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.

HPM-MVS		96.69 38	96.45 41	97.40 45	99.36 15	93.11 62	98.87 1	98.06 66	91.17 133	96.40 55	97.99 57	90.99 55	99.58 61	95.61 52	99.61 11	99.49 31
Chunlin Ren, Qingshan Xu, Shikun Zhang, Jiaqi Yang: Hierarchical Prior Mining for Non-local Multi-View Stereo. ICCV 2023
APDe-MVS		97.82 3	97.73 2	98.08 11	99.15 29	94.82 17	98.81 2	98.30 23	94.76 28	98.30 9	98.90 3	93.77 10	99.68 42	97.93 1	99.69 3	99.75 3

CP-MVS		97.02 22	96.81 24	97.64 38	99.33 18	93.54 50	98.80 3	98.28 26	92.99 77	96.45 54	98.30 40	91.90 38	99.85 13	95.61 52	99.68 4	99.54 23

HPM-MVS_fast		96.51 43	96.27 45	97.22 56	99.32 19	92.74 70	98.74 4	98.06 66	90.57 154	96.77 38	98.35 29	90.21 66	99.53 77	94.80 73	99.63 9	99.38 45

EPP-MVSNet		95.22 76	95.04 73	95.76 116	97.49 119	89.56 165	98.67 5	97.00 190	90.69 143	94.24 107	97.62 86	89.79 72	98.81 148	93.39 101	96.49 137	98.92 84

3Dnovator	91.36 5	95.19 78	94.44 90	97.44 44	96.56 157	93.36 57	98.65 6	98.36 16	94.12 42	89.25 225	98.06 52	82.20 174	99.77 27	93.41 100	99.32 45	99.18 57

XVS		97.18 13	96.96 15	97.81 22	99.38 11	94.03 38	98.59 7	98.20 37	94.85 20	96.59 46	98.29 42	91.70 42	99.80 25	95.66 45	99.40 37	99.62 11

X-MVStestdata		91.71 183	89.67 233	97.81 22	99.38 11	94.03 38	98.59 7	98.20 37	94.85 20	96.59 46	32.69 332	91.70 42	99.80 25	95.66 45	99.40 37	99.62 11

MSP-MVS		97.59 6	97.54 4	97.73 30	99.40 8	93.77 46	98.53 9	98.29 24	95.55 5	98.56 8	97.81 70	93.90 8	99.65 46	96.62 17	99.21 56	99.77 1

HFP-MVS		97.14 16	96.92 17	97.83 19	99.42 6	94.12 34	98.52 10	98.32 20	93.21 68	97.18 28	98.29 42	92.08 32	99.83 19	95.63 50	99.59 13	99.54 23

region2R		97.07 19	96.84 21	97.77 27	99.46 1	93.79 43	98.52 10	98.24 31	93.19 71	97.14 31	98.34 32	91.59 45	99.87 6	95.46 56	99.59 13	99.64 8

ACMMPR		97.07 19	96.84 21	97.79 24	99.44 5	93.88 40	98.52 10	98.31 22	93.21 68	97.15 30	98.33 35	91.35 48	99.86 7	95.63 50	99.59 13	99.62 11

mPP-MVS		96.86 30	96.60 32	97.64 38	99.40 8	93.44 53	98.50 13	98.09 57	93.27 67	95.95 71	98.33 35	91.04 54	99.88 4	95.20 58	99.57 17	99.60 14

3Dnovator+	91.43 4	95.40 69	94.48 88	98.16 9	96.90 140	95.34 11	98.48 14	97.87 95	94.65 32	88.53 234	98.02 55	83.69 143	99.71 34	93.18 103	98.96 72	99.44 38

IS-MVSNet		94.90 86	94.52 86	96.05 107	97.67 110	90.56 140	98.44 15	96.22 228	93.21 68	93.99 111	97.74 75	85.55 122	98.45 178	89.98 152	97.86 100	99.14 61

SteuartSystems-ACMMP		97.62 5	97.53 5	97.87 18	98.39 67	94.25 28	98.43 16	98.27 28	95.34 9	98.11 10	98.56 12	94.53 5	99.71 34	96.57 21	99.62 10	99.65 7
Skip Steuart: Steuart Systems R&D Blog.
MP-MVS		96.77 35	96.45 41	97.72 31	99.39 10	93.80 42	98.41 17	98.06 66	93.37 63	95.54 88	98.34 32	90.59 62	99.88 4	94.83 70	99.54 19	99.49 31
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
QAPM		93.45 126	92.27 145	96.98 64	96.77 147	92.62 74	98.39 18	98.12 50	84.50 272	88.27 240	97.77 73	82.39 171	99.81 23	85.40 239	98.81 76	98.51 112

nrg030		94.05 107	93.31 116	96.27 97	95.22 218	94.59 20	98.34 19	97.46 137	92.93 84	91.21 174	96.64 133	87.23 104	98.22 191	94.99 67	85.80 265	95.98 208

CPTT-MVS		95.57 67	95.19 69	96.70 67	99.27 22	91.48 106	98.33 20	98.11 53	87.79 222	95.17 93	98.03 54	87.09 105	99.61 53	93.51 95	99.42 35	99.02 71

test0726					99.45 2	95.36 8	98.31 21	98.29 24	94.92 18	98.99 2	98.92 2	95.08 2

CSCG		96.05 54	95.91 52	96.46 83	99.24 24	90.47 143	98.30 22	98.57 11	89.01 186	93.97 113	97.57 90	92.62 23	99.76 28	94.66 76	99.27 50	99.15 60

GST-MVS		96.85 31	96.52 37	97.82 21	99.36 15	94.14 33	98.29 23	98.13 48	92.72 91	96.70 39	98.06 52	91.35 48	99.86 7	94.83 70	99.28 49	99.47 35

canonicalmvs		96.02 55	95.45 61	97.75 29	97.59 116	95.15 14	98.28 24	97.60 121	94.52 34	96.27 58	96.12 161	87.65 94	99.18 112	96.20 33	94.82 164	98.91 85

OpenMVS	89.19 12	92.86 147	91.68 162	96.40 86	95.34 207	92.73 71	98.27 25	98.12 50	84.86 267	85.78 275	97.75 74	78.89 231	99.74 29	87.50 204	98.65 82	96.73 187

Vis-MVSNet		95.23 75	94.81 76	96.51 78	97.18 126	91.58 104	98.26 26	98.12 50	94.38 38	94.90 96	98.15 48	82.28 172	98.92 138	91.45 136	98.58 85	99.01 75
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
ACMMP		96.27 50	95.93 51	97.28 51	99.24 24	92.62 74	98.25 27	98.81 3	92.99 77	94.56 101	98.39 27	88.96 76	99.85 13	94.57 78	97.63 106	99.36 47
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
MVSFormer		95.37 70	95.16 70	95.99 110	96.34 170	91.21 117	98.22 28	97.57 124	91.42 123	96.22 59	97.32 100	86.20 115	97.92 236	94.07 83	99.05 68	98.85 91

test_djsdf		93.07 136	92.76 126	94.00 194	93.49 282	88.70 197	98.22 28	97.57 124	91.42 123	90.08 198	95.55 192	82.85 159	97.92 236	94.07 83	91.58 208	95.40 236

DVP-MVS		97.91 1	97.81 1	98.22 7	99.45 2	95.36 8	98.21 30	97.85 99	94.92 18	98.73 4	98.87 4	95.08 2	99.84 16	97.52 2	99.67 6	99.48 33

test_0728_SECOND				98.51 1	99.45 2	95.93 2	98.21 30	98.28 26					99.86 7	97.52 2	99.67 6	99.75 3

PHI-MVS		96.77 35	96.46 40	97.71 33	98.40 65	94.07 36	98.21 30	98.45 15	89.86 165	97.11 34	98.01 56	92.52 26	99.69 40	96.03 38	99.53 20	99.36 47

#test#		97.02 22	96.75 28	97.83 19	99.42 6	94.12 34	98.15 33	98.32 20	92.57 94	97.18 28	98.29 42	92.08 32	99.83 19	95.12 61	99.59 13	99.54 23

FC-MVSNet-test		93.94 111	93.57 104	95.04 151	95.48 201	91.45 109	98.12 34	98.71 5	93.37 63	90.23 187	96.70 128	87.66 93	97.85 242	91.49 134	90.39 228	95.83 213

CS-MVS		95.80 61	95.65 57	96.24 100	97.32 121	91.43 110	98.10 35	97.91 90	93.38 62	95.16 94	94.57 230	90.21 66	98.98 134	95.53 55	98.67 81	98.30 134

FIs		94.09 105	93.70 100	95.27 143	95.70 195	92.03 91	98.10 35	98.68 7	93.36 65	90.39 184	96.70 128	87.63 95	97.94 232	92.25 113	90.50 227	95.84 212

Vis-MVSNet (Re-imp)		94.15 101	93.88 96	94.95 158	97.61 114	87.92 216	98.10 35	95.80 241	92.22 99	93.02 133	97.45 96	84.53 134	97.91 239	88.24 185	97.97 98	99.02 71

VDDNet		93.05 137	92.07 148	96.02 108	96.84 142	90.39 147	98.08 38	95.85 239	86.22 250	95.79 76	98.46 18	67.59 301	99.19 110	94.92 68	94.85 162	98.47 119

TSAR-MVS + MP.		97.42 7	97.33 8	97.69 34	99.25 23	94.24 29	98.07 39	97.85 99	93.72 52	98.57 7	98.35 29	93.69 11	99.40 96	97.06 5	99.46 30	99.44 38
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
Anonymous20231211		90.63 227	89.42 237	94.27 184	98.24 79	89.19 187	98.05 40	97.89 91	79.95 305	88.25 241	94.96 210	72.56 277	98.13 199	89.70 158	85.14 272	95.49 226

WR-MVS_H		92.00 177	91.35 172	93.95 199	95.09 225	89.47 170	98.04 41	98.68 7	91.46 121	88.34 236	94.68 225	85.86 119	97.56 264	85.77 234	84.24 285	94.82 265

Anonymous20240529		91.98 178	90.73 194	95.73 121	98.14 89	89.40 174	97.99 42	97.72 109	79.63 307	93.54 121	97.41 98	69.94 292	99.56 71	91.04 142	91.11 216	98.22 135

SR-MVS		97.01 24	96.86 19	97.47 43	99.09 31	93.27 59	97.98 43	98.07 63	93.75 51	97.45 22	98.48 17	91.43 47	99.59 58	96.22 28	99.27 50	99.54 23

APD-MVS_3200maxsize		96.81 33	96.71 30	97.12 60	99.01 37	92.31 81	97.98 43	98.06 66	93.11 74	97.44 23	98.55 14	90.93 56	99.55 72	96.06 36	99.25 52	99.51 28

tttt0517		92.96 141	92.33 144	94.87 161	97.11 130	87.16 233	97.97 45	92.09 314	90.63 148	93.88 115	97.01 115	76.50 252	99.06 128	90.29 151	95.45 153	98.38 129

SMA-MVS		97.35 9	97.03 11	98.30 5	99.06 35	95.42 7	97.94 46	98.18 41	90.57 154	98.85 3	98.94 1	93.33 13	99.83 19	96.72 15	99.68 4	99.63 9

LFMVS		93.60 121	92.63 132	96.52 75	98.13 90	91.27 114	97.94 46	93.39 306	90.57 154	96.29 57	98.31 38	69.00 294	99.16 114	94.18 82	95.87 145	99.12 65

SD-MVS		97.41 8	97.53 5	97.06 61	98.57 59	94.46 21	97.92 48	98.14 47	94.82 24	99.01 1	98.55 14	94.18 7	97.41 276	96.94 7	99.64 8	99.32 49
Zhenlong Yuan, Jiakai Cao, Zhaoxin Li, Hao Jiang and Zhaoqi Wang: SD-MVS: Segmentation-driven Deformation Multi-View Stereo with Spherical Refinement and EM optimization. AAAI2024
abl_6		96.40 46	96.21 47	96.98 64	98.89 42	92.20 86	97.89 49	98.03 75	93.34 66	97.22 27	98.42 23	87.93 90	99.72 33	95.10 62	99.07 67	99.02 71

UGNet		94.04 108	93.28 117	96.31 93	96.85 141	91.19 120	97.88 50	97.68 114	94.40 36	93.00 134	96.18 158	73.39 276	99.61 53	91.72 127	98.46 86	98.13 138
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
MTMP							97.86 51	82.03 332

alignmvs		95.87 60	95.23 68	97.78 25	97.56 118	95.19 12	97.86 51	97.17 169	94.39 37	96.47 52	96.40 150	85.89 118	99.20 109	96.21 32	95.11 160	98.95 81

VPA-MVSNet		93.24 131	92.48 141	95.51 134	95.70 195	92.39 80	97.86 51	98.66 9	92.30 98	92.09 154	95.37 198	80.49 200	98.40 180	93.95 86	85.86 264	95.75 220

EPNet		95.20 77	94.56 83	97.14 59	92.80 296	92.68 72	97.85 54	94.87 284	96.64 1	92.46 142	97.80 72	86.23 113	99.65 46	93.72 93	98.62 83	99.10 67
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
PS-CasMVS		91.55 191	90.84 190	93.69 214	94.96 230	88.28 205	97.84 55	98.24 31	91.46 121	88.04 244	95.80 175	79.67 216	97.48 269	87.02 214	84.54 283	95.31 242

ETV-MVS		95.53 68	95.47 60	95.71 123	97.06 135	89.63 161	97.82 56	97.87 95	93.57 56	93.92 114	95.04 209	90.61 61	98.95 136	94.62 77	98.68 80	98.54 108

CP-MVSNet		91.89 180	91.24 179	93.82 207	95.05 226	88.57 199	97.82 56	98.19 39	91.70 115	88.21 242	95.76 180	81.96 178	97.52 267	87.86 191	84.65 280	95.37 239

API-MVS		94.84 89	94.49 87	95.90 112	97.90 101	92.00 93	97.80 58	97.48 132	89.19 182	94.81 98	96.71 126	88.84 78	99.17 113	88.91 178	98.76 78	96.53 190

pm-mvs1		90.72 224	89.65 235	93.96 198	94.29 261	89.63 161	97.79 59	96.82 202	89.07 184	86.12 274	95.48 196	78.61 233	97.78 249	86.97 215	81.67 302	94.46 279

testtj		96.93 28	96.56 35	98.05 12	99.10 30	94.66 19	97.78 60	98.22 35	92.74 90	97.59 18	98.20 47	91.96 37	99.86 7	94.21 81	99.25 52	99.63 9

PEN-MVS		91.20 207	90.44 201	93.48 223	94.49 252	87.91 218	97.76 61	98.18 41	91.29 126	87.78 249	95.74 182	80.35 203	97.33 280	85.46 238	82.96 298	95.19 251

PS-MVSNAJss		93.74 117	93.51 108	94.44 177	93.91 269	89.28 183	97.75 62	97.56 127	92.50 95	89.94 200	96.54 143	88.65 81	98.18 196	93.83 92	90.90 221	95.86 209

HQP_MVS		93.78 116	93.43 112	94.82 162	96.21 174	89.99 152	97.74 63	97.51 130	94.85 20	91.34 166	96.64 133	81.32 188	98.60 167	93.02 106	92.23 196	95.86 209

plane_prior2							97.74 63		94.85 20

9.14			96.75 28		98.93 38		97.73 65	98.23 34	91.28 129	97.88 16	98.44 20	93.00 15	99.65 46	95.76 44	99.47 29

jajsoiax		92.42 159	91.89 156	94.03 193	93.33 288	88.50 201	97.73 65	97.53 128	92.00 110	88.85 229	96.50 145	75.62 261	98.11 203	93.88 90	91.56 209	95.48 227

TransMVSNet (Re)		88.94 252	87.56 259	93.08 239	94.35 257	88.45 203	97.73 65	95.23 265	87.47 230	84.26 288	95.29 200	79.86 213	97.33 280	79.44 290	74.44 318	93.45 300

VDD-MVS		93.82 114	93.08 119	96.02 108	97.88 102	89.96 156	97.72 68	95.85 239	92.43 96	95.86 73	98.44 20	68.42 298	99.39 97	96.31 24	94.85 162	98.71 102

APD-MVS		96.95 26	96.60 32	98.01 13	99.03 36	94.93 16	97.72 68	98.10 55	91.50 119	98.01 12	98.32 37	92.33 28	99.58 61	94.85 69	99.51 23	99.53 27
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
thres100view900		92.43 158	91.58 165	94.98 155	97.92 99	89.37 176	97.71 70	94.66 286	92.20 101	93.31 128	94.90 214	78.06 242	99.08 125	81.40 275	94.08 173	96.48 193

v7n		90.76 220	89.86 224	93.45 226	93.54 279	87.60 224	97.70 71	97.37 155	88.85 193	87.65 251	94.08 254	81.08 190	98.10 204	84.68 247	83.79 292	94.66 275

MSLP-MVS++		96.94 27	97.06 10	96.59 73	98.72 46	91.86 96	97.67 72	98.49 12	94.66 31	97.24 26	98.41 26	92.31 30	98.94 137	96.61 18	99.46 30	98.96 79

MAR-MVS		94.22 99	93.46 110	96.51 78	98.00 94	92.19 87	97.67 72	97.47 135	88.13 216	93.00 134	95.84 172	84.86 130	99.51 82	87.99 189	98.17 94	97.83 154
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
LS3D		93.57 123	92.61 134	96.47 81	97.59 116	91.61 101	97.67 72	97.72 109	85.17 262	90.29 186	98.34 32	84.60 132	99.73 30	83.85 258	98.27 90	98.06 143

UA-Net		95.95 58	95.53 58	97.20 58	97.67 110	92.98 66	97.65 75	98.13 48	94.81 25	96.61 44	98.35 29	88.87 77	99.51 82	90.36 149	97.35 116	99.11 66

thres600view7		92.49 157	91.60 164	95.18 146	97.91 100	89.47 170	97.65 75	94.66 286	92.18 105	93.33 127	94.91 213	78.06 242	99.10 120	81.61 272	94.06 176	96.98 177

PGM-MVS		96.81 33	96.53 36	97.65 36	99.35 17	93.53 51	97.65 75	98.98 1	92.22 99	97.14 31	98.44 20	91.17 52	99.85 13	94.35 79	99.46 30	99.57 17

LPG-MVS_test		92.94 143	92.56 135	94.10 189	96.16 179	88.26 206	97.65 75	97.46 137	91.29 126	90.12 194	97.16 107	79.05 224	98.73 155	92.25 113	91.89 204	95.31 242

DTE-MVSNet		90.56 228	89.75 231	93.01 240	93.95 267	87.25 228	97.64 79	97.65 117	90.74 141	87.12 260	95.68 186	79.97 211	97.00 292	83.33 259	81.66 303	94.78 271

mvs_tets		92.31 164	91.76 158	93.94 202	93.41 284	88.29 204	97.63 80	97.53 128	92.04 108	88.76 230	96.45 147	74.62 266	98.09 207	93.91 88	91.48 210	95.45 231

ACMMP_NAP		97.20 12	96.86 19	98.23 6	99.09 31	95.16 13	97.60 81	98.19 39	92.82 87	97.93 14	98.74 8	91.60 44	99.86 7	96.26 25	99.52 21	99.67 6

Anonymous202405211		92.07 175	90.83 191	95.76 116	98.19 86	88.75 195	97.58 82	95.00 274	86.00 253	93.64 118	97.45 96	66.24 308	99.53 77	90.68 146	92.71 189	99.01 75

ACMM	89.79 8	92.96 141	92.50 140	94.35 181	96.30 172	88.71 196	97.58 82	97.36 157	91.40 125	90.53 180	96.65 132	79.77 214	98.75 154	91.24 140	91.64 206	95.59 225
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
tfpnnormal		89.70 247	88.40 251	93.60 217	95.15 221	90.10 148	97.56 84	98.16 44	87.28 236	86.16 273	94.63 228	77.57 247	98.05 214	74.48 305	84.59 282	92.65 305

HPM-MVS++		97.34 10	96.97 14	98.47 2	99.08 33	96.16 1	97.55 85	97.97 87	95.59 4	96.61 44	97.89 60	92.57 24	99.84 16	95.95 39	99.51 23	99.40 42

TranMVSNet+NR-MVSNet		92.50 155	91.63 163	95.14 148	94.76 241	92.07 89	97.53 86	98.11 53	92.90 85	89.56 213	96.12 161	83.16 150	97.60 263	89.30 167	83.20 297	95.75 220

anonymousdsp		92.16 172	91.55 166	93.97 197	92.58 300	89.55 166	97.51 87	97.42 150	89.42 176	88.40 235	94.84 217	80.66 197	97.88 241	91.87 124	91.28 214	94.48 278

VNet		95.89 59	95.45 61	97.21 57	98.07 93	92.94 67	97.50 88	98.15 45	93.87 47	97.52 19	97.61 87	85.29 124	99.53 77	95.81 43	95.27 156	99.16 58

GBi-Net		91.35 200	90.27 207	94.59 170	96.51 160	91.18 121	97.50 88	96.93 194	88.82 196	89.35 219	94.51 232	73.87 270	97.29 282	86.12 227	88.82 240	95.31 242

test1		91.35 200	90.27 207	94.59 170	96.51 160	91.18 121	97.50 88	96.93 194	88.82 196	89.35 219	94.51 232	73.87 270	97.29 282	86.12 227	88.82 240	95.31 242

FMVSNet1		89.88 244	88.31 252	94.59 170	95.41 203	91.18 121	97.50 88	96.93 194	86.62 244	87.41 255	94.51 232	65.94 310	97.29 282	83.04 262	87.43 254	95.31 242

thisisatest0530		93.03 138	92.21 146	95.49 136	97.07 132	89.11 189	97.49 92	92.19 313	90.16 160	94.09 109	96.41 149	76.43 255	99.05 129	90.38 148	95.68 151	98.31 133

EIA-MVS		96.02 55	95.89 53	96.40 86	97.16 127	92.44 79	97.47 93	97.77 102	94.55 33	96.48 51	94.51 232	91.23 51	98.92 138	95.65 48	98.19 92	97.82 155

XXY-MVS		92.16 172	91.23 180	94.95 158	94.75 242	90.94 129	97.47 93	97.43 149	89.14 183	88.90 227	96.43 148	79.71 215	98.24 189	89.56 163	87.68 251	95.67 224

114514_t		93.95 110	93.06 120	96.63 70	99.07 34	91.61 101	97.46 95	97.96 88	77.99 313	93.00 134	97.57 90	86.14 117	99.33 101	89.22 171	99.15 60	98.94 82

tfpn200view9		92.38 161	91.52 168	94.95 158	97.85 103	89.29 181	97.41 96	94.88 281	92.19 103	93.27 130	94.46 237	78.17 239	99.08 125	81.40 275	94.08 173	96.48 193

thres400		92.42 159	91.52 168	95.12 150	97.85 103	89.29 181	97.41 96	94.88 281	92.19 103	93.27 130	94.46 237	78.17 239	99.08 125	81.40 275	94.08 173	96.98 177

FMVSNet2		91.31 203	90.08 216	94.99 153	96.51 160	92.21 84	97.41 96	96.95 192	88.82 196	88.62 232	94.75 222	73.87 270	97.42 275	85.20 242	88.55 246	95.35 240

DeepC-MVS_fast	93.89 2	96.93 28	96.64 31	97.78 25	98.64 54	94.30 25	97.41 96	98.04 73	94.81 25	96.59 46	98.37 28	91.24 50	99.64 52	95.16 59	99.52 21	99.42 41
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
UniMVSNet (Re)		93.31 129	92.55 136	95.61 128	95.39 204	93.34 58	97.39 100	98.71 5	93.14 73	90.10 196	94.83 218	87.71 92	98.03 218	91.67 132	83.99 287	95.46 230

NR-MVSNet		92.34 162	91.27 178	95.53 133	94.95 231	93.05 63	97.39 100	98.07 63	92.65 93	84.46 285	95.71 183	85.00 128	97.77 251	89.71 157	83.52 294	95.78 216

DP-MVS		92.76 152	91.51 170	96.52 75	98.77 44	90.99 126	97.38 102	96.08 233	82.38 290	89.29 222	97.87 63	83.77 142	99.69 40	81.37 278	96.69 133	98.89 88

ACMP	89.59 10	92.62 154	92.14 147	94.05 192	96.40 167	88.20 209	97.36 103	97.25 166	91.52 118	88.30 238	96.64 133	78.46 235	98.72 158	91.86 125	91.48 210	95.23 249
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
pmmvs6		87.81 268	86.19 271	92.69 247	91.32 308	86.30 246	97.34 104	96.41 220	80.59 304	84.05 292	94.37 241	67.37 303	97.67 257	84.75 246	79.51 309	94.09 291

v8		91.29 205	90.53 200	93.57 220	94.15 262	88.12 213	97.34 104	97.06 182	88.99 187	88.32 237	94.26 249	83.08 153	98.01 220	87.62 201	83.92 290	94.57 277

NCCC		97.30 11	97.03 11	98.11 10	98.77 44	95.06 15	97.34 104	98.04 73	95.96 2	97.09 35	97.88 62	93.18 14	99.71 34	95.84 42	99.17 59	99.56 19

v10		91.04 213	90.23 210	93.49 222	94.12 263	88.16 212	97.32 107	97.08 179	88.26 211	88.29 239	94.22 250	82.17 175	97.97 225	86.45 221	84.12 286	94.33 283

V42		91.58 189	90.87 186	93.73 210	94.05 266	88.50 201	97.32 107	96.97 191	88.80 199	89.71 206	94.33 243	82.54 166	98.05 214	89.01 176	85.07 274	94.64 276

DeepC-MVS	93.07 3	96.06 53	95.66 56	97.29 50	97.96 95	93.17 61	97.30 109	98.06 66	93.92 46	93.38 126	98.66 9	86.83 107	99.73 30	95.60 54	99.22 55	98.96 79
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
casdiffmvs		95.64 64	95.49 59	96.08 104	96.76 149	90.45 144	97.29 110	97.44 146	94.00 44	95.46 90	97.98 58	87.52 98	98.73 155	95.64 49	97.33 117	99.08 68

CNVR-MVS		97.68 4	97.44 7	98.37 4	98.90 41	95.86 3	97.27 111	98.08 58	95.81 3	97.87 17	98.31 38	94.26 6	99.68 42	97.02 6	99.49 27	99.57 17

PVSNet_Blended_VisFu		95.27 73	94.91 75	96.38 89	98.20 84	90.86 132	97.27 111	98.25 30	90.21 158	94.18 108	97.27 102	87.48 99	99.73 30	93.53 94	97.77 104	98.55 107

mvs-test1		93.63 120	93.69 101	93.46 225	96.02 185	84.61 267	97.24 113	96.72 205	93.85 48	92.30 149	95.76 180	83.08 153	98.89 143	91.69 130	96.54 136	96.87 183

MTAPA		97.08 18	96.78 26	97.97 16	99.37 13	94.42 23	97.24 113	98.08 58	95.07 15	96.11 62	98.59 10	90.88 58	99.90 1	96.18 34	99.50 25	99.58 15

plane_prior						89.99 152	97.24 113		94.06 43						92.16 200

PAPM_NR		95.01 80	94.59 82	96.26 98	98.89 42	90.68 138	97.24 113	97.73 106	91.80 113	92.93 139	96.62 140	89.13 75	99.14 117	89.21 172	97.78 103	98.97 78

ACMH	87.59 16	90.53 229	89.42 237	93.87 205	96.21 174	87.92 216	97.24 113	96.94 193	88.45 206	83.91 293	96.27 156	71.92 278	98.62 166	84.43 251	89.43 236	95.05 254
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
UniMVSNet_ETH3D		91.34 202	90.22 212	94.68 169	94.86 237	87.86 219	97.23 118	97.46 137	87.99 217	89.90 201	96.92 119	66.35 306	98.23 190	90.30 150	90.99 219	97.96 144

VPNet		92.23 170	91.31 175	94.99 153	95.56 198	90.96 128	97.22 119	97.86 98	92.96 83	90.96 176	96.62 140	75.06 263	98.20 193	91.90 122	83.65 293	95.80 215

DPE-MVS		97.86 2	97.65 3	98.47 2	99.17 28	95.78 4	97.21 120	98.35 19	95.16 13	98.71 6	98.80 6	95.05 4	99.89 3	96.70 16	99.73 1	99.73 5

baseline1		92.82 150	91.90 155	95.55 132	97.20 125	90.77 136	97.19 121	94.58 289	92.20 101	92.36 146	96.34 153	84.16 139	98.21 192	89.20 173	83.90 291	97.68 160

F-COLMAP		93.58 122	92.98 121	95.37 142	98.40 65	88.98 191	97.18 122	97.29 163	87.75 224	90.49 181	97.10 111	85.21 125	99.50 84	86.70 217	96.72 132	97.63 161

UniMVSNet_NR-MVSNet		93.37 127	92.67 131	95.47 139	95.34 207	92.83 68	97.17 123	98.58 10	92.98 82	90.13 192	95.80 175	88.37 86	97.85 242	91.71 128	83.93 288	95.73 222

DU-MVS		92.90 145	92.04 149	95.49 136	94.95 231	92.83 68	97.16 124	98.24 31	93.02 76	90.13 192	95.71 183	83.47 145	97.85 242	91.71 128	83.93 288	95.78 216

baseline		95.58 66	95.42 63	96.08 104	96.78 146	90.41 146	97.16 124	97.45 142	93.69 55	95.65 84	97.85 66	87.29 102	98.68 160	95.66 45	97.25 120	99.13 62

zzz-MVS		97.07 19	96.77 27	97.97 16	99.37 13	94.42 23	97.15 126	98.08 58	95.07 15	96.11 62	98.59 10	90.88 58	99.90 1	96.18 34	99.50 25	99.58 15

Effi-MVS+-dtu		93.08 135	93.21 118	92.68 248	96.02 185	83.25 281	97.14 127	96.72 205	93.85 48	91.20 175	93.44 274	83.08 153	98.30 187	91.69 130	95.73 149	96.50 192

MCST-MVS		97.18 13	96.84 21	98.20 8	99.30 20	95.35 10	97.12 128	98.07 63	93.54 60	96.08 64	97.69 77	93.86 9	99.71 34	96.50 22	99.39 39	99.55 21

MVSTER		93.20 132	92.81 125	94.37 180	96.56 157	89.59 164	97.06 129	97.12 173	91.24 130	91.30 169	95.96 166	82.02 177	98.05 214	93.48 97	90.55 225	95.47 229

Fast-Effi-MVS+-dtu		92.29 166	91.99 152	93.21 236	95.27 213	85.52 255	97.03 130	96.63 216	92.09 106	89.11 226	95.14 206	80.33 204	98.08 208	87.54 203	94.74 167	96.03 207

DP-MVS Recon		95.68 63	95.12 72	97.37 46	99.19 27	94.19 30	97.03 130	98.08 58	88.35 209	95.09 95	97.65 81	89.97 70	99.48 86	92.08 120	98.59 84	98.44 124

save fliter					98.91 40	94.28 26	97.02 132	98.02 76	95.35 8

CANet		96.39 47	96.02 50	97.50 42	97.62 113	93.38 55	97.02 132	97.96 88	95.42 7	94.86 97	97.81 70	87.38 101	99.82 22	96.88 9	99.20 57	99.29 51

FMVSNet3		91.78 182	90.69 196	95.03 152	96.53 159	92.27 83	97.02 132	96.93 194	89.79 170	89.35 219	94.65 227	77.01 250	97.47 271	86.12 227	88.82 240	95.35 240

Baseline_NR-MVSNet		91.20 207	90.62 197	92.95 242	93.83 272	88.03 214	97.01 135	95.12 270	88.42 207	89.70 207	95.13 207	83.47 145	97.44 273	89.66 160	83.24 296	93.37 301

ACMH+	87.92 14	90.20 237	89.18 242	93.25 233	96.48 163	86.45 244	96.99 136	96.68 211	88.83 195	84.79 284	96.22 157	70.16 291	98.53 172	84.42 252	88.04 248	94.77 272

OurMVSNet-221017-0		90.51 230	90.19 214	91.44 275	93.41 284	81.25 292	96.98 137	96.28 224	91.68 116	86.55 270	96.30 154	74.20 269	97.98 222	88.96 177	87.40 256	95.09 252

MP-MVS-pluss		96.70 37	96.27 45	97.98 15	99.23 26	94.71 18	96.96 138	98.06 66	90.67 144	95.55 86	98.78 7	91.07 53	99.86 7	96.58 20	99.55 18	99.38 45
MP-MVS-pluss: MP-MVS-pluss. MP-MVS-pluss
Regformer-3		96.85 31	96.80 25	97.01 62	98.34 70	92.02 92	96.96 138	97.76 103	95.01 17	97.08 36	98.42 23	91.71 41	99.54 74	96.80 11	99.13 62	99.48 33

Regformer-4		96.97 25	96.87 18	97.25 53	98.34 70	92.66 73	96.96 138	98.01 79	95.12 14	97.14 31	98.42 23	91.82 39	99.61 53	96.90 8	99.13 62	99.50 29

v2v482		91.59 188	90.85 189	93.80 208	93.87 271	88.17 211	96.94 141	96.88 199	89.54 172	89.53 214	94.90 214	81.70 184	98.02 219	89.25 170	85.04 276	95.20 250

LCM-MVSNet-Re		92.50 155	92.52 139	92.44 251	96.82 145	81.89 289	96.92 142	93.71 303	92.41 97	84.30 287	94.60 229	85.08 127	97.03 288	91.51 133	97.36 115	98.40 127

COLMAP_ROB	87.81 15	90.40 232	89.28 240	93.79 209	97.95 96	87.13 234	96.92 142	95.89 238	82.83 288	86.88 268	97.18 106	73.77 273	99.29 105	78.44 294	93.62 181	94.95 255
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
EI-MVSNet-Vis-set		96.51 43	96.47 39	96.63 70	98.24 79	91.20 119	96.89 144	97.73 106	94.74 29	96.49 50	98.49 16	90.88 58	99.58 61	96.44 23	98.32 89	99.13 62

EI-MVSNet-UG-set		96.34 48	96.30 44	96.47 81	98.20 84	90.93 130	96.86 145	97.72 109	94.67 30	96.16 61	98.46 18	90.43 63	99.58 61	96.23 27	97.96 99	98.90 86

test_yl		94.78 91	94.23 92	96.43 84	97.74 107	91.22 115	96.85 146	97.10 176	91.23 131	95.71 78	96.93 116	84.30 136	99.31 103	93.10 104	95.12 158	98.75 96

DCV-MVSNet		94.78 91	94.23 92	96.43 84	97.74 107	91.22 115	96.85 146	97.10 176	91.23 131	95.71 78	96.93 116	84.30 136	99.31 103	93.10 104	95.12 158	98.75 96

v1144		91.37 199	90.60 198	93.68 215	93.89 270	88.23 208	96.84 148	97.03 187	88.37 208	89.69 208	94.39 239	82.04 176	97.98 222	87.80 193	85.37 268	94.84 262

v144192		91.06 212	90.28 206	93.39 227	93.66 277	87.23 230	96.83 149	97.07 180	87.43 231	89.69 208	94.28 246	81.48 186	98.00 221	87.18 212	84.92 278	94.93 259

Regformer-1		97.10 17	96.96 15	97.54 41	98.32 73	93.48 52	96.83 149	97.99 85	95.20 12	97.46 21	98.25 45	92.48 27	99.58 61	96.79 13	99.29 47	99.55 21

Regformer-2		97.16 15	96.99 13	97.67 35	98.32 73	93.84 41	96.83 149	98.10 55	95.24 10	97.49 20	98.25 45	92.57 24	99.61 53	96.80 11	99.29 47	99.56 19

Fast-Effi-MVS+		93.46 125	92.75 128	95.59 129	96.77 147	90.03 149	96.81 152	97.13 172	88.19 212	91.30 169	94.27 247	86.21 114	98.63 164	87.66 199	96.46 139	98.12 139

TSAR-MVS + GP.		96.69 38	96.49 38	97.27 52	98.31 75	93.39 54	96.79 153	96.72 205	94.17 41	97.44 23	97.66 80	92.76 18	99.33 101	96.86 10	97.76 105	99.08 68

TAPA-MVS	90.10 7	92.30 165	91.22 181	95.56 130	98.33 72	89.60 163	96.79 153	97.65 117	81.83 294	91.52 161	97.23 105	87.94 89	98.91 140	71.31 316	98.37 88	98.17 137
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
v148		90.99 214	90.38 203	92.81 244	93.83 272	85.80 251	96.78 155	96.68 211	89.45 175	88.75 231	93.93 257	82.96 157	97.82 246	87.83 192	83.25 295	94.80 268

v1921920		90.85 218	90.03 220	93.29 232	93.55 278	86.96 237	96.74 156	97.04 185	87.36 233	89.52 215	94.34 242	80.23 206	97.97 225	86.27 222	85.21 271	94.94 257

v1192		91.07 211	90.23 210	93.58 219	93.70 275	87.82 220	96.73 157	97.07 180	87.77 223	89.58 211	94.32 244	80.90 195	97.97 225	86.52 219	85.48 266	94.95 255

PVSNet_BlendedMVS		94.06 106	93.92 95	94.47 176	98.27 76	89.46 172	96.73 157	98.36 16	90.17 159	94.36 104	95.24 203	88.02 87	99.58 61	93.44 98	90.72 223	94.36 282

TAMVS		94.01 109	93.46 110	95.64 125	96.16 179	90.45 144	96.71 159	96.89 198	89.27 180	93.46 124	96.92 119	87.29 102	97.94 232	88.70 181	95.74 148	98.53 109

MVS_Test		94.89 87	94.62 81	95.68 124	96.83 144	89.55 166	96.70 160	97.17 169	91.17 133	95.60 85	96.11 163	87.87 91	98.76 153	93.01 108	97.17 123	98.72 100

SixPastTwentyTwo		89.15 251	88.54 250	90.98 280	93.49 282	80.28 302	96.70 160	94.70 285	90.78 140	84.15 290	95.57 190	71.78 280	97.71 255	84.63 248	85.07 274	94.94 257

EPNet_dtu		91.71 183	91.28 177	92.99 241	93.76 274	83.71 275	96.69 162	95.28 261	93.15 72	87.02 264	95.95 167	83.37 148	97.38 278	79.46 289	96.84 127	97.88 150
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
PLC	91.00 6	94.11 104	93.43 112	96.13 103	98.58 58	91.15 124	96.69 162	97.39 152	87.29 235	91.37 164	96.71 126	88.39 85	99.52 81	87.33 207	97.13 124	97.73 157
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
testgi		87.97 265	87.21 264	90.24 292	92.86 294	80.76 294	96.67 164	94.97 277	91.74 114	85.52 277	95.83 173	62.66 316	94.47 317	76.25 301	88.36 247	95.48 227

OPM-MVS		93.28 130	92.76 126	94.82 162	94.63 248	90.77 136	96.65 165	97.18 167	93.72 52	91.68 159	97.26 103	79.33 221	98.63 164	92.13 117	92.28 195	95.07 253

HQP-NCC					95.86 188		96.65 165		93.55 57	90.14 188

ACMP_Plane					95.86 188		96.65 165		93.55 57	90.14 188

HQP-MVS		93.19 133	92.74 129	94.54 175	95.86 188	89.33 178	96.65 165	97.39 152	93.55 57	90.14 188	95.87 170	80.95 191	98.50 175	92.13 117	92.10 201	95.78 216

EU-MVSNet		88.72 258	88.90 245	88.20 300	93.15 291	74.21 320	96.63 169	94.22 298	85.18 261	87.32 258	95.97 165	76.16 256	94.98 315	85.27 240	86.17 261	95.41 232

v1240		90.70 225	89.85 225	93.23 234	93.51 281	86.80 238	96.61 170	97.02 188	87.16 238	89.58 211	94.31 245	79.55 218	97.98 222	85.52 237	85.44 267	94.90 260

K. test v3		87.64 269	86.75 268	90.32 291	93.02 293	79.48 308	96.61 170	92.08 315	90.66 146	80.25 308	94.09 253	67.21 304	96.65 299	85.96 232	80.83 306	94.83 263

thres200		92.23 170	91.39 171	94.75 168	97.61 114	89.03 190	96.60 172	95.09 271	92.08 107	93.28 129	94.00 255	78.39 237	99.04 131	81.26 279	94.18 172	96.19 198

WTY-MVS		94.71 93	94.02 94	96.79 66	97.71 109	92.05 90	96.59 173	97.35 158	90.61 150	94.64 100	96.93 116	86.41 112	99.39 97	91.20 141	94.71 168	98.94 82

CNLPA		94.28 98	93.53 107	96.52 75	98.38 68	92.55 76	96.59 173	96.88 199	90.13 161	91.91 156	97.24 104	85.21 125	99.09 123	87.64 200	97.83 101	97.92 147

AdaColmap		94.34 97	93.68 102	96.31 93	98.59 56	91.68 100	96.59 173	97.81 101	89.87 164	92.15 152	97.06 113	83.62 144	99.54 74	89.34 166	98.07 96	97.70 159

IterMVS-LS		92.29 166	91.94 154	93.34 230	96.25 173	86.97 236	96.57 176	97.05 183	90.67 144	89.50 216	94.80 220	86.59 108	97.64 260	89.91 153	86.11 263	95.40 236
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
AllTest		90.23 236	88.98 244	93.98 195	97.94 97	86.64 240	96.51 177	95.54 250	85.38 258	85.49 278	96.77 124	70.28 289	99.15 115	80.02 284	92.87 186	96.15 201

EI-MVSNet		93.03 138	92.88 124	93.48 223	95.77 193	86.98 235	96.44 178	97.12 173	90.66 146	91.30 169	97.64 84	86.56 109	98.05 214	89.91 153	90.55 225	95.41 232

CVMVSNet		91.23 206	91.75 159	89.67 296	95.77 193	74.69 319	96.44 178	94.88 281	85.81 254	92.18 151	97.64 84	79.07 223	95.58 311	88.06 188	95.86 146	98.74 98

OMC-MVS		95.09 79	94.70 80	96.25 99	98.46 61	91.28 113	96.43 180	97.57 124	92.04 108	94.77 99	97.96 59	87.01 106	99.09 123	91.31 138	96.77 129	98.36 131

test_prior4						93.66 47	96.42 181

Effi-MVS+		94.93 85	94.45 89	96.36 91	96.61 151	91.47 107	96.41 182	97.41 151	91.02 138	94.50 102	95.92 168	87.53 97	98.78 150	93.89 89	96.81 128	98.84 93

TEST9					98.70 47	94.19 30	96.41 182	98.02 76	88.17 214	96.03 65	97.56 92	92.74 19	99.59 58

train_agg		96.30 49	95.83 54	97.72 31	98.70 47	94.19 30	96.41 182	98.02 76	88.58 203	96.03 65	97.56 92	92.73 20	99.59 58	95.04 63	99.37 44	99.39 43

MVS_0304		88.79 256	87.57 258	92.46 250	94.65 246	86.15 249	96.40 185	97.17 169	86.44 246	88.02 245	91.71 300	56.68 322	97.03 288	84.47 250	92.58 192	94.19 288

WR-MVS		92.34 162	91.53 167	94.77 167	95.13 223	90.83 133	96.40 185	97.98 86	91.88 112	89.29 222	95.54 193	82.50 167	97.80 247	89.79 156	85.27 270	95.69 223

BH-untuned		92.94 143	92.62 133	93.92 204	97.22 123	86.16 248	96.40 185	96.25 227	90.06 162	89.79 205	96.17 160	83.19 149	98.35 184	87.19 211	97.27 119	97.24 174

TDRefinement		86.53 276	84.76 282	91.85 262	82.23 327	84.25 268	96.38 188	95.35 257	84.97 266	84.09 291	94.94 211	65.76 311	98.34 186	84.60 249	74.52 317	92.97 302

test_8					98.67 49	94.06 37	96.37 189	98.01 79	88.58 203	95.98 70	97.55 94	92.73 20	99.58 61

test_prior3		96.46 45	96.20 48	97.23 54	98.67 49	92.99 64	96.35 190	98.00 81	92.80 88	96.03 65	97.59 88	92.01 34	99.41 94	95.01 64	99.38 40	99.29 51

test_prior2							96.35 190		92.80 88	96.03 65	97.59 88	92.01 34		95.01 64	99.38 40

CDPH-MVS		95.97 57	95.38 64	97.77 27	98.93 38	94.44 22	96.35 190	97.88 93	86.98 240	96.65 43	97.89 60	91.99 36	99.47 87	92.26 111	99.46 30	99.39 43

CDS-MVSNet		94.14 103	93.54 106	95.93 111	96.18 177	91.46 108	96.33 193	97.04 185	88.97 189	93.56 119	96.51 144	87.55 96	97.89 240	89.80 155	95.95 143	98.44 124
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
sss		94.51 95	93.80 98	96.64 68	97.07 132	91.97 94	96.32 194	98.06 66	88.94 190	94.50 102	96.78 123	84.60 132	99.27 106	91.90 122	96.02 141	98.68 104

1112_ss		93.37 127	92.42 142	96.21 101	97.05 137	90.99 126	96.31 195	96.72 205	86.87 243	89.83 204	96.69 130	86.51 110	99.14 117	88.12 187	93.67 179	98.50 114

LTVRE_ROB	88.41 13	90.99 214	89.92 222	94.19 186	96.18 177	89.55 166	96.31 195	97.09 178	87.88 221	85.67 276	95.91 169	78.79 232	98.57 170	81.50 273	89.98 231	94.44 280
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
test_0402		86.46 277	84.79 281	91.45 274	95.02 228	85.55 254	96.29 197	94.89 280	80.90 299	82.21 297	93.97 256	68.21 299	97.29 282	62.98 324	88.68 245	91.51 315

agg_prior1		96.22 52	95.77 55	97.56 40	98.67 49	93.79 43	96.28 198	98.00 81	88.76 200	95.68 80	97.55 94	92.70 22	99.57 69	95.01 64	99.32 45	99.32 49

pmmvs5		89.86 245	88.87 246	92.82 243	92.86 294	86.23 247	96.26 199	95.39 254	84.24 274	87.12 260	94.51 232	74.27 268	97.36 279	87.61 202	87.57 252	94.86 261

xiu_mvs_v1_base_debu		95.01 80	94.76 77	95.75 118	96.58 154	91.71 97	96.25 200	97.35 158	92.99 77	96.70 39	96.63 137	82.67 162	99.44 91	96.22 28	97.46 109	96.11 204

xiu_mvs_v1_base		95.01 80	94.76 77	95.75 118	96.58 154	91.71 97	96.25 200	97.35 158	92.99 77	96.70 39	96.63 137	82.67 162	99.44 91	96.22 28	97.46 109	96.11 204

xiu_mvs_v1_base_debi		95.01 80	94.76 77	95.75 118	96.58 154	91.71 97	96.25 200	97.35 158	92.99 77	96.70 39	96.63 137	82.67 162	99.44 91	96.22 28	97.46 109	96.11 204

MVS_111021_LR		96.24 51	96.19 49	96.39 88	98.23 83	91.35 112	96.24 203	98.79 4	93.99 45	95.80 75	97.65 81	89.92 71	99.24 108	95.87 40	99.20 57	98.58 106

CANet_DTU		94.37 96	93.65 103	96.55 74	96.46 164	92.13 88	96.21 204	96.67 213	94.38 38	93.53 122	97.03 114	79.34 220	99.71 34	90.76 143	98.45 87	97.82 155

MVS_111021_HR		96.68 40	96.58 34	96.99 63	98.46 61	92.31 81	96.20 205	98.90 2	94.30 40	95.86 73	97.74 75	92.33 28	99.38 99	96.04 37	99.42 35	99.28 54

D2MVS		91.30 204	90.95 185	92.35 253	94.71 244	85.52 255	96.18 206	98.21 36	88.89 192	86.60 269	93.82 260	79.92 212	97.95 231	89.29 168	90.95 220	93.56 297

BH-RMVSNet		92.72 153	91.97 153	94.97 156	97.16 127	87.99 215	96.15 207	95.60 247	90.62 149	91.87 157	97.15 109	78.41 236	98.57 170	83.16 260	97.60 107	98.36 131

DI_MVS_plusplus_test		92.01 176	90.77 192	95.73 121	93.34 286	89.78 160	96.14 208	96.18 230	90.58 153	81.80 299	93.50 271	74.95 264	98.90 141	93.51 95	96.94 126	98.51 112

Anonymous20231206		87.09 273	86.14 272	89.93 295	91.22 309	80.35 299	96.11 209	95.35 257	83.57 283	84.16 289	93.02 278	73.54 275	95.61 309	72.16 313	86.14 262	93.84 295

jason		94.84 89	94.39 91	96.18 102	95.52 199	90.93 130	96.09 210	96.52 218	89.28 179	96.01 69	97.32 100	84.70 131	98.77 152	95.15 60	98.91 75	98.85 91
jason: jason.
EG-PatchMatch MVS		87.02 274	85.44 276	91.76 269	92.67 298	85.00 261	96.08 211	96.45 219	83.41 285	79.52 310	93.49 272	57.10 321	97.72 254	79.34 291	90.87 222	92.56 306

1314		92.81 151	92.03 150	95.14 148	95.33 210	89.52 169	96.04 212	97.44 146	87.72 225	86.25 272	95.33 199	83.84 141	98.79 149	89.26 169	97.05 125	97.11 175

1121		94.71 93	93.83 97	97.34 47	98.57 59	93.64 48	96.04 212	97.73 106	81.56 297	95.68 80	97.85 66	90.23 65	99.65 46	87.68 197	99.12 65	98.73 99

MVS		91.71 183	90.44 201	95.51 134	95.20 220	91.59 103	96.04 212	97.45 142	73.44 320	87.36 257	95.60 189	85.42 123	99.10 120	85.97 231	97.46 109	95.83 213

MG-MVS		95.61 65	95.38 64	96.31 93	98.42 64	90.53 141	96.04 212	97.48 132	93.47 61	95.67 83	98.10 49	89.17 74	99.25 107	91.27 139	98.77 77	99.13 62

DeepPCF-MVS	93.97 1	96.61 41	97.09 9	95.15 147	98.09 91	86.63 243	96.00 216	98.15 45	95.43 6	97.95 13	98.56 12	93.40 12	99.36 100	96.77 14	99.48 28	99.45 36

diffmvs		95.25 74	95.13 71	95.63 126	96.43 166	89.34 177	95.99 217	97.35 158	92.83 86	96.31 56	97.37 99	86.44 111	98.67 161	96.26 25	97.19 122	98.87 90

DELS-MVS		96.61 41	96.38 43	97.30 49	97.79 105	93.19 60	95.96 218	98.18 41	95.23 11	95.87 72	97.65 81	91.45 46	99.70 39	95.87 40	99.44 34	99.00 77
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
旧先验2							95.94 219		81.66 295	97.34 25			98.82 147	92.26 111

baseline2		91.63 187	90.86 187	93.94 202	94.33 258	86.32 245	95.92 220	91.64 318	89.37 177	86.94 265	94.69 224	81.62 185	98.69 159	88.64 182	94.57 169	96.81 185

test20.03		86.14 280	85.40 277	88.35 298	90.12 311	80.06 304	95.90 221	95.20 266	88.59 202	81.29 301	93.62 269	71.43 282	92.65 324	71.26 317	81.17 305	92.34 309

MVP-Stereo		90.74 223	90.08 216	92.71 246	93.19 290	88.20 209	95.86 222	96.27 225	86.07 252	84.86 283	94.76 221	77.84 245	97.75 252	83.88 257	98.01 97	92.17 312
Qingsong Yan: MVP-Stereo: A Parallel Multi-View Patchmatch Stereo Method with Dilation Matching for Photogrammetric Application.
DWT-MVSNet_test		90.76 220	89.89 223	93.38 228	95.04 227	83.70 276	95.85 223	94.30 297	88.19 212	90.46 182	92.80 280	73.61 274	98.50 175	88.16 186	90.58 224	97.95 146

lupinMVS		94.99 84	94.56 83	96.29 96	96.34 170	91.21 117	95.83 224	96.27 225	88.93 191	96.22 59	96.88 121	86.20 115	98.85 145	95.27 57	99.05 68	98.82 94

mvs_anonymous		93.82 114	93.74 99	94.06 191	96.44 165	85.41 257	95.81 225	97.05 183	89.85 167	90.09 197	96.36 152	87.44 100	97.75 252	93.97 85	96.69 133	99.02 71

新几何2							95.79 226

无先验							95.79 226	97.87 95	83.87 280				99.65 46	87.68 197		98.89 88

OpenMVS_ROB	81.14 20	84.42 288	82.28 290	90.83 282	90.06 312	84.05 272	95.73 228	94.04 300	73.89 319	80.17 309	91.53 302	59.15 319	97.64 260	66.92 322	89.05 239	90.80 318

原ACMM2							95.67 229

BH-w/o		92.14 174	91.75 159	93.31 231	96.99 139	85.73 252	95.67 229	95.69 243	88.73 201	89.26 224	94.82 219	82.97 156	98.07 211	85.26 241	96.32 140	96.13 203

TR-MVS		91.48 194	90.59 199	94.16 188	96.40 167	87.33 225	95.67 229	95.34 260	87.68 226	91.46 162	95.52 194	76.77 251	98.35 184	82.85 264	93.61 182	96.79 186

HY-MVS	89.66 9	93.87 112	92.95 122	96.63 70	97.10 131	92.49 78	95.64 232	96.64 214	89.05 185	93.00 134	95.79 178	85.77 121	99.45 90	89.16 175	94.35 170	97.96 144

RPSCF		90.75 222	90.86 187	90.42 290	96.84 142	76.29 317	95.61 233	96.34 222	83.89 278	91.38 163	97.87 63	76.45 253	98.78 150	87.16 213	92.23 196	96.20 197

MS-PatchMatch		90.27 234	89.77 229	91.78 267	94.33 258	84.72 266	95.55 234	96.73 204	86.17 251	86.36 271	95.28 202	71.28 283	97.80 247	84.09 253	98.14 95	92.81 304

PAPR		94.18 100	93.42 114	96.48 80	97.64 112	91.42 111	95.55 234	97.71 113	88.99 187	92.34 148	95.82 174	89.19 73	99.11 119	86.14 226	97.38 114	98.90 86

PatchFormer-LS_test		91.68 186	91.18 183	93.19 237	95.24 217	83.63 278	95.53 236	95.44 253	89.82 168	91.37 164	92.58 285	80.85 196	98.52 173	89.65 161	90.16 230	97.42 172

Test_1112_low_res		92.84 149	91.84 157	95.85 114	97.04 138	89.97 155	95.53 236	96.64 214	85.38 258	89.65 210	95.18 204	85.86 119	99.10 120	87.70 195	93.58 184	98.49 116

FMVSNet5		87.29 272	85.79 274	91.78 267	94.80 240	87.28 226	95.49 238	95.28 261	84.09 276	83.85 294	91.82 297	62.95 315	94.17 318	78.48 293	85.34 269	93.91 293

PVSNet_Blended		94.87 88	94.56 83	95.81 115	98.27 76	89.46 172	95.47 239	98.36 16	88.84 194	94.36 104	96.09 164	88.02 87	99.58 61	93.44 98	98.18 93	98.40 127

xiu_mvs_v2_base		95.32 72	95.29 67	95.40 141	97.22 123	90.50 142	95.44 240	97.44 146	93.70 54	96.46 53	96.18 158	88.59 84	99.53 77	94.79 75	97.81 102	96.17 199

ab-mvs		93.57 123	92.55 136	96.64 68	97.28 122	91.96 95	95.40 241	97.45 142	89.81 169	93.22 132	96.28 155	79.62 217	99.46 88	90.74 144	93.11 185	98.50 114

MIMVSNet1		84.93 287	83.05 288	90.56 288	89.56 316	84.84 265	95.40 241	95.35 257	83.91 277	80.38 306	92.21 295	57.23 320	93.34 322	70.69 319	82.75 301	93.50 298

ET-MVSNet_ETH3D		91.49 193	90.11 215	95.63 126	96.40 167	91.57 105	95.34 243	93.48 305	90.60 152	75.58 316	95.49 195	80.08 208	96.79 297	94.25 80	89.76 234	98.52 110

test222					98.24 79	92.21 84	95.33 244	97.60 121	79.22 309	95.25 91	97.84 69	88.80 79			99.15 60	98.72 100

XVG-ACMP-BASELINE		90.93 216	90.21 213	93.09 238	94.31 260	85.89 250	95.33 244	97.26 164	91.06 137	89.38 218	95.44 197	68.61 296	98.60 167	89.46 164	91.05 217	94.79 270

PS-MVSNAJ		95.37 70	95.33 66	95.49 136	97.35 120	90.66 139	95.31 246	97.48 132	93.85 48	96.51 49	95.70 185	88.65 81	99.65 46	94.80 73	98.27 90	96.17 199

XVG-OURS-SEG-HR		93.86 113	93.55 105	94.81 164	97.06 135	88.53 200	95.28 247	97.45 142	91.68 116	94.08 110	97.68 78	82.41 170	98.90 141	93.84 91	92.47 193	96.98 177

CLD-MVS		92.98 140	92.53 138	94.32 183	96.12 183	89.20 185	95.28 247	97.47 135	92.66 92	89.90 201	95.62 188	80.58 198	98.40 180	92.73 109	92.40 194	95.38 238
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
DPM-MVS		95.69 62	94.92 74	98.01 13	98.08 92	95.71 6	95.27 249	97.62 120	90.43 157	95.55 86	97.07 112	91.72 40	99.50 84	89.62 162	98.94 73	98.82 94

PatchMatch-RL		92.90 145	92.02 151	95.56 130	98.19 86	90.80 134	95.27 249	97.18 167	87.96 218	91.86 158	95.68 186	80.44 201	98.99 133	84.01 254	97.54 108	96.89 182

testdata1							95.26 251		93.10 75

test0.0.03 1		89.37 250	88.70 247	91.41 276	92.47 301	85.63 253	95.22 252	92.70 310	91.11 135	86.91 267	93.65 268	79.02 226	93.19 323	78.00 295	89.18 238	95.41 232

CHOSEN 1792x2688		94.15 101	93.51 108	96.06 106	98.27 76	89.38 175	95.18 253	98.48 14	85.60 257	93.76 117	97.11 110	83.15 151	99.61 53	91.33 137	98.72 79	99.19 56

IB-MVS	87.33 17	89.91 242	88.28 253	94.79 166	95.26 216	87.70 222	95.12 254	93.95 302	89.35 178	87.03 263	92.49 286	70.74 287	99.19 110	89.18 174	81.37 304	97.49 170
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
testing_2		87.33 271	85.03 279	94.22 185	87.77 322	89.32 180	94.97 255	97.11 175	89.22 181	71.64 319	88.73 309	55.16 324	97.94 232	91.95 121	88.73 244	95.41 232

DSMNet-mixed		86.34 278	86.12 273	87.00 305	89.88 314	70.43 323	94.93 256	90.08 323	77.97 314	85.42 280	92.78 281	74.44 267	93.96 319	74.43 306	95.14 157	96.62 189

XVG-OURS		93.72 118	93.35 115	94.80 165	97.07 132	88.61 198	94.79 257	97.46 137	91.97 111	93.99 111	97.86 65	81.74 183	98.88 144	92.64 110	92.67 191	96.92 181

SCA		91.84 181	91.18 183	93.83 206	95.59 197	84.95 263	94.72 258	95.58 249	90.82 139	92.25 150	93.69 264	75.80 258	98.10 204	86.20 224	95.98 142	98.45 121

pmmvs4		90.93 216	89.85 225	94.17 187	93.34 286	90.79 135	94.60 259	96.02 234	84.62 270	87.45 253	95.15 205	81.88 181	97.45 272	87.70 195	87.87 250	94.27 287

HyFIR lowres test		93.66 119	92.92 123	95.87 113	98.24 79	89.88 157	94.58 260	98.49 12	85.06 264	93.78 116	95.78 179	82.86 158	98.67 161	91.77 126	95.71 150	99.07 70

MDA-MVSNet-bldmvs		85.00 286	82.95 289	91.17 279	93.13 292	83.33 280	94.56 261	95.00 274	84.57 271	65.13 324	92.65 282	70.45 288	95.85 305	73.57 310	77.49 311	94.33 283

PMMVS		92.86 147	92.34 143	94.42 179	94.92 233	86.73 239	94.53 262	96.38 221	84.78 269	94.27 106	95.12 208	83.13 152	98.40 180	91.47 135	96.49 137	98.12 139

pmmvs-eth3d		86.22 279	84.45 283	91.53 272	88.34 319	87.25 228	94.47 263	95.01 273	83.47 284	79.51 311	89.61 307	69.75 293	95.71 308	83.13 261	76.73 314	91.64 313

LF4IMVS		87.94 266	87.25 262	89.98 294	92.38 304	80.05 305	94.38 264	95.25 264	87.59 228	84.34 286	94.74 223	64.31 313	97.66 259	84.83 244	87.45 253	92.23 310

thisisatest0515		92.29 166	91.30 176	95.25 144	96.60 152	88.90 193	94.36 265	92.32 312	87.92 219	93.43 125	94.57 230	77.28 249	99.00 132	89.42 165	95.86 146	97.86 151

GA-MVS		91.38 198	90.31 204	94.59 170	94.65 246	87.62 223	94.34 266	96.19 229	90.73 142	90.35 185	93.83 258	71.84 279	97.96 229	87.22 210	93.61 182	98.21 136

IterMVS		90.15 239	89.67 233	91.61 271	95.48 201	83.72 274	94.33 267	96.12 232	89.99 163	87.31 259	94.15 252	75.78 260	96.27 303	86.97 215	86.89 258	94.83 263
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
IterMVS-SCA-FT		90.31 233	89.81 227	91.82 264	95.52 199	84.20 270	94.30 268	96.15 231	90.61 150	87.39 256	94.27 247	75.80 258	96.44 300	87.34 206	86.88 259	94.82 265

test-LLR		91.42 196	91.19 182	92.12 256	94.59 249	80.66 295	94.29 269	92.98 308	91.11 135	90.76 178	92.37 288	79.02 226	98.07 211	88.81 179	96.74 130	97.63 161

TESTMET0.1,1		90.06 240	89.42 237	91.97 259	94.41 256	80.62 297	94.29 269	91.97 316	87.28 236	90.44 183	92.47 287	68.79 295	97.67 257	88.50 184	96.60 135	97.61 165

test-mter		90.19 238	89.54 236	92.12 256	94.59 249	80.66 295	94.29 269	92.98 308	87.68 226	90.76 178	92.37 288	67.67 300	98.07 211	88.81 179	96.74 130	97.63 161

CMPMVS	62.92 21	85.62 284	84.92 280	87.74 302	89.14 317	73.12 322	94.17 272	96.80 203	73.98 318	73.65 318	94.93 212	66.36 305	97.61 262	83.95 256	91.28 214	92.48 308
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
N_pmnet		78.73 296	78.71 296	78.79 310	92.80 296	46.50 335	94.14 273	43.71 338	78.61 311	80.83 302	91.66 301	74.94 265	96.36 301	67.24 321	84.45 284	93.50 298

CostFormer		91.18 210	90.70 195	92.62 249	94.84 238	81.76 290	94.09 274	94.43 291	84.15 275	92.72 141	93.77 262	79.43 219	98.20 193	90.70 145	92.18 199	97.90 148

tpm		90.25 235	89.74 232	91.76 269	93.92 268	79.73 306	93.98 275	93.54 304	88.28 210	91.99 155	93.25 276	77.51 248	97.44 273	87.30 208	87.94 249	98.12 139

TinyColmap		86.82 275	85.35 278	91.21 277	94.91 235	82.99 282	93.94 276	94.02 301	83.58 282	81.56 300	94.68 225	62.34 317	98.13 199	75.78 302	87.35 257	92.52 307

miper_lstm_enhance		90.50 231	90.06 219	91.83 263	95.33 210	83.74 273	93.86 277	96.70 210	87.56 229	87.79 248	93.81 261	83.45 147	96.92 294	87.39 205	84.62 281	94.82 265

USDC		88.94 252	87.83 257	92.27 254	94.66 245	84.96 262	93.86 277	95.90 237	87.34 234	83.40 295	95.56 191	67.43 302	98.19 195	82.64 268	89.67 235	93.66 296

tpm2		89.96 241	89.21 241	92.23 255	94.91 235	81.25 292	93.78 279	94.42 292	80.62 303	91.56 160	93.44 274	76.44 254	97.94 232	85.60 236	92.08 203	97.49 170

ppachtmachnet_test		88.35 263	87.29 261	91.53 272	92.45 302	83.57 279	93.75 280	95.97 235	84.28 273	85.32 281	94.18 251	79.00 230	96.93 293	75.71 303	84.99 277	94.10 289

new-patchmatchnet		83.18 290	81.87 291	87.11 304	86.88 323	75.99 318	93.70 281	95.18 267	85.02 265	77.30 314	88.40 311	65.99 309	93.88 320	74.19 309	70.18 322	91.47 317

MSDG		91.42 196	90.24 209	94.96 157	97.15 129	88.91 192	93.69 282	96.32 223	85.72 256	86.93 266	96.47 146	80.24 205	98.98 134	80.57 281	95.05 161	96.98 177

EPMVS		90.70 225	89.81 227	93.37 229	94.73 243	84.21 269	93.67 283	88.02 325	89.50 174	92.38 145	93.49 272	77.82 246	97.78 249	86.03 230	92.68 190	98.11 142

cascas		91.20 207	90.08 216	94.58 174	94.97 229	89.16 188	93.65 284	97.59 123	79.90 306	89.40 217	92.92 279	75.36 262	98.36 183	92.14 116	94.75 166	96.23 196

UnsupCasMVSNet_eth		85.99 281	84.45 283	90.62 287	89.97 313	82.40 287	93.62 285	97.37 155	89.86 165	78.59 313	92.37 288	65.25 312	95.35 314	82.27 270	70.75 321	94.10 289

our_test_3		88.78 257	87.98 256	91.20 278	92.45 302	82.53 284	93.61 286	95.69 243	85.77 255	84.88 282	93.71 263	79.99 210	96.78 298	79.47 288	86.24 260	94.28 286

PM-MVS		83.48 289	81.86 292	88.31 299	87.83 321	77.59 314	93.43 287	91.75 317	86.91 241	80.63 304	89.91 305	44.42 329	95.84 306	85.17 243	76.73 314	91.50 316

tpmrst		91.44 195	91.32 174	91.79 266	95.15 221	79.20 310	93.42 288	95.37 256	88.55 205	93.49 123	93.67 267	82.49 168	98.27 188	90.41 147	89.34 237	97.90 148

PAPM		91.52 192	90.30 205	95.20 145	95.30 212	89.83 158	93.38 289	96.85 201	86.26 249	88.59 233	95.80 175	84.88 129	98.15 198	75.67 304	95.93 144	97.63 161

testmvs		13.36 309	16.33 311	4.48 322	5.04 338	2.26 340	93.18 290	3.28 340	2.70 334	8.24 336	21.66 333	2.29 342	2.19 337	7.58 334	2.96 334	9.00 333

YYNet1		85.87 282	84.23 285	90.78 286	92.38 304	82.46 286	93.17 291	95.14 269	82.12 292	67.69 320	92.36 291	78.16 241	95.50 313	77.31 298	79.73 308	94.39 281

MDA-MVSNet_test_wron		85.87 282	84.23 285	90.80 285	92.38 304	82.57 283	93.17 291	95.15 268	82.15 291	67.65 321	92.33 294	78.20 238	95.51 312	77.33 297	79.74 307	94.31 285

PatchmatchNet		91.91 179	91.35 172	93.59 218	95.38 205	84.11 271	93.15 293	95.39 254	89.54 172	92.10 153	93.68 266	82.82 160	98.13 199	84.81 245	95.32 155	98.52 110
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
tpmvs		89.83 246	89.15 243	91.89 261	94.92 233	80.30 301	93.11 294	95.46 252	86.28 248	88.08 243	92.65 282	80.44 201	98.52 173	81.47 274	89.92 232	96.84 184

MDTV_nov1_ep13_2view						70.35 324	93.10 295		83.88 279	93.55 120		82.47 169		86.25 223		98.38 129

MDTV_nov1_ep13			90.76 193		95.22 218	80.33 300	93.03 296	95.28 261	88.14 215	92.84 140	93.83 258	81.34 187	98.08 208	82.86 263	94.34 171

PVSNet	86.66 18	92.24 169	91.74 161	93.73 210	97.77 106	83.69 277	92.88 297	96.72 205	87.91 220	93.00 134	94.86 216	78.51 234	99.05 129	86.53 218	97.45 113	98.47 119

dp		88.90 254	88.26 254	90.81 283	94.58 251	76.62 316	92.85 298	94.93 279	85.12 263	90.07 199	93.07 277	75.81 257	98.12 202	80.53 282	87.42 255	97.71 158

test_post1							92.81 299				16.58 336	80.53 199	97.68 256	86.20 224

pmmvs3		79.97 294	77.50 297	87.39 303	82.80 326	79.38 309	92.70 300	90.75 322	70.69 321	78.66 312	87.47 317	51.34 326	93.40 321	73.39 311	69.65 323	89.38 321

tpm cat1		88.36 262	87.21 264	91.81 265	95.13 223	80.55 298	92.58 301	95.70 242	74.97 317	87.45 253	91.96 296	78.01 244	98.17 197	80.39 283	88.74 243	96.72 188

PCF-MVS	89.48 11	91.56 190	89.95 221	96.36 91	96.60 152	92.52 77	92.51 302	97.26 164	79.41 308	88.90 227	96.56 142	84.04 140	99.55 72	77.01 300	97.30 118	97.01 176
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
test123		13.04 310	15.66 312	5.18 321	4.51 339	3.45 339	92.50 303	1.81 341	2.50 335	7.58 337	20.15 334	3.67 341	2.18 338	7.13 335	1.07 335	9.90 332

GG-mvs-BLEND				93.62 216	93.69 276	89.20 185	92.39 304	83.33 331		87.98 247	89.84 306	71.00 285	96.87 295	82.08 271	95.40 154	94.80 268

new_pmnet		82.89 291	81.12 294	88.18 301	89.63 315	80.18 303	91.77 305	92.57 311	76.79 316	75.56 317	88.23 313	61.22 318	94.48 316	71.43 315	82.92 299	89.87 320

MIMVSNet		88.50 261	86.76 267	93.72 212	94.84 238	87.77 221	91.39 306	94.05 299	86.41 247	87.99 246	92.59 284	63.27 314	95.82 307	77.44 296	92.84 188	97.57 168

FPMVS		71.27 298	69.85 299	75.50 312	74.64 329	59.03 331	91.30 307	91.50 319	58.80 325	57.92 327	88.28 312	29.98 333	85.53 329	53.43 326	82.84 300	81.95 324

gg-mvs-nofinetune		87.82 267	85.61 275	94.44 177	94.46 253	89.27 184	91.21 308	84.61 330	80.88 300	89.89 203	74.98 324	71.50 281	97.53 266	85.75 235	97.21 121	96.51 191

ADS-MVSNet2		89.45 248	88.59 249	92.03 258	95.86 188	82.26 288	90.93 309	94.32 296	83.23 286	91.28 172	91.81 298	79.01 228	95.99 304	79.52 286	91.39 212	97.84 152

ADS-MVSNet		89.89 243	88.68 248	93.53 221	95.86 188	84.89 264	90.93 309	95.07 272	83.23 286	91.28 172	91.81 298	79.01 228	97.85 242	79.52 286	91.39 212	97.84 152

UnsupCasMVSNet_bld		82.13 293	79.46 295	90.14 293	88.00 320	82.47 285	90.89 311	96.62 217	78.94 310	75.61 315	84.40 319	56.63 323	96.31 302	77.30 299	66.77 325	91.63 314

PVSNet_0	82.17 19	85.46 285	83.64 287	90.92 281	95.27 213	79.49 307	90.55 312	95.60 247	83.76 281	83.00 296	89.95 304	71.09 284	97.97 225	82.75 266	60.79 326	95.31 242

CHOSEN 280x420		93.12 134	92.72 130	94.34 182	96.71 150	87.27 227	90.29 313	97.72 109	86.61 245	91.34 166	95.29 200	84.29 138	98.41 179	93.25 102	98.94 73	97.35 173

CR-MVSNet		90.82 219	89.77 229	93.95 199	94.45 254	87.19 231	90.23 314	95.68 245	86.89 242	92.40 143	92.36 291	80.91 193	97.05 286	81.09 280	93.95 177	97.60 166

RPMNet		88.52 260	86.72 269	93.95 199	94.45 254	87.19 231	90.23 314	94.99 276	77.87 315	92.40 143	87.55 316	80.17 207	97.05 286	68.84 320	93.95 177	97.60 166

LCM-MVSNet		72.55 297	69.39 300	82.03 308	70.81 333	65.42 329	90.12 316	94.36 295	55.02 326	65.88 323	81.72 320	24.16 337	89.96 325	74.32 308	68.10 324	90.71 319

Patchmtry		88.64 259	87.25 262	92.78 245	94.09 264	86.64 240	89.82 317	95.68 245	80.81 302	87.63 252	92.36 291	80.91 193	97.03 288	78.86 292	85.12 273	94.67 274

PatchT		88.87 255	87.42 260	93.22 235	94.08 265	85.10 260	89.51 318	94.64 288	81.92 293	92.36 146	88.15 314	80.05 209	97.01 291	72.43 312	93.65 180	97.54 169

JIA-IIPM		88.26 264	87.04 266	91.91 260	93.52 280	81.42 291	89.38 319	94.38 293	80.84 301	90.93 177	80.74 321	79.22 222	97.92 236	82.76 265	91.62 207	96.38 195

Patchmatch-test		89.42 249	87.99 255	93.70 213	95.27 213	85.11 259	88.98 320	94.37 294	81.11 298	87.10 262	93.69 264	82.28 172	97.50 268	74.37 307	94.76 165	98.48 118

MVS-HIRNet		82.47 292	81.21 293	86.26 307	95.38 205	69.21 326	88.96 321	89.49 324	66.28 322	80.79 303	74.08 326	68.48 297	97.39 277	71.93 314	95.47 152	92.18 311

Patchmatch-RL test		87.38 270	86.24 270	90.81 283	88.74 318	78.40 313	88.12 322	93.17 307	87.11 239	82.17 298	89.29 308	81.95 179	95.60 310	88.64 182	77.02 312	98.41 126

PMMVS2		70.19 299	66.92 301	80.01 309	76.35 328	65.67 328	86.22 323	87.58 327	64.83 324	62.38 325	80.29 322	26.78 335	88.49 327	63.79 323	54.07 327	85.88 322

ambc				86.56 306	83.60 325	70.00 325	85.69 324	94.97 277		80.60 305	88.45 310	37.42 330	96.84 296	82.69 267	75.44 316	92.86 303

ANet_high		63.94 301	59.58 303	77.02 311	61.24 335	66.06 327	85.66 325	87.93 326	78.53 312	42.94 330	71.04 327	25.42 336	80.71 330	52.60 327	30.83 330	84.28 323

EMVS		52.08 305	51.31 307	54.39 318	72.62 332	45.39 336	83.84 326	75.51 335	41.13 330	40.77 332	59.65 331	30.08 332	73.60 333	28.31 332	29.90 331	44.18 330

E-PMN		53.28 303	52.56 306	55.43 317	74.43 330	47.13 334	83.63 327	76.30 334	42.23 329	42.59 331	62.22 330	28.57 334	74.40 332	31.53 331	31.51 329	44.78 329

PMVS	53.92 22	58.58 302	55.40 304	68.12 315	51.00 336	48.64 333	78.86 328	87.10 329	46.77 328	35.84 334	74.28 325	8.76 338	86.34 328	42.07 329	73.91 319	69.38 326
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
tmp_tt		51.94 306	53.82 305	46.29 319	33.73 337	45.30 337	78.32 329	67.24 337	18.02 332	50.93 329	87.05 318	52.99 325	53.11 335	70.76 318	25.29 332	40.46 331

MVE	50.73 23	53.25 304	48.81 308	66.58 316	65.34 334	57.50 332	72.49 330	70.94 336	40.15 331	39.28 333	63.51 329	6.89 340	73.48 334	38.29 330	42.38 328	68.76 327
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
Gipuma		67.86 300	65.41 302	75.18 313	92.66 299	73.45 321	66.50 331	94.52 290	53.33 327	57.80 328	66.07 328	30.81 331	89.20 326	48.15 328	78.88 310	62.90 328
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
test_normal		79.26 295	75.88 298	89.42 297	84.67 324	76.93 315	58.84 332	97.02 188	89.63 171	59.33 326	75.16 323	46.20 328	97.48 269	87.30 208	84.76 279	93.85 294

wuyk23d		25.11 307	24.57 310	26.74 320	73.98 331	39.89 338	57.88 333	9.80 339	12.27 333	10.39 335	6.97 337	7.03 339	36.44 336	25.43 333	17.39 333	3.89 334

test_part1				0.00 323		0.00 341	0.00 334	98.26 29				0.00 343	0.00 339	0.00 336	0.00 336	0.00 335

cdsmvs_eth3d_5k		23.24 308	30.99 309	0.00 323	0.00 340	0.00 341	0.00 334	97.63 119	0.00 336	0.00 338	96.88 121	84.38 135	0.00 339	0.00 336	0.00 336	0.00 335

pcd_1.5k_mvsjas		7.39 312	9.85 314	0.00 323	0.00 340	0.00 341	0.00 334	0.00 342	0.00 336	0.00 338	0.00 338	88.65 81	0.00 339	0.00 336	0.00 336	0.00 335

sosnet-low-res		0.00 313	0.00 315	0.00 323	0.00 340	0.00 341	0.00 334	0.00 342	0.00 336	0.00 338	0.00 338	0.00 343	0.00 339	0.00 336	0.00 336	0.00 335

sosnet		0.00 313	0.00 315	0.00 323	0.00 340	0.00 341	0.00 334	0.00 342	0.00 336	0.00 338	0.00 338	0.00 343	0.00 339	0.00 336	0.00 336	0.00 335

uncertanet		0.00 313	0.00 315	0.00 323	0.00 340	0.00 341	0.00 334	0.00 342	0.00 336	0.00 338	0.00 338	0.00 343	0.00 339	0.00 336	0.00 336	0.00 335

Regformer		0.00 313	0.00 315	0.00 323	0.00 340	0.00 341	0.00 334	0.00 342	0.00 336	0.00 338	0.00 338	0.00 343	0.00 339	0.00 336	0.00 336	0.00 335

ab-mvs-re		8.06 311	10.74 313	0.00 323	0.00 340	0.00 341	0.00 334	0.00 342	0.00 336	0.00 338	96.69 130	0.00 343	0.00 339	0.00 336	0.00 336	0.00 335

uanet		0.00 313	0.00 315	0.00 323	0.00 340	0.00 341	0.00 334	0.00 342	0.00 336	0.00 338	0.00 338	0.00 343	0.00 339	0.00 336	0.00 336	0.00 335

save filter2										97.90 15	98.30 40	92.94 16	99.81 23	96.61 18	99.61 11	99.44 38

test_0728_THIRD									94.78 27	98.73 4	98.87 4	95.87 1	99.84 16	97.45 4	99.72 2	99.77 1

GSMVS																98.45 121

test_part2					99.28 21	95.74 5				98.10 11

sam_mvs1												82.76 161				98.45 121

sam_mvs												81.94 180

MTGPA								98.08 58

test_post											17.58 335	81.76 182	98.08 208

patchmatchnet-post											90.45 303	82.65 165	98.10 204

gm-plane-assit					93.22 289	78.89 312			84.82 268		93.52 270		98.64 163	87.72 194

test9_res														94.81 72	99.38 40	99.45 36

agg_prior2														93.94 87	99.38 40	99.50 29

agg_prior					98.67 49	93.79 43		98.00 81		95.68 80			99.57 69

TestCases				93.98 195	97.94 97	86.64 240		95.54 250	85.38 258	85.49 278	96.77 124	70.28 289	99.15 115	80.02 284	92.87 186	96.15 201

test_prior				97.23 54	98.67 49	92.99 64		98.00 81					99.41 94			99.29 51

新几何1				97.32 48	98.60 55	93.59 49		97.75 104	81.58 296	95.75 77	97.85 66	90.04 69	99.67 44	86.50 220	99.13 62	98.69 103

旧先验1					98.38 68	93.38 55		97.75 104			98.09 50	92.30 31			99.01 70	99.16 58

原ACMM1				96.38 89	98.59 56	91.09 125		97.89 91	87.41 232	95.22 92	97.68 78	90.25 64	99.54 74	87.95 190	99.12 65	98.49 116

testdata2													99.67 44	85.96 232

segment_acmp												92.89 17

testdata				95.46 140	98.18 88	88.90 193		97.66 115	82.73 289	97.03 37	98.07 51	90.06 68	98.85 145	89.67 159	98.98 71	98.64 105

test12				97.65 36	98.46 61	94.26 27		97.66 115		95.52 89		90.89 57	99.46 88		99.25 52	99.22 55

plane_prior7					96.21 174	89.98 154

plane_prior6					96.10 184	90.00 150						81.32 188

plane_prior5								97.51 130					98.60 167	93.02 106	92.23 196	95.86 209

plane_prior4											96.64 133

plane_prior3						90.00 150			94.46 35	91.34 166

plane_prior1					96.14 182

n2								0.00 342

nn								0.00 342

door-mid								91.06 321

lessismore_v0				90.45 289	91.96 307	79.09 311		87.19 328		80.32 307	94.39 239	66.31 307	97.55 265	84.00 255	76.84 313	94.70 273

LGP-MVS_train				94.10 189	96.16 179	88.26 206		97.46 137	91.29 126	90.12 194	97.16 107	79.05 224	98.73 155	92.25 113	91.89 204	95.31 242

test11								97.88 93

door								91.13 320

HQP5-MVS						89.33 178

BP-MVS														92.13 117

HQP4-MVS										90.14 188			98.50 175			95.78 216

HQP3-MVS								97.39 152							92.10 201

HQP2-MVS												80.95 191

NP-MVS					95.99 187	89.81 159					95.87 170

ACMMP++_ref															90.30 229

ACMMP++															91.02 218

Test By Simon												88.73 80

ITE_SJBPF				92.43 252	95.34 207	85.37 258		95.92 236	91.47 120	87.75 250	96.39 151	71.00 285	97.96 229	82.36 269	89.86 233	93.97 292

DeepMVS_CX				74.68 314	90.84 310	64.34 330		81.61 333	65.34 323	67.47 322	88.01 315	48.60 327	80.13 331	62.33 325	73.68 320	79.58 325

High-resolution multi-view benchmark