81 345	33.67 324	34.50 339	67.67 337

PMMVS2		37.93 320	33.61 321	50.92 331	46.31 348	24.76 351	60.55 340	50.05 349	28.94 343	20.93 342	47.59 337	4.41 356	65.13 346	25.14 339	18.55 344	62.87 339

PMVS	26.43 22	31.84 323	28.16 325	42.89 334	25.87 356	27.58 349	50.92 344	49.78 351	21.37 346	14.17 349	40.81 344	2.01 357	66.62 344	9.61 350	38.88 336	34.49 348
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
MVE	24.84 23	24.35 327	19.77 331	38.09 336	34.56 354	26.92 350	26.57 349	38.87 354	11.73 351	11.37 350	27.44 347	1.37 358	50.42 351	11.41 349	14.60 346	36.93 346
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
PNet_i23d		32.77 322	29.98 324	41.11 335	48.05 345	29.17 347	65.82 332	50.02 350	21.42 345	14.74 348	37.19 346	1.11 359	55.11 350	19.75 345	11.77 347	39.06 345

wuyk23d		11.30 331	10.95 332	12.33 343	48.05 345	19.89 354	25.89 350	1.92 359	3.58 352	3.12 354	1.37 355	0.64 360	15.77 356	6.23 353	7.77 353	1.35 353

wuykxyi23d		29.03 325	23.09 330	46.84 332	31.67 355	28.82 348	43.46 346	57.72 348	14.39 350	7.52 353	20.84 351	0.64 360	60.29 349	21.57 343	10.04 349	51.40 344

test123		6.92 334	9.21 335	0.08 344	0.03 359	0.05 359	81.65 283	0.01 361	0.02 355	0.14 356	0.85 357	0.03 362	0.02 357	0.12 355	0.00 356	0.16 354

testmvs		7.23 333	9.62 334	0.06 345	0.04 358	0.02 360	84.98 258	0.02 360	0.03 354	0.18 355	1.21 356	0.01 363	0.02 357	0.14 354	0.01 354	0.13 355

sosnet-low-res		0.00 336	0.00 337	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	0.00 358	0.00 364	0.00 359	0.00 356	0.00 356	0.00 357

sosnet		0.00 336	0.00 337	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	0.00 358	0.00 364	0.00 359	0.00 356	0.00 356	0.00 357

uncertanet		0.00 336	0.00 337	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	0.00 358	0.00 364	0.00 359	0.00 356	0.00 356	0.00 357

Regformer		0.00 336	0.00 337	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	0.00 358	0.00 364	0.00 359	0.00 356	0.00 356	0.00 357

ab-mvs-re		7.91 332	10.55 333	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	94.95 43	0.00 364	0.00 359	0.00 356	0.00 356	0.00 357

uanet		0.00 336	0.00 337	0.00 346	0.00 360	0.00 361	0.00 352	0.00 362	0.00 356	0.00 357	0.00 358	0.00 364	0.00 359	0.00 356	0.00 356	0.00 357

GSMVS																94.68 65

test_part3							94.96 31		68.52 213		97.23 2		98.90 7	91.52 6

test_part2					96.29 7	68.16 61				90.78 4

MTGPA								92.23 115

MTMP								32.52 356

gm-plane-assit					88.42 160	67.04 88			78.62 52		91.83 105		97.37 50	76.57 96

test9_res														89.41 12	94.96 11	95.29 41

agg_prior2														86.41 34	94.75 19	95.33 38

agg_prior					94.16 33	66.97 89		93.31 76		84.49 38			96.75 86

test_prior4						67.18 84	93.92 54

test_prior				86.42 55	94.71 23	67.35 78		93.10 88					96.84 82			95.05 53

旧先验2							92.00 115		59.37 287	87.54 17			93.47 200	75.39 102

新几何2							91.41 143

无先验							92.71 88	92.61 106	62.03 270				97.01 68	66.63 169		93.97 95

原ACMM2							92.01 113

testdata2													96.09 100	61.26 217

testdata1							89.21 201		77.55 63

plane_prior7					86.94 181	61.51 216

plane_prior5								91.31 153					95.55 123	76.74 94	78.53 149	88.39 181

plane_prior4											89.14 142

plane_prior3						61.95 212			79.09 44	72.53 136

plane_prior2							93.13 75		78.81 49

plane_prior1					87.15 178

plane_prior						62.42 202	93.85 57		79.38 37						78.80 147

n2								0.00 362

nn								0.00 362

door-mid								66.01 343

test11								93.01 90

door								66.57 342

HQP5-MVS						63.66 181

HQP-NCC					87.54 173		94.06 45		79.80 33	74.18 118

ACMP_Plane					87.54 173		94.06 45		79.80 33	74.18 118

BP-MVS														77.63 91

HQP4-MVS										74.18 118			95.61 119			88.63 176

HQP3-MVS								91.70 139							78.90 145

NP-MVS					87.41 176	63.04 191					90.30 124

ACMMP++_ref															71.63 200

ACMMP++															69.72 215

High-resolution multi-view benchmark