High-res multi-view results

This table lists the benchmark results for the high-res multi-view scenario. The following metrics are evaluated:

Accuracy [%]: The fraction of the reconstruction which is closer to the ground truth than the evaluation threshold distance (*). Larger is better.
Completeness [%]: The fraction of the ground truth which is closer to the reconstruction than the evaluation threshold distance (*). Larger is better.
F₁ score [%]: The harmonic mean of accuracy and completeness, used to rank methods based on both metrics. Larger is better.
Time [s]: The runtime of the method. Smaller is better.

(*) For exact definitions, detailing how potentially incomplete ground truth is taken into account, see our paper.

The datasets are grouped into different categories, and result averages are computed for a category and method if results of the method are available for all datasets within the category. Note that the category "all" includes both the high-res multi-view and the low-res many-view scenarios.

Methods with suffix _ROB may participate in the Robust Vision Challenge.

Click a dataset result cell to show a visualization of the reconstruction. For training datasets, ground truth and accuracy / completeness visualizations are also available. The visualizations may not work with mobile browsers.

Method	all	high-res multi-view	indoor	outdoor	courty.	delive.	electro	facade	kicker	meadow	office	pipes	playgr.	relief	relief.	terrace	terrai.

SED-MVS		97.92 1	98.27 2	97.52 1	98.88 12	99.60 1	98.80 5	95.08 7	98.57 2	95.63 2	96.98 9	99.73 1	97.67 2	97.26 10	95.86 22	99.04 15	99.89 5

MSP-MVS		97.74 2	98.32 1	97.06 7	98.66 15	99.35 8	98.66 8	94.75 13	98.22 5	93.60 6	97.99 1	98.58 8	97.41 5	98.24 2	95.95 18	99.27 4	99.91 1
Zhenlong Yuan, Cong Liu, Fei Shen, Zhaoxin Li, Jingguo luo, Tianlu Mao and Zhaoqi Wang: MSP-MVS: Multi-granularity Segmentation Prior Guided Multi-View Stereo. AAAI2025
DVP-MVS++		97.71 3	98.01 6	97.37 2	98.98 6	99.58 3	98.79 6	95.06 8	98.24 4	94.66 3	96.35 15	99.20 4	97.63 3	97.20 12	95.68 23	99.08 13	99.84 7

DPE-MVS		97.69 4	98.16 3	97.14 5	99.01 5	99.52 5	99.12 3	95.38 2	98.00 8	93.31 9	97.71 2	99.61 3	96.94 6	96.99 16	95.45 27	99.09 12	99.81 9
Kehua Chen, Zhenlong Yuan, Tianlu Mao, Zhaoqi Wang: Dual-Level Precision Edges Guided Multi-View Stereo with Accurate Planarization. AAAI2025
DVP-MVS		97.61 5	97.87 7	97.30 3	98.94 11	99.60 1	98.21 13	95.11 4	98.39 3	95.83 1	94.40 29	99.70 2	96.79 7	97.16 13	95.95 18	98.92 26	99.90 2
Zhenlong Yuan, Jinguo Luo, Fei Shen, Zhaoxin Li, Cong Liu, Tianlu Mao, Zhaoqi Wang: DVP-MVS: Synergize Depth-Edge and Visibility Prior for Multi-View Stereo. AAAI2025
CNVR-MVS		97.60 6	98.08 4	97.03 8	99.14 2	99.55 4	98.67 7	95.32 3	97.91 9	92.55 11	97.11 6	97.23 14	97.49 4	98.16 3	97.05 5	99.04 15	99.55 19

APDe-MVS		97.31 7	97.51 12	97.08 6	98.95 10	99.29 13	98.58 10	95.11 4	97.69 14	94.16 4	96.91 10	96.81 18	96.57 10	96.71 20	95.39 29	99.08 13	99.79 10
Zhaojie Zeng, Yuesong Wang, Tao Guan: Matching Ambiguity-Resilient Multi-View Stereo via Adaptive Patch Deformation. Pattern Recognition
SF-MVS		97.17 8	97.18 15	97.17 4	99.11 3	99.20 15	99.05 4	95.55 1	97.39 17	93.56 7	97.48 4	96.71 20	96.75 8	95.73 32	94.40 46	98.98 20	99.33 24

NCCC		97.01 9	97.74 8	96.16 11	99.02 4	99.35 8	98.63 9	95.04 9	97.84 11	88.95 24	96.83 12	97.02 17	96.39 15	97.44 7	96.51 9	98.90 28	99.16 40

SMA-MVS		96.96 10	97.65 11	96.15 12	98.98 6	99.31 12	97.91 18	94.68 15	97.52 15	90.59 18	94.54 28	99.20 4	96.54 12	97.29 9	96.48 10	98.22 64	99.19 36
Yufeng Yin; Xiaoyan Liu; Zichao Zhang: SMA-MVS: Segmentation-Guided Multi-Scale Anchor Deformation Patch Multi-View Stereo. IEEE Transactions on Circuits and Systems for Video Technology
MCST-MVS		96.93 11	98.07 5	95.61 18	98.98 6	99.44 6	98.04 14	95.04 9	98.10 6	86.55 30	97.65 3	97.56 11	95.60 23	97.67 6	96.45 11	99.43 1	99.61 18

HPM-MVS++		96.91 12	97.70 9	96.00 13	98.97 9	99.16 17	97.82 20	94.81 12	98.04 7	89.61 21	96.56 14	98.60 7	96.39 15	97.09 14	95.22 31	98.39 58	99.22 32

SD-MVS		96.87 13	97.69 10	95.92 14	96.38 47	99.25 14	97.76 21	94.75 13	97.72 12	92.46 13	95.94 16	99.09 6	96.48 14	96.01 29	96.08 16	97.68 95	99.73 13
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
APD-MVS		96.79 14	96.99 18	96.56 9	98.76 14	98.87 26	98.42 11	94.93 11	97.70 13	91.83 14	95.52 19	95.94 25	96.63 9	95.94 30	95.47 26	98.80 34	99.47 22
Yuesong Wang, Zhaojie Zeng and etc.: Adaptive Patch Deformation for Textureless-Resilient Multi-View Stereo. CVPR2023
TSAR-MVS + MP.		96.50 15	97.08 16	95.82 16	96.12 51	98.97 23	98.00 15	94.13 20	97.89 10	91.49 15	95.11 24	97.52 12	96.26 19	96.27 27	94.07 56	98.91 27	99.74 12
Zhenlong Yuan, Jiakai Cao, Zhaoqi Wang, Zhaoxin Li: TSAR-MVS: Textureless-aware Segmentation and Correlative Refinement Guided Multi-View Stereo. Pattern Recognition
SteuartSystems-ACMMP		96.20 16	97.22 14	95.01 22	98.40 22	99.11 18	97.93 17	93.62 23	96.28 29	87.45 27	97.05 8	96.00 24	94.23 31	96.83 19	95.97 17	98.40 57	99.27 29
Skip Steuart: Steuart Systems R&D Blog.
HFP-MVS		96.09 17	96.41 23	95.72 17	98.58 17	98.84 27	97.95 16	93.08 27	96.96 22	90.24 19	96.60 13	94.40 31	96.52 13	95.13 42	94.33 47	97.93 85	98.59 65

ACMMP_NAP		95.81 18	96.50 22	95.01 22	98.79 13	99.17 16	97.52 26	94.20 19	96.19 30	85.71 35	93.80 32	96.20 23	95.89 20	96.62 22	94.98 37	97.93 85	98.52 69

train_agg		95.72 19	97.37 13	93.80 28	97.82 31	98.92 24	97.84 19	93.50 24	96.86 24	81.35 54	97.10 7	97.71 9	94.19 32	96.02 28	95.37 30	98.07 72	99.64 16

ACMMPR		95.59 20	95.89 25	95.25 20	98.41 21	98.74 29	97.69 24	92.73 31	96.88 23	88.95 24	95.33 21	92.91 38	95.79 21	94.73 52	94.33 47	97.92 87	98.32 79

DeepC-MVS_fast	91.53 1	95.57 21	95.67 28	95.45 19	98.57 18	99.00 22	97.76 21	94.41 17	97.06 19	86.84 29	86.39 45	92.27 43	96.38 17	97.89 5	98.06 3	98.73 39	99.01 48
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
MSLP-MVS++		95.49 22	94.84 33	96.25 10	98.64 16	98.63 32	98.35 12	92.37 33	95.04 48	92.62 10	87.12 44	93.79 32	96.55 11	93.53 72	96.78 6	98.98 20	98.99 49

CP-MVS		95.43 23	95.67 28	95.14 21	98.24 27	98.60 33	97.45 27	92.80 29	95.98 33	89.21 23	95.22 22	93.60 33	95.43 24	94.37 59	93.22 75	97.68 95	98.72 56

DPM-MVS		95.36 24	95.84 26	94.82 24	96.70 43	98.49 43	99.27 1	95.09 6	96.71 25	83.87 43	86.34 47	96.44 22	95.06 26	98.35 1	98.82 1	98.89 29	95.69 133

MP-MVS		95.24 25	95.96 24	94.40 26	98.32 24	98.38 48	97.12 29	92.87 28	95.17 46	85.50 36	95.68 17	94.91 29	94.58 28	95.11 43	93.76 61	98.05 75	98.68 58
Rongxuan Tan, Qing Wang, et al.: MP-MVS: Multi-Scale Windows PatchMatch and Planar Prior Multi-View Stereo.
TSAR-MVS + ACMM		94.99 26	97.02 17	92.61 38	97.19 37	98.71 31	97.74 23	93.21 26	96.97 21	79.27 72	94.09 30	97.14 15	90.84 67	96.64 21	95.94 20	97.42 112	99.67 15

X-MVS		94.70 27	95.71 27	93.52 32	98.38 23	98.56 35	96.99 30	92.62 32	95.58 37	81.00 62	94.57 27	93.49 34	94.16 34	94.82 48	94.29 50	97.99 81	98.68 58

PGM-MVS		94.64 28	95.49 30	93.66 30	98.55 19	98.51 41	97.63 25	87.77 46	94.45 52	84.92 39	97.23 5	91.90 45	95.22 25	94.56 55	93.80 60	97.87 91	97.97 90

TSAR-MVS + GP.		94.59 29	96.60 21	92.25 39	90.25 94	98.17 55	96.22 36	86.53 52	97.49 16	87.26 28	95.21 23	97.06 16	94.07 36	94.34 61	94.20 52	99.18 5	99.71 14

PHI-MVS		94.49 30	96.72 20	91.88 41	97.06 38	98.88 25	94.99 47	89.13 41	96.15 31	79.70 68	96.91 10	95.78 26	91.87 57	94.65 53	95.68 23	98.53 49	98.98 51

AdaColmap		94.28 31	92.94 45	95.84 15	98.32 24	98.33 50	96.06 38	94.62 16	96.29 28	91.22 16	89.89 38	85.50 74	96.38 17	91.85 101	90.89 90	98.44 53	97.81 93

DeepPCF-MVS	91.00 2	94.15 32	96.87 19	90.97 49	96.82 41	99.33 11	89.40 102	92.76 30	98.76 1	82.36 50	88.74 39	95.49 28	90.58 74	98.13 4	97.80 4	93.88 192	99.88 6

CPTT-MVS		94.11 33	93.99 38	94.25 27	96.58 44	97.66 63	97.31 28	91.94 34	94.84 49	88.72 26	92.51 33	93.04 37	95.78 22	91.51 104	89.97 107	95.15 181	98.37 76

EPNet		93.69 34	95.34 31	91.76 42	96.98 40	98.47 45	95.40 44	86.79 49	95.47 39	82.84 47	95.66 18	89.17 51	90.47 76	95.25 41	94.69 41	98.10 69	98.68 58
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
ACMMP		93.32 35	93.59 41	93.00 36	97.03 39	98.24 51	95.27 45	91.66 37	95.20 44	83.25 45	95.39 20	85.52 72	92.80 48	92.60 91	90.21 103	98.01 78	97.99 88
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
CANet		93.23 36	93.72 40	92.65 37	95.48 54	99.09 20	96.55 34	86.74 50	95.28 42	85.22 37	77.30 75	91.25 47	92.60 50	97.06 15	96.63 7	99.31 2	99.45 23

CDPH-MVS		93.22 37	95.08 32	91.04 48	97.57 34	98.49 43	96.74 32	89.35 40	95.19 45	73.57 101	90.26 36	91.59 46	90.68 71	95.09 45	96.15 14	98.31 63	98.81 54

CSCG		93.16 38	92.65 47	93.76 29	98.32 24	99.09 20	96.12 37	89.91 39	93.15 61	89.64 20	83.62 55	88.91 54	92.40 52	91.09 109	93.70 62	96.14 164	98.99 49

MVS_111021_LR		93.05 39	94.53 35	91.32 46	96.43 46	98.38 48	92.81 62	87.20 48	95.94 35	81.45 53	94.75 25	86.08 68	92.12 55	94.83 47	93.34 69	97.89 90	98.42 75

3Dnovator+	86.26 7	92.90 40	92.45 49	93.42 33	97.25 36	98.45 47	95.82 39	85.71 58	93.83 56	89.55 22	72.31 105	92.28 42	94.01 38	95.10 44	95.92 21	98.17 65	99.23 31

MVS_111021_HR		92.73 41	94.83 34	90.28 55	96.27 48	99.10 19	92.77 63	86.15 55	93.41 59	77.11 90	93.82 31	87.39 60	90.61 72	95.60 34	95.15 33	98.79 35	99.32 25

PLC	89.12 3	92.67 42	90.84 59	94.81 25	97.69 32	96.10 93	95.42 43	91.70 35	95.82 36	92.52 12	81.24 61	86.01 69	94.36 29	92.44 95	90.27 100	97.19 121	93.99 161
Jie Liao, Yanping Fu, Qingan Yan, Chunxia xiao: Pyramid Multi-View Stereo with Local Consistency. Pacific Graphics 2019
3Dnovator	85.78 8	92.53 43	91.96 51	93.20 34	97.99 28	98.47 45	95.78 40	85.94 56	93.07 63	86.40 31	73.43 97	89.00 53	94.08 35	94.74 51	96.44 12	99.01 19	98.57 66

DeepC-MVS	88.77 4	92.39 44	91.74 53	93.14 35	96.21 49	98.55 38	96.30 35	93.84 21	93.06 64	81.09 59	74.69 90	85.20 78	93.48 42	95.41 37	96.13 15	97.92 87	99.18 37
Andreas Kuhn, Christian Sormann, Mattia Rossi, Oliver Erdler, Friedrich Fraundorfer: DeepC-MVS: Deep Confidence Prediction for Multi-View Stereo Reconstruction. 3DV 2020
OMC-MVS		92.05 45	91.88 52	92.25 39	96.51 45	97.94 57	93.18 59	88.97 43	96.53 26	84.47 41	80.79 63	87.85 56	93.25 46	92.48 94	91.81 83	97.12 122	95.73 132

MVSTER		91.91 46	93.43 44	90.14 56	89.81 101	92.32 134	94.53 50	81.32 89	96.00 32	84.77 40	85.41 52	92.39 41	91.32 59	96.41 23	94.01 58	99.11 8	97.45 102

MVS_0304		91.90 47	92.93 46	90.69 53	93.66 64	98.78 28	96.73 33	85.43 62	93.13 62	78.11 84	77.02 78	89.09 52	91.10 63	96.98 17	96.54 8	99.11 8	98.96 52

CS-MVS-test		91.76 48	93.47 42	89.76 59	94.64 59	98.22 53	88.13 112	81.58 86	97.02 20	82.47 49	85.49 51	85.41 76	93.28 44	95.33 39	93.61 64	98.45 52	99.22 32

QAPM		91.68 49	91.97 50	91.34 45	97.86 30	98.72 30	95.60 42	85.72 57	90.86 79	77.14 89	76.06 79	90.35 48	92.69 49	94.10 64	94.60 43	99.04 15	99.09 42

CS-MVS		91.55 50	92.49 48	90.45 54	94.00 62	97.91 59	91.17 83	81.40 88	95.22 43	83.51 44	82.37 59	82.29 84	94.07 36	96.36 26	94.03 57	98.56 46	99.22 32

CNLPA		91.53 51	89.74 72	93.63 31	96.75 42	97.63 65	91.16 84	91.70 35	96.38 27	90.82 17	69.66 116	85.52 72	93.76 39	90.44 115	91.14 89	97.55 105	97.40 103

ETV-MVS		91.51 52	94.06 37	88.54 71	89.39 107	97.52 66	89.48 99	80.88 92	97.09 18	79.41 70	87.87 40	86.18 67	92.95 47	95.94 30	94.33 47	99.13 7	99.52 21

EC-MVSNet		91.25 53	93.45 43	88.68 69	88.90 113	96.18 92	91.66 72	76.70 125	95.57 38	82.00 51	84.18 53	89.28 50	94.17 33	95.64 33	94.19 53	98.68 41	99.14 41

DELS-MVS		91.09 54	90.56 67	91.71 43	95.82 52	98.59 34	95.74 41	86.68 51	85.86 107	85.12 38	72.71 100	81.36 87	88.06 96	97.31 8	98.27 2	98.86 32	99.82 8
Christian Sormann, Emanuele Santellani, Mattia Rossi, Andreas Kuhn, Friedrich Fraundorfer: DELS-MVS: Deep Epipolar Line Search for Multi-View Stereo. Winter Conference on Applications of Computer Vision (WACV), 2023
TAPA-MVS	87.40 6	90.98 55	90.71 61	91.30 47	96.14 50	97.66 63	94.80 48	89.00 42	94.74 51	77.42 88	80.22 64	86.70 63	92.27 53	91.65 103	90.17 105	98.15 68	93.83 165
Andrea Romanoni, Matteo Matteucci: TAPA-MVS: Textureless-Aware PAtchMatch Multi-View Stereo. ICCV 2019
PVSNet_BlendedMVS		90.74 56	90.66 63	90.82 51	94.75 57	98.54 39	91.30 80	86.53 52	95.43 40	85.75 33	78.66 70	70.67 126	87.60 97	96.37 24	95.08 35	98.98 20	99.90 2

PVSNet_Blended		90.74 56	90.66 63	90.82 51	94.75 57	98.54 39	91.30 80	86.53 52	95.43 40	85.75 33	78.66 70	70.67 126	87.60 97	96.37 24	95.08 35	98.98 20	99.90 2

CHOSEN 280x420		90.61 58	94.27 36	86.35 91	93.12 68	98.16 56	89.99 95	69.62 180	92.48 68	76.89 93	87.28 43	96.72 19	90.31 78	94.81 49	92.33 80	98.17 65	98.08 86

MAR-MVS		90.44 59	91.17 57	89.59 60	97.48 35	97.92 58	90.96 87	79.80 97	95.07 47	77.03 91	80.83 62	79.10 97	94.68 27	93.16 77	94.46 45	97.59 104	97.63 95
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
PCF-MVS	88.14 5	90.42 60	89.56 78	91.41 44	94.44 60	98.18 54	94.35 51	94.33 18	84.55 120	76.61 94	75.84 82	88.47 55	91.29 60	90.37 118	90.66 96	97.46 108	98.88 53
Andreas Kuhn, Shan Lin, Oliver Erdler: Plane Completion and Filtering for Multi-View Stereo Reconstruction. GCPR 2019
OpenMVS	83.41 11	89.84 61	88.89 84	90.95 50	97.63 33	98.51 41	94.64 49	85.47 61	88.14 93	78.39 81	65.06 131	85.42 75	91.04 65	93.06 80	93.70 62	98.53 49	98.37 76

EIA-MVS		89.82 62	91.48 55	87.89 80	89.16 109	97.31 68	88.99 103	80.92 91	94.29 53	77.65 86	82.16 60	79.77 95	91.90 56	94.61 54	93.03 77	98.70 40	99.21 35

sasdasda		89.62 63	89.87 70	89.33 62	90.47 87	97.02 74	93.46 56	79.67 100	92.45 69	81.05 60	82.84 56	73.00 113	93.71 40	90.38 116	94.85 38	97.65 99	98.54 67

canonicalmvs		89.62 63	89.87 70	89.33 62	90.47 87	97.02 74	93.46 56	79.67 100	92.45 69	81.05 60	82.84 56	73.00 113	93.71 40	90.38 116	94.85 38	97.65 99	98.54 67

TSAR-MVS + COLMAP		89.59 65	89.64 75	89.53 61	93.32 67	96.51 84	95.03 46	88.53 44	95.98 33	69.10 117	91.81 34	64.53 149	93.40 43	93.53 72	91.35 88	97.77 92	93.75 168

HQP-MVS		89.57 66	90.57 66	88.41 73	92.77 69	94.71 109	94.24 52	87.97 45	93.44 58	68.18 120	91.75 35	71.54 125	89.90 81	92.31 98	91.43 86	97.39 113	98.80 55

MGCFI-Net		89.36 67	89.66 74	89.02 66	90.40 91	96.92 77	93.26 58	79.54 104	92.10 71	80.11 66	82.55 58	72.65 116	93.26 45	90.24 120	94.69 41	97.53 106	98.46 73

MVS_Test		89.02 68	90.20 68	87.64 82	89.83 100	97.05 73	92.30 66	77.59 121	92.89 65	75.01 99	77.36 74	76.10 107	92.27 53	95.30 40	95.42 28	98.83 33	97.30 107

CLD-MVS		88.99 69	88.07 87	90.07 57	89.61 103	94.94 106	93.82 55	85.70 59	92.73 67	82.73 48	79.97 65	69.59 130	90.44 77	90.32 119	89.93 109	98.10 69	99.04 45
Zhaoxin Li, Wangmeng Zuo, Zhaoqi Wang, Lei Zhang: Confidence-based Large-scale Dense Multi-view Stereo. IEEE Transaction on Image Processing, 2020
baseline		88.91 70	89.94 69	87.70 81	89.44 106	96.74 82	91.62 74	77.92 118	93.79 57	78.76 76	77.55 73	78.46 100	89.38 87	92.26 99	92.52 79	99.10 10	98.23 80

PMMVS		88.56 71	91.22 56	85.47 99	90.04 96	95.60 102	86.62 127	78.49 113	93.86 55	70.62 112	90.00 37	80.08 93	91.64 58	92.36 96	89.80 113	95.40 176	96.84 116

test2506		88.38 72	88.02 89	88.80 68	91.55 78	97.78 60	90.87 89	83.36 71	84.51 121	83.06 46	74.13 93	76.93 104	85.39 107	94.34 61	93.33 71	98.60 42	95.10 150

baseline1		88.16 73	88.15 86	88.17 77	90.02 97	94.79 108	91.85 71	83.89 65	87.37 99	75.67 97	73.75 95	79.89 94	88.44 95	94.41 56	93.33 71	99.18 5	93.55 170

thisisatest0530		87.99 74	90.76 60	84.75 103	88.36 118	96.82 79	87.65 117	79.67 100	91.77 73	70.93 108	79.94 66	87.65 58	84.21 117	92.98 83	89.07 125	97.66 98	97.13 110

tttt0517		87.93 75	90.71 61	84.68 104	88.33 119	96.76 81	87.42 120	79.67 100	91.74 74	70.83 109	79.91 67	87.61 59	84.21 117	92.88 88	89.07 125	97.62 102	97.03 112

CANet_DTU		87.91 76	91.57 54	83.64 111	90.96 81	97.12 71	91.90 70	75.97 133	92.83 66	53.16 175	86.02 48	79.02 98	90.80 68	95.40 38	94.15 54	99.03 18	96.47 127

diffmvs		87.86 77	87.40 95	88.39 74	88.57 116	96.10 93	91.24 82	83.15 74	90.62 80	79.13 74	72.45 103	67.71 136	90.07 80	92.58 92	93.31 74	98.17 65	99.03 46
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
IS_MVSNet		87.83 78	90.66 63	84.53 105	90.08 95	96.79 80	88.16 111	79.89 96	85.44 109	72.20 103	75.50 86	87.14 61	80.21 145	95.53 35	95.22 31	96.65 138	99.02 47

EPP-MVSNet		87.72 79	89.74 72	85.37 100	89.11 110	95.57 103	86.31 128	79.44 105	85.83 108	75.73 96	77.23 76	90.05 49	84.78 113	91.22 107	90.25 101	96.83 129	98.04 87

casdiffmvs_mvg		87.64 80	86.46 102	89.01 67	89.45 105	96.09 95	92.69 64	83.42 70	84.60 119	80.01 67	68.55 119	70.29 128	90.51 75	93.93 67	93.59 65	97.96 82	98.18 81
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
ET-MVSNet_ETH3D		87.63 81	91.08 58	83.59 112	67.96 216	96.30 91	92.06 68	78.47 114	91.95 72	69.87 114	87.57 42	84.14 82	94.34 30	88.58 133	92.10 81	98.88 30	96.93 113

DI_MVS_plusplus_trai		87.63 81	87.13 97	88.22 76	88.61 115	95.92 98	94.09 54	81.41 87	87.00 102	78.38 82	59.70 150	80.52 91	89.08 90	94.37 59	93.34 69	97.73 93	99.05 44

casdiffmvs		87.59 83	86.69 101	88.64 70	89.06 111	96.32 90	90.18 92	83.21 73	87.74 97	80.20 65	67.99 121	68.34 134	90.79 69	93.83 68	94.08 55	98.41 56	98.50 71
Fangjinhua Wang, Qingshan Xu, Yew-Soon Ong, Marc Pollefeys: Lightweight and Accurate Multi-View Stereo With Confidence-Aware Diffusion Model. IEEE T-PAMI 2025
PVSNet_Blended_VisFu		87.44 84	88.72 85	85.95 95	92.02 73	97.26 69	86.88 125	82.66 81	83.86 127	79.16 73	66.96 125	84.91 79	77.26 162	94.97 46	93.48 66	97.73 93	99.64 16

FMVSNet3		87.19 85	87.32 96	87.04 89	82.82 154	90.21 149	92.88 61	76.53 128	91.69 75	81.31 55	64.81 134	80.64 88	89.79 85	94.80 50	94.76 40	98.88 30	94.32 157

LS3D		87.19 85	85.48 109	89.18 64	94.96 56	95.47 104	92.02 69	93.36 25	88.69 91	67.01 121	70.56 112	72.10 120	92.47 51	89.96 123	89.93 109	95.25 178	91.68 179

ACMP	85.16 9	87.15 87	87.04 98	87.27 86	90.80 83	94.45 112	89.41 101	83.09 78	89.15 87	76.98 92	86.35 46	65.80 143	86.94 100	88.45 134	87.52 144	96.42 153	97.56 100
Qingshan Xu and Wenbing Tao: Planar Prior Assisted PatchMatch Multi-View Stereo. AAAI 2020
UGNet		87.04 88	89.59 77	84.07 107	90.94 82	95.95 97	86.02 130	81.65 85	85.94 106	78.54 80	78.00 72	85.40 77	69.62 182	91.83 102	91.53 85	97.63 101	98.51 70
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
LGP-MVS_train		86.95 89	87.65 92	86.12 94	91.77 76	93.84 118	93.04 60	82.77 80	88.04 94	65.33 126	87.69 41	67.09 140	86.79 101	90.20 121	88.99 128	97.05 124	97.71 94

PatchMatch-RL		86.75 90	85.43 110	88.29 75	94.06 61	96.37 89	86.82 126	82.94 79	88.94 89	79.59 69	79.83 68	59.17 162	89.46 86	91.12 108	88.81 132	96.88 128	93.78 166

FA-MVS(training)		86.74 91	88.01 90	85.26 101	89.86 98	96.99 76	88.54 108	64.26 196	89.04 88	81.30 58	66.74 127	81.52 86	89.11 89	94.04 65	90.37 99	98.47 51	97.37 104

baseline2		86.51 92	89.35 81	83.19 114	85.70 139	94.88 107	85.75 135	77.13 123	89.87 84	70.65 111	79.03 69	79.14 96	81.51 138	93.70 69	90.22 102	98.38 59	98.60 64

thres100view900		86.48 93	85.08 112	88.12 78	90.54 84	96.90 78	92.39 65	84.82 63	84.16 125	71.65 104	70.86 109	60.49 157	91.23 62	93.65 70	90.19 104	98.10 69	99.32 25

ACMM	84.23 10	86.40 94	84.64 115	88.46 72	91.90 74	91.93 140	88.11 113	85.59 60	88.61 92	79.13 74	75.31 87	66.25 141	89.86 84	89.88 124	87.64 141	96.16 163	92.86 175
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
GBi-Net		86.16 95	86.00 105	86.35 91	81.81 160	89.52 158	91.40 76	76.53 128	91.69 75	81.31 55	64.81 134	80.64 88	88.72 91	90.54 112	90.72 92	98.34 60	94.08 158

test1		86.16 95	86.00 105	86.35 91	81.81 160	89.52 158	91.40 76	76.53 128	91.69 75	81.31 55	64.81 134	80.64 88	88.72 91	90.54 112	90.72 92	98.34 60	94.08 158

tfpn200view9		86.07 97	84.76 114	87.61 83	90.54 84	96.39 86	91.35 79	83.15 74	84.16 125	71.65 104	70.86 109	60.49 157	90.91 66	92.89 85	89.34 116	98.05 75	99.17 38

DCV-MVSNet		85.90 98	85.88 107	85.93 96	87.86 124	88.37 175	89.45 100	77.46 122	87.33 100	77.51 87	76.06 79	75.76 109	88.48 94	87.40 143	88.89 131	94.80 187	97.37 104

Vis-MVSNet (Re-imp)		85.89 99	89.62 76	81.55 125	89.85 99	96.08 96	87.55 118	79.80 97	84.80 116	66.55 123	73.70 96	86.71 62	68.25 189	94.40 57	94.53 44	97.32 116	97.09 111

MSDG		85.81 100	82.29 139	89.93 58	95.52 53	92.61 129	91.51 75	91.46 38	85.12 113	78.56 78	63.25 140	69.01 132	85.31 110	88.45 134	88.23 135	97.21 120	89.33 190

thres200		85.80 101	84.38 116	87.46 84	90.51 86	96.39 86	91.64 73	83.15 74	81.59 136	71.54 106	70.24 113	60.41 159	89.88 82	92.89 85	89.85 112	98.06 73	99.26 30

ECVR-MVS		85.74 102	83.80 124	88.00 79	91.55 78	97.78 60	90.87 89	83.36 71	84.51 121	78.21 83	58.65 155	62.75 153	85.39 107	94.34 61	93.33 71	98.60 42	95.25 144

OPM-MVS		85.69 103	82.79 132	89.06 65	93.42 65	94.21 116	94.21 53	87.61 47	72.68 161	70.79 110	71.09 107	67.27 139	90.74 70	91.29 106	89.05 127	97.61 103	93.94 163
Ray L. Khuboni, Hongjun Xu: Octagram Propagation Matching for Multi-Scale View Stereopsis (OPM-MVS).
thres400		85.59 104	84.08 119	87.36 85	90.45 89	96.60 83	90.95 88	83.67 68	80.99 139	71.17 107	69.08 118	60.25 160	89.88 82	93.14 78	89.34 116	98.02 77	99.17 38

CostFormer		85.47 105	86.98 99	83.71 110	88.70 114	94.02 117	88.07 114	62.72 198	89.78 85	78.68 77	72.69 101	78.37 101	87.35 99	85.96 156	89.32 120	96.73 135	98.72 56

test1111		85.17 106	83.46 127	87.17 87	91.36 80	97.75 62	90.06 94	83.44 69	83.41 129	75.25 98	58.08 158	62.19 155	84.39 116	94.39 58	93.38 68	98.54 48	95.00 152

thres600view7		85.14 107	83.58 126	86.96 90	90.37 93	96.39 86	90.33 91	83.15 74	80.46 140	70.60 113	67.96 122	60.04 161	89.22 88	92.89 85	88.28 134	98.06 73	99.08 43

test-LLR		85.11 108	89.49 79	80.00 134	85.32 143	94.49 110	82.27 165	74.18 142	87.83 95	56.70 153	75.55 84	86.26 64	82.75 131	93.06 80	90.60 97	98.77 36	98.65 62

FMVSNet2		84.89 109	84.02 121	85.91 97	81.81 160	89.52 158	91.40 76	75.79 134	84.45 123	79.39 71	58.75 153	74.35 111	88.72 91	93.51 74	93.46 67	98.34 60	94.08 158

FC-MVSNet-train		84.88 110	84.08 119	85.82 98	89.21 108	91.74 141	85.87 131	81.20 90	81.71 135	74.66 100	73.38 98	64.99 147	86.60 102	90.75 110	88.08 136	97.36 114	97.90 91

EPNet_dtu		84.87 111	89.01 82	80.05 133	95.25 55	92.88 127	88.84 105	84.11 64	91.69 75	49.28 191	85.69 49	78.95 99	65.39 194	92.22 100	91.66 84	97.43 111	89.95 186
Wanjuan Su, Wenbing Tao: Efficient Edge-Preserving Multi-View Stereo Network for Depth Estimation. AAAI 2023
Effi-MVS+		84.80 112	85.71 108	83.73 109	87.94 123	95.76 99	90.08 93	73.45 149	85.12 113	62.66 135	72.39 104	64.97 148	90.59 73	92.95 84	90.69 95	97.67 97	98.12 83

UA-Net		84.69 113	87.64 93	81.25 127	90.38 92	95.67 100	87.33 121	79.41 106	72.07 165	66.48 124	75.09 88	92.48 40	66.88 190	94.03 66	94.25 51	97.01 127	89.88 187

TESTMET0.1,1		84.62 114	89.49 79	78.94 143	82.18 157	94.49 110	82.27 165	70.94 169	87.83 95	56.70 153	75.55 84	86.26 64	82.75 131	93.06 80	90.60 97	98.77 36	98.65 62

CHOSEN 1792x2688		84.59 115	84.30 118	84.93 102	93.71 63	98.23 52	89.91 96	77.96 117	84.81 115	65.93 125	45.19 200	71.76 124	83.13 129	95.46 36	95.13 34	98.94 25	99.53 20

Anonymous20231211		84.23 116	81.71 144	87.17 87	87.38 131	93.59 121	88.95 104	82.14 83	83.82 128	78.56 78	48.09 194	73.89 112	91.25 61	86.38 150	88.06 138	94.74 188	98.14 82

MDTV_nov1_ep13		84.17 117	88.03 88	79.66 136	86.00 137	94.41 113	85.05 137	66.01 192	90.36 81	64.34 131	77.13 77	84.56 80	82.71 133	87.12 147	88.92 129	93.84 194	93.69 169

test-mter		84.06 118	89.00 83	78.29 148	81.92 158	94.23 115	81.07 175	70.38 174	87.12 101	56.10 162	74.75 89	85.80 70	81.81 137	92.52 93	90.10 106	98.43 54	98.49 72

IB-MVS	79.58 12	83.83 119	84.81 113	82.68 117	91.85 75	97.35 67	75.75 194	82.57 82	86.55 104	84.01 42	70.90 108	65.43 145	63.18 200	84.19 170	89.92 111	98.74 38	99.31 27
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
EPMVS		83.71 120	86.76 100	80.16 132	89.72 102	95.64 101	84.68 138	59.73 203	89.61 86	62.67 134	72.65 102	81.80 85	86.22 104	86.23 152	88.03 139	97.96 82	93.35 171

HyFIR lowres test		83.43 121	82.94 130	84.01 108	93.41 66	97.10 72	87.21 122	74.04 144	80.15 142	64.98 127	41.09 208	76.61 106	86.51 103	93.31 75	93.01 78	97.91 89	99.30 28

PatchmatchNet		83.28 122	87.57 94	78.29 148	87.46 129	94.95 105	83.36 147	59.43 206	90.20 83	58.10 148	74.29 92	86.20 66	84.13 119	85.27 162	87.39 145	97.25 119	94.67 155
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Pablo Speciale, Marc Pollefeys: PatchmatchNet: Learned Multi-View Patchmatch Stereo.
SCA		83.26 123	87.76 91	78.00 153	87.45 130	92.20 135	82.63 161	58.42 208	90.30 82	58.23 146	75.74 83	87.75 57	83.97 122	86.10 155	87.64 141	97.30 117	94.62 156

GeoE		83.17 124	82.86 131	83.53 113	87.24 132	93.78 119	87.94 115	72.75 154	82.19 133	69.76 115	60.54 147	65.95 142	86.01 105	89.41 128	89.72 114	97.47 107	98.43 74

CDS-MVSNet		83.13 125	83.73 125	82.43 123	84.52 148	92.92 126	88.26 110	77.67 120	72.08 164	69.08 118	66.96 125	74.66 110	78.61 151	90.70 111	91.96 82	96.46 152	96.86 115
Khang Truong Giang, Soohwan Song, Sungho Jo: Curvature-guided dynamic scale networks for Multi-view Stereo. ICLR 2022
RPSCF		82.91 126	81.86 141	84.13 106	88.25 120	88.32 176	87.67 116	80.86 93	84.78 117	76.57 95	85.56 50	76.00 108	84.61 114	78.20 205	76.52 208	86.81 214	83.63 207

Vis-MVSNet		82.88 127	86.04 104	79.20 141	87.77 127	96.42 85	86.10 129	76.70 125	74.82 155	61.38 138	70.70 111	77.91 102	64.83 196	93.22 76	93.19 76	98.43 54	96.01 130
Jingyang Zhang, Yao Yao, Shiwei Li, Zixin Luo, Tian Fang: Visibility-aware Multiview Stereo Network. BMVC 2020
dps		82.63 128	82.64 135	82.62 119	87.81 126	92.81 128	84.39 139	61.96 199	86.43 105	81.63 52	69.72 115	67.60 138	84.42 115	82.51 183	83.90 182	95.52 172	95.50 141

IterMVS-LS		82.62 129	82.75 134	82.48 120	87.09 133	87.48 189	87.19 123	72.85 152	79.09 143	66.63 122	65.22 129	72.14 119	84.06 121	88.33 137	91.39 87	97.03 126	95.60 140
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
Fast-Effi-MVS+		82.61 130	82.51 137	82.72 116	85.49 142	93.06 125	87.17 124	71.39 166	84.18 124	64.59 129	63.03 141	58.89 163	90.22 79	91.39 105	90.83 91	97.44 109	96.21 129

tpm cat1		82.39 131	82.32 138	82.47 121	88.13 121	92.42 133	87.43 119	62.79 197	85.30 110	78.05 85	60.14 148	72.10 120	83.20 128	82.26 186	85.67 161	95.23 179	98.35 78

dmvs_re		82.31 132	81.55 145	83.19 114	83.15 153	93.17 124	88.68 107	83.72 66	82.73 131	61.70 136	67.43 124	55.43 171	83.35 127	87.51 142	89.27 123	98.56 46	95.31 143

MS-PatchMatch		82.16 133	82.18 140	82.12 124	91.65 77	93.50 122	89.51 98	71.95 160	81.48 137	64.45 130	59.58 152	77.54 103	77.23 163	89.88 124	85.62 162	97.94 84	87.68 194

tpmrst		81.71 134	83.87 123	79.20 141	89.01 112	93.67 120	84.22 140	60.14 201	87.45 98	59.49 142	64.97 132	71.86 123	85.30 111	84.72 166	86.30 153	97.04 125	98.09 85

RPMNet		81.47 135	86.24 103	75.90 171	86.72 134	92.12 137	82.82 159	55.76 214	85.21 111	53.73 173	63.45 138	83.16 83	80.13 146	92.34 97	89.52 115	96.23 161	97.90 91

CR-MVSNet		81.44 136	85.29 111	76.94 162	86.53 135	92.12 137	83.86 141	58.37 209	85.21 111	56.28 157	59.60 151	80.39 92	80.50 143	92.77 89	89.32 120	96.12 165	97.59 98

Effi-MVS+-dtu		81.18 137	82.77 133	79.33 139	84.70 147	92.54 131	85.81 132	71.55 164	78.84 144	57.06 152	71.98 106	63.77 151	85.09 112	88.94 130	87.62 143	91.79 207	95.68 135

test0.0.03 1		80.99 138	84.37 117	77.05 160	85.32 143	89.79 154	78.43 185	74.18 142	84.78 117	57.98 151	76.06 79	72.88 115	69.14 186	88.02 139	87.70 140	97.27 118	91.37 180

Fast-Effi-MVS+-dtu		80.57 139	83.44 128	77.22 158	83.98 151	91.52 143	85.78 134	64.54 195	80.38 141	50.28 187	74.06 94	62.89 152	82.00 136	89.10 129	88.91 130	96.75 133	97.21 109

FMVSNet5		80.56 140	82.53 136	78.26 150	73.80 210	81.52 208	82.26 167	68.36 185	88.85 90	64.21 132	69.09 117	84.38 81	83.49 126	87.13 146	86.76 150	97.44 109	79.95 210

ADS-MVSNet		80.25 141	82.96 129	77.08 159	87.86 124	92.60 130	81.82 172	56.19 213	86.95 103	56.16 160	68.19 120	72.42 118	83.70 125	82.05 187	85.45 167	96.75 133	93.08 174

FMVSNet1		80.18 142	78.07 156	82.65 118	78.55 184	87.57 188	88.41 109	73.93 145	70.16 170	73.57 101	49.80 183	64.45 150	85.35 109	90.54 112	90.72 92	96.10 166	93.21 172

USDC		80.10 143	79.33 152	81.00 129	86.36 136	91.71 142	88.74 106	75.77 135	81.90 134	54.90 167	67.67 123	52.05 176	83.94 123	88.44 136	86.25 154	96.31 156	87.28 198

COLMAP_ROB	75.69 15	79.47 144	76.90 163	82.46 122	92.20 70	90.53 145	85.30 136	83.69 67	78.27 147	61.47 137	58.26 156	62.75 153	78.28 154	82.41 184	82.13 195	93.83 196	83.98 206
Johannes L. SchÃ¶nberger, Enliang Zheng, Marc Pollefeys, Jan-Michael Frahm: Pixelwise View Selection for Unstructured Multi-View Stereo. ECCV 2016
pmmvs4		79.32 145	77.78 158	81.11 128	80.18 169	88.96 170	83.39 145	76.07 131	81.27 138	69.35 116	58.66 154	51.19 179	82.01 135	87.16 145	84.39 179	95.66 170	92.82 176

PatchT		79.28 146	83.88 122	73.93 180	85.54 141	90.95 144	66.14 211	56.53 212	83.21 130	56.28 157	56.50 160	76.80 105	80.50 143	92.77 89	89.32 120	98.57 45	97.59 98

ACMH	78.51 14	79.27 147	78.08 155	80.65 130	89.52 104	90.40 146	80.45 177	79.77 99	69.54 175	54.85 168	64.83 133	56.16 169	83.94 123	84.58 168	86.01 158	95.41 175	95.03 151
Qingshan Xu and Wenbing Tao: Multi-Scale Geometric Consistency Guided Multi-View Stereo. CVPR 2019
TAMVS		79.23 148	78.95 154	79.56 137	81.89 159	92.52 132	82.97 154	73.70 146	67.27 181	64.97 128	61.66 146	65.06 146	78.61 151	87.12 147	88.07 137	95.23 179	90.95 182

ACMH+	79.09 13	79.12 149	77.22 162	81.35 126	88.50 117	90.36 147	82.14 169	79.38 108	72.78 160	58.59 143	62.31 145	56.44 168	84.10 120	82.03 188	84.05 180	95.40 176	92.55 177

UniMVSNet_NR-MVSNet		78.89 150	78.04 157	79.88 135	79.40 175	89.70 155	82.92 156	80.17 94	76.37 153	58.56 144	57.10 159	54.92 172	81.44 139	83.51 175	87.12 147	96.76 132	97.60 96

tpm		78.87 151	81.33 148	76.00 169	85.57 140	90.19 150	82.81 160	59.66 204	78.35 146	51.40 182	66.30 128	67.92 135	80.94 141	83.28 178	85.73 159	95.65 171	97.56 100

GA-MVS		78.86 152	80.42 149	77.05 160	83.27 152	92.17 136	83.24 149	75.73 136	73.75 157	46.27 201	62.43 143	57.12 165	76.94 165	93.14 78	89.34 116	96.83 129	95.00 152

IterMVS		78.85 153	81.36 146	75.93 170	84.27 150	85.74 195	83.83 143	66.35 190	76.82 148	50.48 185	63.48 137	68.82 133	73.99 170	89.68 126	89.34 116	96.63 141	95.67 136
Fangjinhua Wang, Silvano Galliani, Christoph Vogel, Marc Pollefeys: IterMVS: Iterative Probability Estimation for Efficient Multi-View Stereo.
IterMVS-SCA-FT		78.71 154	81.34 147	75.64 175	84.31 149	85.67 196	83.51 144	66.14 191	76.67 149	50.38 186	63.45 138	69.02 131	73.23 172	89.66 127	89.22 124	96.24 160	95.67 136

UniMVSNet (Re)		78.00 155	77.52 159	78.57 146	79.66 174	90.36 147	82.09 170	77.86 119	76.38 152	60.26 139	54.63 166	52.07 175	75.31 168	84.97 165	86.10 156	96.22 162	98.11 84

DU-MVS		77.98 156	76.71 164	79.46 138	78.68 181	89.26 164	82.92 156	79.06 110	76.52 150	58.56 144	54.89 164	48.35 193	81.44 139	83.16 180	87.21 146	96.08 167	97.60 96

FC-MVSNet-test		77.95 157	81.85 142	73.39 185	82.31 155	88.99 169	79.33 181	74.24 141	78.75 145	47.40 199	70.22 114	72.09 122	60.78 206	86.66 149	85.62 162	96.30 157	90.61 183

NR-MVSNet		77.21 158	76.41 165	78.14 152	80.18 169	89.26 164	83.38 146	79.06 110	76.52 150	56.59 155	54.89 164	45.32 203	72.89 174	85.39 161	86.12 155	96.71 136	97.36 106

thisisatest0515		77.13 159	79.36 151	74.52 177	79.79 173	89.65 156	73.54 199	73.69 147	74.10 156	58.14 147	62.79 142	60.57 156	66.49 192	88.08 138	85.16 172	95.49 174	95.15 148

gg-mvs-nofinetune		77.08 160	79.79 150	73.92 181	85.95 138	97.23 70	92.18 67	52.65 217	46.19 220	27.79 224	38.27 212	85.63 71	85.67 106	96.95 18	95.62 25	99.30 3	98.67 61

TranMVSNet+NR-MVSNet		77.02 161	75.76 167	78.49 147	78.46 187	88.24 177	83.03 153	79.97 95	73.49 159	54.73 169	54.00 169	48.74 188	78.15 156	82.36 185	86.90 149	96.59 143	96.55 121

CVMVSNet		76.86 162	79.09 153	74.26 178	85.29 145	89.44 161	79.91 180	78.47 114	68.94 178	44.45 206	62.35 144	69.70 129	64.50 197	85.82 157	87.03 148	92.94 202	90.33 184

Baseline_NR-MVSNet		76.71 163	74.56 174	79.23 140	78.68 181	84.15 204	82.45 163	78.87 112	75.83 154	60.05 140	47.92 195	50.18 185	79.06 150	83.16 180	83.86 183	96.26 158	96.80 117

v2v482		76.25 164	74.78 171	77.96 154	78.50 186	89.14 167	83.05 152	76.02 132	68.78 179	54.11 170	51.36 175	48.59 190	79.49 148	83.53 174	85.60 165	96.59 143	96.49 126

V42		76.21 165	75.04 170	77.58 155	78.68 181	89.33 163	82.93 155	74.64 139	69.84 172	56.13 161	50.42 180	50.93 180	76.30 167	83.32 176	84.89 176	96.83 129	96.54 122

v8		75.89 166	74.74 172	77.23 157	79.09 177	88.00 180	83.19 150	71.08 168	70.03 171	56.29 156	50.50 178	50.88 181	77.06 164	83.32 176	84.99 174	96.68 137	95.49 142

TinyColmap		75.75 167	73.19 185	78.74 145	84.82 146	87.69 184	81.59 173	74.62 140	71.81 166	54.01 171	55.79 163	44.42 208	82.89 130	84.61 167	83.76 184	94.50 189	84.22 205

MIMVSNet		75.71 168	77.26 160	73.90 182	70.93 212	88.71 173	79.98 179	57.67 211	73.58 158	58.08 150	53.93 170	58.56 164	79.41 149	90.04 122	89.97 107	97.34 115	86.04 199

UniMVSNet_ETH3D		75.63 169	71.59 194	80.35 131	81.03 164	89.90 153	83.25 148	76.58 127	60.08 202	64.19 133	42.89 207	45.01 204	82.14 134	80.20 198	86.75 151	94.90 184	96.29 128

pm-mvs1		75.61 170	74.19 176	77.26 156	80.16 171	88.79 171	81.49 174	75.49 138	59.49 204	58.09 149	48.32 191	55.53 170	72.35 175	88.61 132	85.48 166	95.99 168	93.12 173

v10		75.57 171	74.67 173	76.62 165	78.73 180	87.46 190	83.14 151	69.41 181	69.27 176	53.44 174	49.73 184	49.21 187	78.44 153	86.17 154	85.18 171	96.53 148	95.65 139

v1144		75.54 172	74.55 175	76.69 163	78.33 190	88.77 172	82.89 158	72.76 153	67.18 183	51.73 179	49.34 186	48.37 191	78.10 157	86.22 153	85.24 169	96.35 155	96.74 118

TDRefinement		75.54 172	73.22 183	78.25 151	87.65 128	89.65 156	85.81 132	79.28 109	71.14 168	56.06 163	52.17 173	51.96 177	68.74 188	81.60 189	80.58 197	91.94 205	85.45 200

pmmvs5		75.46 174	75.12 169	75.87 172	79.39 176	89.44 161	78.12 187	72.27 158	65.98 188	51.54 180	55.83 162	46.23 198	76.80 166	88.77 131	85.73 159	97.07 123	93.84 164

tfpnnormal		75.27 175	72.12 191	78.94 143	82.30 156	88.52 174	82.41 164	79.41 106	58.03 205	55.59 165	43.83 206	44.71 205	77.35 160	87.70 141	85.45 167	96.60 142	96.61 120

anonymousdsp		75.14 176	77.25 161	72.69 188	76.68 200	89.26 164	75.26 196	68.44 184	65.53 191	46.65 200	58.16 157	56.67 167	73.96 171	87.84 140	86.05 157	95.13 182	97.22 108

v148		74.98 177	73.52 181	76.69 163	78.84 179	89.02 168	78.78 183	76.82 124	67.22 182	59.61 141	49.18 187	47.94 195	70.57 181	80.76 193	83.99 181	95.52 172	96.52 124

v1192		74.96 178	73.92 177	76.17 166	77.76 193	88.19 179	82.54 162	71.94 161	66.84 184	50.07 189	48.10 193	46.14 199	78.28 154	86.30 151	85.23 170	96.41 154	96.67 119

v144192		74.76 179	73.64 178	76.06 168	77.58 194	88.23 178	81.87 171	71.63 163	66.03 187	51.08 183	48.63 190	46.77 197	77.59 159	84.53 169	84.76 177	96.64 140	96.54 122

v1921920		74.60 180	73.56 180	75.81 173	77.43 196	87.94 181	82.18 168	71.33 167	66.48 186	49.23 193	47.84 196	45.56 201	78.03 158	85.70 159	84.92 175	96.65 138	96.50 125

v1240		74.04 181	73.04 187	75.20 176	77.19 198	87.69 184	80.93 176	70.72 173	65.08 192	48.47 194	47.31 197	44.71 205	77.33 161	85.50 160	85.07 173	96.59 143	95.94 131

testgi		73.22 182	75.84 166	70.16 199	81.67 163	85.50 199	71.45 201	70.81 171	69.56 174	44.74 205	74.52 91	49.25 186	58.45 207	84.10 172	83.37 188	93.86 193	84.56 204

CP-MVSNet		73.19 183	72.37 189	74.15 179	77.54 195	86.77 193	76.34 190	72.05 159	65.66 190	51.47 181	50.49 179	43.66 209	70.90 177	80.93 192	83.40 187	96.59 143	95.66 138

WR-MVS		72.93 184	73.57 179	72.19 191	78.14 191	87.71 183	76.21 192	73.02 151	67.78 180	50.09 188	50.35 181	50.53 183	61.27 205	80.42 196	83.10 191	94.43 190	95.11 149

TransMVSNet (Re)		72.90 185	70.51 198	75.69 174	80.88 165	85.26 201	79.25 182	78.43 116	56.13 211	52.81 176	46.81 198	48.20 194	66.77 191	85.18 164	83.70 185	95.98 169	88.28 193

WR-MVS_H		72.69 186	72.80 188	72.56 190	77.94 192	87.83 182	75.26 196	71.53 165	64.75 193	52.19 178	49.83 182	48.62 189	61.96 203	81.12 191	82.44 193	96.50 149	95.00 152

SixPastTwentyTwo		72.65 187	73.22 183	71.98 194	78.40 188	87.64 186	70.09 204	70.37 175	66.49 185	47.60 197	65.09 130	45.94 200	73.09 173	78.94 200	78.66 203	92.33 203	89.82 188

LTVRE_ROB	71.82 16	72.62 188	71.77 192	73.62 183	80.74 166	87.59 187	80.42 178	70.37 175	49.73 215	37.12 218	59.76 149	42.52 214	80.92 142	83.20 179	85.61 164	92.13 204	93.95 162
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
PS-CasMVS		72.37 189	71.47 196	73.43 184	77.32 197	86.43 194	75.99 193	71.94 161	63.37 196	49.24 192	49.07 188	42.42 215	69.60 183	80.59 195	83.18 190	96.48 151	95.23 146

MVS-HIRNet		72.32 190	73.45 182	71.00 197	80.58 167	89.97 151	68.51 208	55.28 215	70.89 169	52.27 177	39.09 210	57.11 166	75.02 169	85.76 158	86.33 152	94.36 191	85.00 202

PEN-MVS		72.24 191	71.30 197	73.33 186	77.08 199	85.57 197	76.75 188	72.52 156	63.89 195	48.12 195	50.79 176	43.09 212	69.03 187	78.54 202	83.46 186	96.50 149	93.76 167

v7n		72.11 192	71.66 193	72.63 189	75.26 205	86.85 191	76.74 189	68.77 183	62.70 199	49.40 190	45.92 199	43.51 210	70.63 180	84.16 171	83.21 189	94.99 183	95.25 144

EG-PatchMatch MVS		71.81 193	71.54 195	72.12 192	80.53 168	89.94 152	78.51 184	66.56 189	57.38 207	47.46 198	44.28 205	52.22 174	63.10 201	85.22 163	84.42 178	96.56 147	87.35 197

CMPMVS	54.54 17	71.74 194	67.94 203	76.16 167	90.41 90	93.25 123	78.32 186	75.60 137	59.81 203	53.95 172	44.64 203	51.22 178	70.70 178	74.59 211	75.88 209	88.01 211	76.23 213
M. Jancosek, T. Pajdla: Multi-View Reconstruction Preserving Weakly-Supported Surfaces. CVPR 2011
MDTV_nov1_ep13_2view		71.65 195	73.08 186	69.97 200	75.22 206	86.81 192	73.98 198	59.61 205	69.75 173	48.01 196	54.21 168	53.06 173	69.19 185	78.50 203	80.43 198	93.84 194	88.79 191

pmnet_mix02		71.64 196	72.36 190	70.81 198	78.39 189	85.57 197	68.64 206	73.65 148	72.13 162	45.07 204	56.01 161	50.61 182	65.34 195	76.21 208	76.60 207	93.75 197	89.35 189

gm-plane-assit		71.33 197	75.18 168	66.83 203	79.06 178	75.57 215	48.05 222	60.33 200	48.28 216	34.67 222	44.34 204	67.70 137	79.78 147	97.25 11	96.21 13	99.10 10	96.92 114

DTE-MVSNet		71.19 198	70.45 199	72.06 193	76.61 201	84.59 203	75.61 195	72.32 157	63.12 198	45.70 203	50.72 177	43.02 213	65.89 193	77.53 207	82.23 194	96.26 158	91.93 178

pmmvs6		70.29 199	67.90 204	73.07 187	76.17 202	85.31 200	76.29 191	70.75 172	47.39 218	55.33 166	37.15 216	50.49 184	69.55 184	82.96 182	80.85 196	90.34 210	91.18 181

PM-MVS		70.17 200	69.42 201	71.04 196	70.82 213	81.26 210	71.25 202	67.80 186	69.16 177	51.04 184	53.15 172	34.93 219	72.19 176	80.30 197	76.95 206	93.16 201	90.21 185

pmmvs-eth3d		69.59 201	67.57 206	71.95 195	70.04 214	80.05 211	71.48 200	70.00 179	62.57 200	55.99 164	44.92 201	35.73 218	70.64 179	81.56 190	79.69 199	93.55 198	88.43 192

N_pmnet		68.54 202	67.83 205	69.38 201	75.77 203	81.90 207	66.21 210	72.53 155	65.91 189	46.09 202	44.67 202	45.48 202	63.82 199	74.66 210	77.39 205	91.87 206	84.77 203

Anonymous20231206		68.09 203	68.68 202	67.39 202	75.16 207	82.55 205	69.33 205	70.06 178	63.34 197	42.28 209	37.91 214	43.12 211	52.67 210	83.56 173	82.71 192	94.84 186	87.59 195

EU-MVSNet		68.07 204	70.25 200	65.52 204	74.68 209	81.30 209	68.53 207	70.31 177	62.40 201	37.43 217	54.62 167	48.36 192	51.34 211	78.32 204	79.27 200	90.84 208	87.47 196

GG-mvs-BLEND		65.67 205	93.78 39	32.89 218	0.47 229	99.35 8	96.92 31	0.22 227	93.28 60	0.51 230	84.07 54	92.50 39	0.62 227	93.59 71	93.86 59	98.59 44	99.79 10

test20.03		65.17 206	67.41 207	62.55 206	75.35 204	79.31 212	62.22 212	68.83 182	56.50 210	35.35 221	51.97 174	44.70 207	40.01 216	80.69 194	79.25 201	93.55 198	79.47 212

MDA-MVSNet-bldmvs		62.23 207	61.13 211	63.52 205	58.94 220	82.44 206	60.71 215	73.28 150	57.22 208	38.42 215	49.63 185	27.64 226	62.83 202	54.98 218	74.16 210	86.96 213	81.83 209

new_pmnet		61.60 208	62.68 209	60.35 209	63.02 217	74.93 216	60.97 214	58.86 207	64.21 194	35.38 220	39.51 209	39.89 216	57.37 208	72.78 212	72.56 212	86.49 215	74.85 215

new-patchmatchnet		60.74 209	59.78 213	61.87 207	69.52 215	76.67 214	57.99 218	65.78 193	52.63 213	38.47 214	38.08 213	32.92 222	48.88 213	68.50 213	69.87 213	90.56 209	79.75 211

pmmvs3		60.52 210	60.87 212	60.12 210	61.38 218	71.62 217	57.42 219	53.94 216	48.09 217	35.95 219	38.62 211	32.19 225	64.12 198	75.33 209	77.99 204	87.89 212	82.28 208

MIMVSNet1		60.51 211	61.43 210	59.44 211	48.75 223	77.21 213	60.98 213	66.84 188	52.09 214	38.74 213	29.29 219	39.40 217	48.08 214	77.60 206	78.87 202	93.22 200	75.56 214

test_method		60.40 212	66.30 208	53.52 213	37.48 227	64.10 221	55.56 220	42.45 222	71.79 167	41.87 210	33.74 217	46.80 196	61.71 204	79.18 199	73.33 211	82.01 217	95.17 147

FPMVS		56.54 213	52.82 215	60.87 208	74.90 208	67.58 220	67.69 209	65.38 194	57.86 206	41.51 211	37.83 215	34.19 220	41.21 215	55.88 217	53.09 219	74.55 220	63.31 218

WB-MVS		47.20 214	51.37 216	42.35 216	71.55 211	57.66 223	32.77 226	70.86 170	47.39 218	6.95 229	48.14 192	32.52 223	12.95 224	61.73 216	61.27 216	59.00 224	50.85 222

PMVS	42.57 18	45.71 215	42.61 218	49.32 214	61.35 219	37.82 226	36.96 224	60.10 202	37.20 221	41.50 212	28.53 220	33.11 221	28.82 221	53.45 219	48.70 221	67.22 222	59.42 219
Y. Furukawa, J. Ponce: Accurate, dense, and robust multiview stereopsis. PAMI (2010)
Gipuma		43.95 216	42.62 217	45.50 215	50.79 222	41.20 225	35.55 225	52.51 218	52.95 212	29.09 223	12.92 222	11.48 229	38.15 217	62.01 215	66.62 215	66.89 223	51.17 220
S. Galliani, K. Lasinger, K. Schindler: Massively Parallel Multiview Stereopsis by Surface Normal Diffusion. ICCV 2015
PMMVS2		41.25 217	42.55 219	39.74 217	43.25 224	55.05 224	38.15 223	47.11 221	31.78 222	11.83 226	21.16 221	19.12 227	20.98 223	49.95 221	56.09 218	77.09 218	64.68 217

E-PMN		27.87 218	24.36 221	31.97 219	41.27 226	25.56 229	16.62 228	49.16 219	22.00 224	9.90 227	11.75 224	7.86 231	29.57 220	22.22 223	34.70 222	45.27 225	46.41 223

MVE	32.98 19	27.61 219	29.89 220	24.94 221	21.97 228	37.22 227	15.56 230	38.83 223	17.49 225	14.72 225	11.64 226	5.62 232	21.26 222	35.20 222	50.95 220	37.29 227	51.13 221
Simon Fuhrmann, Fabian Langguth, Michael Goesele: MVE - A Multi-View Reconstruction Environment. EUROGRAPHICS Workshops on Graphics and Cultural Heritage (2014)
EMVS		26.96 220	22.96 222	31.63 220	41.91 225	25.73 228	16.30 229	49.10 220	22.38 223	9.03 228	11.22 227	8.12 230	29.93 219	20.16 224	31.04 223	43.49 226	42.04 224

testmvs		5.16 221	8.14 223	1.69 222	0.36 230	1.65 230	3.02 231	0.66 225	7.17 226	0.50 231	12.58 223	0.69 233	4.67 225	5.42 225	5.65 224	0.92 228	23.86 226

test123		4.39 222	7.11 224	1.21 223	0.11 231	1.16 231	1.67 232	0.35 226	5.91 227	0.16 232	11.65 225	0.16 234	4.45 226	1.72 226	4.92 225	0.51 229	24.28 225

uanet_test		0.00 223	0.00 225	0.00 224	0.00 232	0.00 232	0.00 233	0.00 228	0.00 228	0.00 233	0.00 228	0.00 235	0.00 228	0.00 227	0.00 226	0.00 230	0.00 227

sosnet-low-res		0.00 223	0.00 225	0.00 224	0.00 232	0.00 232	0.00 233	0.00 228	0.00 228	0.00 233	0.00 228	0.00 235	0.00 228	0.00 227	0.00 226	0.00 230	0.00 227

sosnet		0.00 223	0.00 225	0.00 224	0.00 232	0.00 232	0.00 233	0.00 228	0.00 228	0.00 233	0.00 228	0.00 235	0.00 228	0.00 227	0.00 226	0.00 230	0.00 227

TPM-MVS					99.19 1	99.43 7	99.16 2			85.97 32	94.75 25	97.40 13	97.76 1			98.95 24	95.69 133
Ray Leroy Khuboni and Hongjun Xu: Textureless Resilient Propagation Matching in Multiple View Stereosis (TPM-MVS). SATNAC 2025
RE-MVS-def										43.17 207

9.14												97.59 10

SR-MVS					98.52 20			93.70 22				96.63 21

Anonymous202405211			81.72 143		88.09 122	94.27 114	89.62 97	82.14 83	82.27 132		48.83 189	72.58 117	91.08 64	87.40 143	88.70 133	94.90 184	97.99 88

our_test_3					78.55 184	84.98 202	70.12 203

ambc			57.08 214		58.68 221	67.71 219	60.07 216		57.13 209	42.79 208	30.00 218	11.64 228	50.18 212	78.89 201	69.14 214	82.64 216	85.02 201

MTAPA										93.37 8		95.71 27

MTMP										93.84 5		94.86 30

Patchmatch-RL test							19.65 227

tmp_tt				57.89 212	79.94 172	59.29 222	52.84 221	36.65 224	94.77 50	68.22 119	72.96 99	65.62 144	33.65 218	66.20 214	58.02 217	76.06 219

XVS					92.16 71	98.56 35	91.04 85			81.00 62		93.49 34				98.00 79

X-MVStestdata					92.16 71	98.56 35	91.04 85			81.00 62		93.49 34				98.00 79

mPP-MVS					97.95 29							92.24 44

NP-MVS									94.12 54

Patchmtry						92.08 139	83.86 141	58.37 209		56.28 157

DeepMVS_CX						70.68 218	59.61 217	67.36 187	72.12 163	38.41 216	53.88 171	32.44 224	55.15 209	50.88 220		74.35 221	68.42 216

High-resolution multi-view benchmark