This table lists the benchmark results for the low-res two-view scenario. This benchmark evaluates the Middlebury stereo metrics (for all metrics, smaller is better):

The mask determines whether the metric is evaluated for all pixels with ground truth, or only for pixels which are visible in both images (non-occluded).
The coverage selector allows to limit the table to results for all pixels (dense), or a given minimum fraction of pixels.

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

Click one or more dataset result cells or column headers to show visualizations. Most visualizations are only available for training datasets. The visualizations may not work with mobile browsers.




Method Infoalllakes. 1llakes. 1ssand box 1lsand box 1sstora. room 1lstora. room 1sstora. room 2lstora. room 2sstora. room 2 1lstora. room 2 1sstora. room 2 2lstora. room 2 2sstora. room 3lstora. room 3stunnel 1ltunnel 1stunnel 2ltunnel 2stunnel 3ltunnel 3s
sorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
CREStereotwo views1.09
1
0.51
6
2.16
1
0.11
1
0.24
7
1.67
1
0.03
1
4.24
2
6.91
19
0.80
1
1.73
2
0.44
2
0.64
1
1.58
1
0.61
1
0.00
1
0.00
1
0.00
1
0.00
1
0.01
1
0.02
1
PMTNettwo views1.48
2
0.43
2
2.77
2
0.31
3
0.17
4
2.94
10
0.33
9
3.66
1
5.27
7
1.11
2
1.28
1
0.28
1
0.92
5
4.32
9
0.90
2
4.89
129
0.00
1
0.00
1
0.00
1
0.03
6
0.05
2
DPM-Stereotwo views2.12
3
0.98
19
3.94
14
0.20
2
0.11
1
4.73
32
0.13
3
9.27
17
4.40
4
6.46
19
5.03
6
0.45
3
0.72
2
3.79
7
1.94
6
0.00
1
0.01
33
0.00
1
0.00
1
0.05
10
0.24
25
Gwc-CoAtRStwo views2.16
4
0.42
1
3.13
5
1.28
10
0.31
9
3.52
14
0.39
11
6.56
4
6.31
15
5.38
12
5.11
7
5.47
14
1.70
9
2.00
2
1.08
3
0.01
21
0.19
91
0.00
1
0.25
91
0.02
4
0.16
18
FENettwo views2.35
5
0.59
8
3.28
7
2.32
43
0.27
8
2.78
8
0.13
3
7.68
8
6.09
11
5.38
12
2.12
3
4.84
10
3.33
20
4.98
11
2.66
8
0.02
28
0.03
50
0.00
1
0.00
1
0.04
8
0.40
31
RAFT-Stereopermissivetwo views2.60
6
0.46
4
2.94
3
1.10
7
0.18
5
3.61
16
0.60
21
6.88
6
3.50
1
9.13
46
17.23
78
1.77
4
0.77
3
2.03
3
1.65
4
0.00
1
0.01
33
0.00
1
0.00
1
0.01
1
0.21
21
Lahav Lipson, Zachary Teed, and Jia Deng: RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matching. 3DV
R-Stereo Traintwo views2.60
6
0.46
4
2.94
3
1.10
7
0.18
5
3.61
16
0.60
21
6.88
6
3.50
1
9.13
46
17.23
78
1.77
4
0.77
3
2.03
3
1.65
4
0.00
1
0.01
33
0.00
1
0.00
1
0.01
1
0.21
21
ACVNettwo views2.86
8
0.96
16
3.87
10
0.86
5
0.61
20
2.63
6
0.54
15
8.93
12
6.12
12
5.09
7
9.32
27
5.70
17
3.84
26
2.68
5
5.56
40
0.00
1
0.00
1
0.00
1
0.00
1
0.26
55
0.23
23
DN-CSS_ROBtwo views3.00
9
2.14
70
7.33
50
2.67
64
0.78
26
3.13
11
0.07
2
6.68
5
4.22
3
5.51
15
12.12
48
2.33
6
1.21
6
8.69
37
2.51
7
0.03
36
0.00
1
0.00
1
0.00
1
0.42
76
0.15
13
DMCAtwo views3.08
10
1.10
24
6.20
41
1.82
27
0.90
32
3.80
22
0.55
17
9.76
20
5.42
8
6.56
21
6.47
12
7.06
27
2.09
10
4.05
8
5.46
39
0.00
1
0.01
33
0.00
1
0.00
1
0.11
24
0.16
18
HITNettwo views3.11
11
1.38
36
5.35
30
2.23
37
0.12
2
5.59
40
0.58
19
9.95
22
5.72
10
6.62
23
7.91
17
3.90
7
3.24
18
4.39
10
4.58
29
0.01
21
0.00
1
0.00
1
0.00
1
0.08
15
0.57
47
Vladimir Tankovich, Christian Häne, Yinda Zhang, Adarsh Kowdle, Sean Fanello, Sofien Bouaziz: HITNet: Hierarchical Iterative Tile Refinement Network for Real-time Stereo Matching. CVPR 2021
ccstwo views3.26
12
0.58
7
3.71
8
2.06
32
0.53
18
2.06
2
0.49
13
13.61
70
14.34
74
4.14
6
5.97
10
5.46
13
2.40
13
5.25
12
3.12
14
0.09
60
0.08
70
0.12
90
0.10
75
0.22
49
0.94
81
AdaStereotwo views3.34
13
0.74
9
4.00
16
3.10
81
0.51
17
4.22
26
1.25
41
12.84
58
8.39
38
6.33
18
9.55
30
5.77
19
1.54
8
5.53
13
2.74
10
0.02
28
0.00
1
0.00
1
0.00
1
0.03
6
0.15
13
Xiao Song, Guorun Yang, Xinge Zhu, Hui Zhou, Zhe Wang, Jianping Shi: AdaStereo: A Simple and Efficient Approach for Adaptive Stereo Matching. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2021.
DMCA-RVCcopylefttwo views3.59
14
1.98
65
8.68
67
2.30
42
1.48
50
3.54
15
0.55
17
8.40
10
5.43
9
9.37
51
5.90
9
8.34
37
3.21
16
7.19
19
4.46
26
0.06
55
0.11
76
0.08
86
0.02
49
0.21
46
0.40
31
NOSS_ROBtwo views3.67
15
0.86
13
3.81
9
2.53
57
1.11
39
6.10
45
0.72
32
11.08
38
12.36
62
5.45
14
8.91
26
5.65
15
2.09
10
8.30
32
3.16
15
0.20
75
0.07
69
0.00
1
0.00
1
0.11
24
0.80
74
GANet-RSSMtwo views3.67
15
1.16
27
4.56
21
3.24
85
2.03
66
2.63
6
0.50
14
11.53
47
6.83
17
6.25
17
10.58
36
7.35
29
6.65
61
6.77
17
2.94
12
0.00
1
0.00
1
0.08
86
0.00
1
0.26
55
0.12
6
BEATNet_4xtwo views3.69
17
2.12
69
8.11
60
1.82
27
0.16
3
5.30
36
1.04
38
10.87
33
6.14
14
7.09
30
8.42
24
4.99
11
4.42
34
5.89
14
6.23
45
0.04
45
0.05
58
0.00
1
0.00
1
0.32
69
0.84
75
PSMNet-RSSMtwo views3.69
17
1.36
35
4.54
20
2.44
50
1.90
61
2.35
4
0.69
29
11.77
49
6.38
16
6.60
22
11.85
46
7.70
32
5.61
54
6.66
16
3.67
17
0.00
1
0.00
1
0.02
61
0.01
44
0.17
39
0.10
3
CFNet_RVCtwo views3.70
19
1.59
49
3.87
10
1.68
18
2.42
72
3.20
12
0.66
26
8.92
11
7.76
27
5.09
7
9.95
31
10.77
53
5.43
50
7.41
21
4.71
31
0.00
1
0.00
1
0.03
69
0.01
44
0.30
63
0.11
4
CFNet-ftpermissivetwo views3.70
19
1.59
49
3.87
10
1.68
18
2.42
72
3.20
12
0.66
26
8.93
12
7.76
27
5.09
7
9.95
31
10.77
53
5.43
50
7.41
21
4.71
31
0.00
1
0.00
1
0.03
69
0.01
44
0.30
63
0.11
4
cf-rtwo views3.76
21
1.43
37
5.34
29
2.66
62
2.79
79
2.85
9
0.62
24
9.08
15
6.91
19
7.06
27
11.90
47
9.64
45
3.83
25
6.23
15
4.53
28
0.00
1
0.00
1
0.00
1
0.00
1
0.19
43
0.17
20
MLCVtwo views3.76
21
1.52
44
6.93
48
1.66
17
0.31
9
4.31
28
0.32
8
7.92
9
7.68
26
9.55
52
12.12
48
5.17
12
2.74
15
10.00
58
4.49
27
0.01
21
0.00
1
0.00
1
0.00
1
0.16
35
0.27
27
DeepPruner_ROBtwo views3.82
23
1.87
56
5.65
35
1.31
11
1.64
51
3.62
18
0.81
35
14.47
78
5.03
6
8.03
36
10.78
38
6.95
24
2.26
12
8.92
41
4.32
24
0.07
57
0.03
50
0.00
1
0.01
44
0.37
74
0.29
28
acv_fttwo views3.90
24
0.96
16
5.85
37
2.56
60
4.70
113
7.27
60
0.84
36
9.17
16
6.12
12
10.00
55
9.32
27
5.70
17
4.07
31
2.68
5
8.24
55
0.00
1
0.00
1
0.00
1
0.00
1
0.35
73
0.23
23
STTStereotwo views3.90
24
1.43
37
8.20
62
3.03
77
2.28
70
3.64
19
0.64
25
10.11
23
7.09
21
10.20
56
7.08
14
8.73
42
3.57
21
7.44
23
4.00
19
0.00
1
0.02
42
0.01
51
0.01
44
0.04
8
0.40
31
hitnet-ftcopylefttwo views3.95
26
1.16
27
3.92
13
0.89
6
0.73
24
4.88
34
0.54
15
9.78
21
7.91
32
7.42
33
10.24
33
12.50
65
4.84
41
8.15
29
5.56
40
0.03
36
0.00
1
0.06
80
0.00
1
0.24
53
0.12
6
ccs_robtwo views3.96
27
1.98
65
5.57
33
2.74
66
0.92
33
6.29
48
0.21
6
10.74
30
9.59
48
7.08
29
11.51
43
8.47
39
2.64
14
8.28
30
2.71
9
0.00
1
0.00
1
0.00
1
0.00
1
0.29
61
0.13
8
ac_64two views3.98
28
0.86
13
5.10
26
3.17
82
3.43
91
5.80
42
0.70
30
10.87
33
4.96
5
8.94
41
6.46
11
10.70
52
5.52
52
7.44
23
5.10
37
0.03
36
0.00
1
0.00
1
0.00
1
0.14
28
0.31
29
CFNettwo views4.03
29
1.97
64
6.30
43
2.70
65
1.64
51
5.41
38
0.22
7
12.14
52
10.28
49
5.24
11
11.15
41
6.43
22
3.70
23
8.97
44
4.01
20
0.00
1
0.00
1
0.00
1
0.00
1
0.30
63
0.13
8
iResNettwo views4.04
30
1.46
41
8.72
69
3.25
86
0.36
11
4.54
31
0.47
12
9.06
14
10.40
52
7.98
34
12.82
56
4.06
8
3.22
17
8.82
39
4.95
35
0.03
36
0.01
33
0.00
1
0.00
1
0.14
28
0.55
45
NLCA_NET_v2_RVCtwo views4.11
31
1.53
46
6.87
47
3.04
78
3.33
89
4.34
30
0.59
20
11.36
44
7.92
33
8.98
42
7.98
18
9.72
47
3.82
24
7.92
26
4.43
25
0.15
66
0.02
42
0.03
69
0.03
55
0.07
11
0.14
10
Zhibo Rao, Mingyi He, Yuchao Dai, Zhidong Zhu, Bo Li, and Renjie He.: NLCA-Net: A non-local context attention network for stereo matching.
FADNet-RVC-Resampletwo views4.11
31
2.25
73
13.88
94
1.64
15
0.66
21
5.91
44
2.34
53
10.87
33
9.09
45
6.53
20
4.34
5
5.69
16
4.71
39
8.01
28
4.64
30
0.18
72
0.22
94
0.17
102
0.12
78
0.46
82
0.59
48
CC-Net-ROBtwo views4.12
33
1.54
47
6.86
46
2.98
74
3.04
83
4.18
25
0.67
28
11.10
39
8.07
36
9.05
44
8.35
23
9.47
43
3.99
29
7.94
27
4.73
33
0.14
65
0.02
42
0.03
69
0.03
55
0.07
11
0.25
26
GwcNet-RSSMtwo views4.13
34
1.71
52
6.25
42
2.50
53
1.93
64
2.39
5
1.49
45
9.59
19
8.71
41
6.66
24
15.04
65
8.47
39
3.84
26
6.87
18
6.74
47
0.00
1
0.00
1
0.04
75
0.00
1
0.29
61
0.15
13
FADNet_RVCtwo views4.24
35
2.32
79
14.23
97
1.18
9
0.79
27
5.64
41
0.60
21
11.50
45
7.19
22
3.68
3
8.08
21
4.14
9
4.06
30
10.29
61
7.47
51
0.36
91
0.44
105
0.18
103
0.47
108
0.93
103
1.23
90
HSMtwo views4.25
36
1.10
24
4.14
17
1.78
24
2.14
67
6.67
54
1.08
39
12.92
59
6.90
18
7.06
27
8.17
22
14.17
74
6.03
56
9.16
45
3.11
13
0.00
1
0.00
1
0.00
1
0.00
1
0.09
20
0.41
36
FADNet-RVCtwo views4.36
37
2.62
89
14.51
100
2.01
31
0.44
15
4.27
27
1.27
42
12.61
56
7.89
30
3.76
4
4.05
4
10.63
51
4.34
33
9.87
54
7.66
52
0.03
36
0.09
74
0.00
1
0.03
55
0.65
91
0.50
40
HSM-Net_RVCpermissivetwo views4.40
38
0.44
3
3.22
6
0.72
4
0.68
22
7.06
57
1.82
49
12.57
55
8.48
39
9.04
43
12.20
51
14.98
77
5.14
45
7.32
20
4.19
21
0.02
28
0.02
42
0.00
1
0.00
1
0.02
4
0.14
10
Gengshan Yang, Joshua Manela, Michael Happold, and Deva Ramanan: Hierarchical Deep Stereo Matching on High-resolution Images. CVPR 2019
DSFCAtwo views4.48
39
1.03
22
6.84
45
1.61
14
1.68
55
7.26
59
4.46
76
11.27
43
7.51
23
11.01
66
6.89
13
8.49
41
4.54
35
10.41
63
6.18
44
0.03
36
0.06
63
0.02
61
0.03
55
0.13
26
0.14
10
FADNettwo views4.55
40
2.35
80
13.60
93
1.87
29
0.80
28
4.74
33
0.75
33
14.47
78
12.41
63
4.07
5
5.45
8
9.64
45
5.03
43
8.86
40
4.82
34
0.20
75
0.04
54
0.07
82
0.05
65
1.27
111
0.40
31
HGLStereotwo views4.58
41
1.04
23
6.02
40
3.30
89
5.14
116
6.36
49
0.71
31
11.04
37
7.60
25
8.04
37
12.85
57
6.90
23
8.51
75
7.85
25
5.93
42
0.02
28
0.02
42
0.00
1
0.00
1
0.14
28
0.15
13
TDLMtwo views4.58
41
1.79
55
5.03
25
2.45
51
1.32
47
4.04
23
7.15
99
11.60
48
11.46
58
6.91
26
12.69
55
5.94
20
5.03
43
9.74
50
3.80
18
0.67
107
0.00
1
0.01
51
0.05
65
0.23
51
1.78
107
CBMV_ROBtwo views4.66
43
0.85
12
3.94
14
1.64
15
0.75
25
8.38
65
1.90
50
11.19
40
10.36
50
9.20
50
15.21
66
7.28
28
6.99
64
9.48
48
5.24
38
0.04
45
0.05
58
0.00
1
0.00
1
0.07
11
0.64
55
iResNet_ROBtwo views4.67
44
1.77
53
5.93
38
2.47
52
0.95
34
3.78
20
0.36
10
15.65
87
17.36
86
8.75
39
11.23
42
7.05
26
4.60
37
8.94
42
4.23
22
0.01
21
0.00
1
0.00
1
0.00
1
0.16
35
0.15
13
CVANet_RVCtwo views4.68
45
1.93
61
5.57
33
2.61
61
1.86
60
4.12
24
4.94
79
12.11
50
9.29
46
8.05
38
11.08
40
10.28
49
6.48
59
9.52
49
3.19
16
0.45
95
0.01
33
0.00
1
0.12
78
0.23
51
1.75
106
iResNetv2_ROBtwo views4.72
46
2.26
75
8.87
72
3.20
84
1.25
44
4.32
29
1.73
48
14.22
76
10.62
54
10.53
60
12.13
50
7.77
33
5.54
53
8.29
31
2.76
11
0.00
1
0.00
1
0.00
1
0.00
1
0.42
76
0.55
45
NVstereo2Dtwo views4.82
47
1.29
32
7.98
59
3.68
97
3.48
92
8.02
62
3.10
59
11.25
41
15.50
78
5.14
10
7.71
15
8.21
36
4.68
38
10.19
60
4.30
23
0.02
28
0.19
91
0.13
93
0.36
102
0.07
11
1.12
88
StereoDRNet-Refinedtwo views4.84
48
0.96
16
5.13
27
2.22
36
0.46
16
9.10
68
0.16
5
10.72
29
9.40
47
13.41
82
12.50
53
10.02
48
4.58
36
8.60
35
8.71
61
0.01
21
0.00
1
0.01
51
0.05
65
0.24
53
0.53
43
Rohan Chabra, Julian Straub, Chris Sweeney, Richard Newcombe, Henry Fuchs: StereoDRNet. CVPR
DLCB_ROBtwo views4.87
49
1.45
40
4.75
23
2.52
56
1.13
42
6.25
47
3.01
57
10.44
28
8.34
37
11.05
67
13.43
59
14.04
71
3.90
28
9.74
50
6.80
48
0.00
1
0.00
1
0.00
1
0.00
1
0.08
15
0.37
30
RASNettwo views5.00
50
1.19
30
5.69
36
3.68
97
4.80
114
5.04
35
1.29
43
10.26
24
8.99
44
5.89
16
11.65
44
10.55
50
9.76
79
9.96
57
10.32
79
0.20
75
0.00
1
0.00
1
0.00
1
0.09
20
0.73
68
SGM-Foresttwo views5.40
51
0.84
11
4.19
18
1.43
12
0.68
22
10.12
78
6.41
90
12.13
51
11.86
60
11.50
69
14.43
63
11.62
59
5.20
46
9.93
56
6.16
43
0.16
68
0.04
54
0.00
1
0.00
1
0.08
15
1.24
91
Johannes L. Schönberger, Sudipta Sinha, Marc Pollefeys: Learning to Fuse Proposals from Multiple Scanline Optimizations in Semi-Global Matching. ECCV 2018
PA-Nettwo views5.40
51
2.35
80
9.55
79
2.75
67
2.21
68
8.74
66
3.92
66
13.05
62
13.35
70
7.30
32
7.98
18
8.08
34
6.39
58
11.17
67
9.30
68
0.02
28
0.03
50
0.00
1
0.03
55
0.16
35
1.55
99
Zhibo Rao, Mingyi He, Yuchao Dai, Zhelun Shen: Patch Attention Network with Generative Adversarial Model for Semi-Supervised Binocular Disparity Prediction.
AANet_RVCtwo views5.41
53
2.61
88
8.17
61
2.25
39
1.30
45
2.27
3
1.64
46
10.82
31
18.54
93
8.01
35
18.09
83
8.39
38
3.31
19
11.37
70
8.98
65
1.15
114
0.38
101
0.02
61
0.00
1
0.17
39
0.70
65
PSMNet_ROBtwo views5.41
53
2.45
83
7.39
53
2.36
45
1.90
61
10.11
77
6.54
91
16.32
93
7.59
24
6.84
25
10.83
39
12.06
63
4.13
32
8.73
38
10.24
77
0.02
28
0.01
33
0.01
51
0.11
77
0.26
55
0.40
31
CBMVpermissivetwo views5.97
55
1.52
44
4.51
19
1.99
30
0.53
18
10.02
76
6.94
96
13.27
65
13.80
73
13.08
81
15.93
73
11.36
58
8.11
71
9.82
52
7.27
50
0.08
58
0.15
85
0.00
1
0.00
1
0.30
63
0.74
70
Konstantinos Batsos, Changjiang Cai, Philippos Mordohai: CBMV: A Coalesced Bidirectional Matching Volume for Disparity Estimation. Computer Vision and Pattern Recognition (CVPR) 2018
StereoDRNettwo views6.02
56
2.40
82
8.22
63
3.43
91
4.43
104
10.57
81
4.21
68
19.44
107
11.24
57
12.86
76
8.06
20
11.64
60
4.79
40
8.65
36
9.32
69
0.01
21
0.10
75
0.00
1
0.03
55
0.34
72
0.69
64
DRN-Testtwo views6.23
57
1.51
42
7.53
55
3.07
80
3.69
97
12.17
90
3.23
62
20.80
116
11.11
56
12.64
75
13.02
58
10.91
57
5.26
47
8.49
33
10.12
75
0.05
52
0.05
58
0.04
75
0.04
64
0.28
59
0.65
56
NCCL2two views6.27
58
2.46
84
7.00
49
2.36
45
1.11
39
9.48
71
11.14
112
13.01
61
10.40
52
9.94
54
9.49
29
22.11
109
7.42
67
9.90
55
8.56
58
0.01
21
0.00
1
0.02
61
0.03
55
0.20
44
0.73
68
ETE_ROBtwo views6.28
59
2.60
87
9.22
76
1.71
22
0.84
30
6.44
50
6.78
93
13.16
63
8.80
42
13.01
79
15.93
73
21.01
107
6.18
57
10.97
64
8.11
54
0.04
45
0.01
33
0.00
1
0.02
49
0.16
35
0.70
65
XPNet_ROBtwo views6.46
60
1.88
57
7.41
54
2.81
68
0.95
34
6.45
51
6.82
94
13.58
68
8.80
42
15.50
91
15.86
71
19.59
101
7.12
66
11.41
71
9.86
73
0.02
28
0.04
54
0.00
1
0.03
55
0.20
44
0.79
73
DANettwo views6.46
60
1.58
48
9.58
81
4.10
102
3.99
100
9.13
69
1.34
44
10.37
27
7.76
27
13.76
86
15.66
69
16.14
81
7.03
65
13.07
80
13.62
99
0.19
73
0.12
79
0.02
61
0.02
49
0.15
34
1.61
100
NaN_ROBtwo views6.51
62
2.25
73
8.77
70
1.69
20
1.79
57
9.71
73
10.94
110
16.20
91
16.13
80
13.04
80
8.86
25
11.77
61
5.94
55
12.88
79
8.01
53
0.05
52
0.13
82
0.07
82
0.20
86
0.32
69
1.50
96
DISCOtwo views6.58
63
0.98
19
6.44
44
3.75
99
1.18
43
11.19
83
3.29
64
12.92
59
16.72
82
7.27
31
11.74
45
17.34
86
6.60
60
19.59
114
11.59
89
0.00
1
0.00
1
0.00
1
0.00
1
0.45
81
0.50
40
RYNettwo views6.82
64
1.51
42
7.35
52
1.79
25
4.52
108
16.68
101
4.41
75
14.19
75
17.43
87
11.64
70
7.86
16
15.03
78
8.81
76
10.10
59
14.16
101
0.04
45
0.13
82
0.15
96
0.05
65
0.08
15
0.53
43
S-Stereotwo views6.85
65
0.93
15
10.76
85
3.34
90
3.51
93
6.55
52
6.01
86
13.75
72
23.86
107
9.65
53
15.54
68
11.81
62
8.24
72
8.59
34
11.23
87
0.13
64
0.16
86
0.00
1
0.42
105
0.14
28
2.39
114
GANettwo views6.86
66
1.78
54
5.98
39
3.05
79
1.12
41
10.12
78
9.38
108
13.41
67
8.67
40
11.46
68
29.87
109
13.94
70
8.25
73
12.06
74
6.33
46
0.40
93
0.06
63
0.16
99
0.33
99
0.18
42
0.59
48
GwcNetcopylefttwo views6.95
67
2.89
95
13.32
92
2.82
69
5.53
118
11.76
86
2.14
52
14.95
82
17.69
89
10.83
64
18.14
84
10.83
55
6.66
62
9.83
53
10.19
76
0.03
36
0.06
63
0.03
69
0.08
71
0.64
90
0.66
63
GANetREF_RVCpermissivetwo views6.97
68
3.59
103
9.14
75
4.00
100
0.40
12
13.17
92
11.27
114
15.95
89
17.51
88
10.62
61
10.66
37
12.15
64
7.97
70
13.18
81
5.04
36
0.45
95
0.00
1
0.01
51
0.15
81
3.47
126
0.60
50
Zhang, Feihu and Prisacariu, Victor and Yang, Ruigang and Torr, Philip HS: GA-Net: Guided Aggregation Net for End- to-end Stereo Matching. CVPR 2019
FAT-Stereotwo views6.98
69
0.79
10
7.57
56
2.54
59
1.80
58
6.16
46
4.33
74
19.16
106
18.82
94
10.75
63
17.08
77
18.23
92
13.56
94
9.25
46
8.44
56
0.03
36
0.16
86
0.01
51
0.10
75
0.14
28
0.65
56
LALA_ROBtwo views7.02
70
2.54
85
8.51
64
1.53
13
0.96
36
9.96
75
8.71
106
16.70
94
12.68
66
12.99
78
13.90
62
24.47
111
5.40
49
11.61
73
9.50
70
0.04
45
0.05
58
0.00
1
0.02
49
0.17
39
0.65
56
edge stereotwo views7.12
71
1.19
30
7.83
58
2.32
43
2.46
74
6.90
55
3.26
63
15.58
86
12.52
64
15.68
92
18.34
85
21.83
108
13.05
90
11.53
72
8.68
60
0.04
45
0.06
63
0.11
88
0.03
55
0.30
63
0.63
53
DeepPrunerFtwo views7.18
72
3.58
102
25.29
119
4.27
107
7.20
125
3.78
20
4.64
77
14.03
73
20.86
102
8.85
40
10.57
35
7.35
29
7.88
69
12.16
77
10.66
82
0.25
80
0.16
86
0.29
112
0.43
106
0.77
99
0.65
56
Anonymous Stereotwo views7.27
73
4.41
112
26.56
121
3.67
96
3.09
85
5.56
39
15.31
120
10.85
32
11.63
59
10.31
57
10.32
34
7.46
31
3.67
22
14.42
90
15.13
104
0.64
105
0.88
112
0.00
1
0.07
70
0.33
71
1.16
89
RGCtwo views7.40
74
2.77
91
7.71
57
4.43
110
4.87
115
8.88
67
2.96
56
16.05
90
12.76
67
11.96
73
19.57
87
18.32
93
11.57
86
13.89
88
9.58
72
0.83
110
0.02
42
0.24
111
0.41
104
0.40
75
0.77
72
RPtwo views7.41
75
1.90
59
7.33
50
4.26
106
5.25
117
6.64
53
3.46
65
12.44
54
16.23
81
10.36
58
23.00
93
18.72
94
16.59
104
11.10
65
8.47
57
1.10
113
0.02
42
0.12
90
0.16
82
0.43
78
0.65
56
Abc-Nettwo views7.43
76
2.28
76
8.58
65
3.28
87
4.48
105
8.34
63
4.27
71
20.45
114
13.66
71
9.17
48
17.52
80
16.49
82
14.31
99
13.71
84
8.95
63
1.73
120
0.12
79
0.01
51
0.50
110
0.21
46
0.60
50
Xing Li, Yangyu Fan, Guoyun Lv, and Haoyue Ma: Area-based Correlation and Non-local Attention Network for Stereo Matching. The Visual Computer
NCC-stereotwo views7.43
76
2.28
76
8.58
65
3.28
87
4.48
105
8.34
63
4.27
71
20.45
114
13.66
71
9.17
48
17.52
80
16.49
82
14.31
99
13.71
84
8.95
63
1.73
120
0.12
79
0.01
51
0.50
110
0.21
46
0.60
50
Nwc_Nettwo views7.54
78
1.91
60
8.71
68
4.16
103
3.42
90
10.77
82
2.00
51
20.33
113
12.19
61
9.10
45
24.04
97
17.67
88
13.58
95
12.14
76
9.12
66
0.94
112
0.00
1
0.00
1
0.02
49
0.22
49
0.41
36
psmorigintwo views7.74
79
2.19
72
18.99
109
2.51
54
0.89
31
7.08
58
2.74
55
11.25
41
7.90
31
20.49
105
22.36
91
20.16
102
11.95
87
13.84
86
10.61
81
0.04
45
0.18
89
0.06
80
0.24
89
0.27
58
1.07
86
ADCReftwo views7.76
80
2.31
78
19.09
110
2.82
69
3.29
88
11.88
87
3.06
58
11.52
46
10.89
55
13.69
85
25.85
103
8.20
35
8.89
78
8.95
43
22.66
113
0.16
68
0.08
70
0.16
99
0.34
100
0.47
84
0.91
78
pmcnntwo views8.17
81
1.70
51
11.49
87
3.18
83
3.07
84
9.66
72
6.06
88
13.60
69
17.71
90
15.23
90
24.02
95
27.61
117
8.81
76
10.36
62
9.51
71
0.08
58
0.08
70
0.00
1
0.00
1
0.31
68
0.87
76
stereogantwo views8.18
82
1.32
34
9.05
74
3.59
93
4.00
101
20.16
111
3.13
60
17.25
99
20.16
99
10.68
62
19.19
86
20.56
106
10.70
81
14.74
94
7.00
49
0.34
89
0.04
54
0.21
108
0.08
71
0.70
95
0.65
56
SuperBtwo views8.29
83
3.96
108
24.92
117
2.96
73
1.33
48
9.83
74
4.15
67
10.92
36
29.79
116
11.86
71
12.51
54
6.07
21
6.97
63
11.13
66
14.73
102
0.15
66
0.11
76
0.35
114
0.26
94
12.64
133
1.11
87
AF-Nettwo views8.37
84
2.16
71
9.01
73
3.65
95
4.58
111
9.40
70
2.63
54
17.71
101
21.32
104
10.38
59
24.51
98
20.43
103
16.67
105
12.54
78
10.76
85
0.70
108
0.00
1
0.12
90
0.00
1
0.47
84
0.42
38
CSANtwo views8.41
85
2.86
94
9.35
78
2.43
49
0.82
29
13.59
94
11.68
115
15.80
88
21.98
105
14.71
88
16.46
75
18.00
90
14.42
101
13.68
83
10.10
74
0.10
61
0.13
82
0.15
96
0.19
84
0.43
78
1.28
92
PWCDC_ROBbinarytwo views8.50
86
4.40
111
9.57
80
6.16
123
4.54
110
10.46
80
1.03
37
15.30
83
28.16
113
10.88
65
30.88
113
6.96
25
11.31
83
14.57
92
10.40
80
3.10
128
0.02
42
0.00
1
0.00
1
1.55
115
0.63
53
STTStereo_v2two views8.60
87
1.93
61
12.59
89
6.04
120
3.66
95
25.93
120
4.26
69
10.35
25
8.05
34
19.71
101
24.73
99
14.15
72
15.49
102
11.30
68
10.69
83
0.50
102
0.46
106
0.19
105
0.25
91
0.80
100
0.91
78
ADCP+two views8.60
87
2.78
92
17.44
107
1.69
20
4.27
103
16.95
104
5.29
81
13.38
66
12.52
64
13.57
84
17.57
82
10.86
56
11.28
82
18.43
110
23.91
119
0.03
36
0.05
58
0.01
51
0.18
83
0.47
84
1.31
93
G-Nettwo views8.60
87
1.93
61
12.59
89
6.04
120
3.66
95
25.93
120
4.26
69
10.35
25
8.05
34
19.71
101
24.73
99
14.15
72
15.49
102
11.30
68
10.69
83
0.50
102
0.46
106
0.19
105
0.25
91
0.80
100
0.91
78
PWC_ROBbinarytwo views8.89
90
4.49
113
15.62
102
2.92
71
4.49
107
7.89
61
1.23
40
17.29
100
20.00
97
16.54
94
28.06
106
14.37
75
11.98
89
14.93
95
15.43
105
0.47
99
0.00
1
0.05
78
0.00
1
0.67
92
1.31
93
aanetorigintwo views9.02
91
3.75
106
28.68
123
2.16
34
3.86
98
5.84
43
5.40
83
5.93
3
10.39
51
28.12
119
31.02
114
15.98
80
13.74
96
9.38
47
12.55
95
0.25
80
0.22
94
0.22
109
0.24
89
1.24
109
1.47
95
PASMtwo views9.11
92
5.68
118
25.06
118
4.33
108
4.18
102
7.01
56
8.48
104
12.37
53
20.95
103
13.49
83
15.37
67
18.80
95
11.52
85
14.70
93
12.70
96
0.93
111
1.19
119
0.32
113
1.21
122
1.66
116
2.22
113
MDST_ROBtwo views9.13
93
1.31
33
10.16
83
4.36
109
2.48
76
28.66
122
7.64
101
19.96
108
14.97
75
27.53
118
24.02
95
9.47
43
5.27
48
16.31
101
8.61
59
0.63
104
0.19
91
0.00
1
0.00
1
0.08
15
0.90
77
FBW_ROBtwo views9.17
94
2.08
68
10.31
84
2.51
54
1.66
54
13.22
93
6.04
87
23.09
123
20.11
98
19.55
99
15.80
70
18.97
98
11.96
88
19.34
111
11.38
88
0.66
106
0.30
100
2.15
130
0.90
117
1.15
108
2.16
111
XQCtwo views9.46
95
4.84
116
18.88
108
4.07
101
3.09
85
15.97
99
6.65
92
16.26
92
26.63
111
12.63
74
12.35
52
15.75
79
10.62
80
16.48
102
21.19
110
0.44
94
0.96
113
0.05
78
0.32
97
1.10
105
0.99
82
MSMD_ROBtwo views9.73
96
1.43
37
5.42
31
1.81
26
0.42
14
16.83
102
4.30
73
14.29
77
15.34
76
23.06
107
40.17
127
25.63
114
22.62
118
13.37
82
8.85
62
0.48
101
0.03
50
0.00
1
0.00
1
0.13
26
0.47
39
RTSCtwo views9.87
97
4.34
110
16.44
104
4.60
112
2.38
71
12.54
91
0.75
33
18.19
103
37.31
126
16.16
93
15.86
71
13.32
67
7.56
68
20.93
116
23.50
117
0.34
89
1.07
115
0.02
61
0.35
101
0.71
96
0.99
82
ADCPNettwo views10.18
98
3.40
100
33.24
125
2.38
47
1.74
56
17.75
107
7.65
102
13.20
64
13.03
69
14.09
87
21.72
89
19.36
99
11.40
84
15.43
98
23.00
115
0.28
86
1.15
118
0.39
117
1.48
126
0.68
93
2.21
112
WCMA_ROBtwo views10.20
99
1.99
67
9.33
77
3.01
76
2.46
74
15.78
98
7.67
103
12.75
57
15.41
77
25.08
116
33.42
119
27.06
116
19.86
110
13.87
87
12.47
94
1.28
116
0.38
101
0.18
103
0.20
86
0.28
59
1.54
97
PDISCO_ROBtwo views10.27
100
2.79
93
13.30
91
10.58
130
9.96
131
22.78
114
5.99
84
20.01
109
28.57
114
11.88
72
14.49
64
20.54
104
4.92
42
17.01
106
10.29
78
5.79
131
0.42
103
0.14
95
0.13
80
2.53
122
3.25
121
MFN_U_SF_DS_RVCtwo views10.36
101
5.37
117
16.61
106
2.95
72
3.02
82
24.75
117
15.44
122
14.68
80
19.73
95
19.77
103
23.97
94
16.98
84
8.40
74
17.58
108
9.23
67
1.59
119
1.75
123
0.16
99
2.63
130
0.75
97
1.85
108
SHDtwo views10.45
102
3.67
105
14.03
96
4.81
113
4.52
108
11.93
88
1.66
47
21.36
120
38.40
128
18.84
98
21.72
89
17.63
87
13.07
91
15.65
100
17.71
107
0.36
91
1.09
116
0.01
51
0.32
97
0.50
87
1.67
101
FCDSN-DCtwo views10.72
103
1.16
27
5.01
24
2.16
34
1.31
46
17.46
106
4.80
78
17.79
102
19.87
96
25.02
115
32.85
115
33.57
123
23.52
122
16.80
104
12.13
90
0.11
63
0.00
1
0.00
1
0.00
1
0.09
20
0.76
71
Dominik Hirner, Friedrich Fraundorfer: FCDSN-DC: An accurate but lightweight end-to-end trainable neural network for stereo estimation with depth completion.
SGM_RVCbinarytwo views10.77
104
1.11
26
4.66
22
2.66
62
0.41
13
21.03
112
6.87
95
20.02
110
15.54
79
25.86
117
24.97
101
33.59
124
20.51
113
20.89
115
14.78
103
0.31
88
0.26
98
0.22
109
0.26
94
0.43
78
1.01
84
Heiko Hirschmueller: Stereo processing by semiglobal matching and mutual information. TPAMI 2008, Volume 30(2), pp. 328-341
ADCLtwo views10.77
104
3.23
96
21.50
113
2.26
41
1.96
65
23.65
115
9.07
107
14.69
81
24.88
109
15.21
89
27.20
104
12.68
66
14.02
98
16.94
105
24.43
121
0.26
83
0.18
89
0.74
120
0.67
116
0.80
100
1.02
85
DPSNettwo views10.88
106
2.57
86
20.40
112
2.24
38
1.90
61
24.92
118
17.00
128
21.35
119
28.61
115
12.87
77
13.74
60
17.01
85
20.14
112
15.36
96
14.01
100
0.81
109
0.98
114
0.20
107
0.65
115
1.30
112
1.54
97
ADCMidtwo views10.97
107
4.49
113
23.66
115
2.53
57
2.52
78
11.46
85
6.31
89
15.31
84
12.85
68
23.57
111
22.99
92
17.92
89
21.04
114
19.57
113
28.05
126
0.47
99
0.43
104
1.10
124
1.32
124
1.73
118
2.10
109
FC-DCNNcopylefttwo views11.29
108
1.01
21
5.50
32
2.25
39
1.65
53
18.23
108
5.36
82
18.99
105
20.65
101
28.95
120
34.83
122
33.83
125
23.18
121
17.04
107
13.37
97
0.17
71
0.01
33
0.02
61
0.02
49
0.10
23
0.70
65
SANettwo views11.48
109
3.34
98
13.93
95
2.40
48
1.01
38
16.49
100
12.16
117
20.05
112
37.28
125
19.62
100
28.04
105
25.61
113
20.10
111
14.43
91
12.39
93
0.10
61
0.06
63
0.03
69
0.06
69
0.76
98
1.73
105
AnyNet_C32two views11.82
110
7.08
121
25.83
120
4.55
111
5.88
121
16.92
103
16.97
127
14.14
74
18.11
92
16.87
95
21.66
88
14.78
76
13.87
97
23.54
121
31.83
131
0.24
79
0.28
99
0.35
114
0.53
112
1.35
113
1.71
104
MeshStereopermissivetwo views11.94
111
1.88
57
5.24
28
2.11
33
1.44
49
20.09
110
4.95
80
21.74
122
16.75
83
33.10
125
35.07
123
39.28
131
22.72
119
19.45
112
13.59
98
0.19
73
0.06
63
0.02
61
0.00
1
0.46
82
0.65
56
C. Zhang, Z. Li, Y. Cheng, R. Cai, H. Chao, Y. Rui: MeshStereo: A Global Stereo Model with Mesh Alignment Regularization for View Interpolation. ICCV 2015
MFN_U_SF_RVCtwo views13.54
112
4.69
115
28.43
122
4.24
105
2.50
77
24.26
116
5.99
84
28.23
128
20.18
100
24.06
114
29.38
108
30.35
120
16.92
106
20.98
117
21.39
111
1.35
117
1.13
117
1.02
122
1.41
125
1.88
120
2.50
116
ADCStwo views13.76
113
6.46
119
30.74
124
3.62
94
2.93
80
14.28
96
11.22
113
20.02
110
34.42
121
23.20
108
25.42
102
19.51
100
19.62
109
24.54
122
33.88
133
0.27
85
0.23
96
0.37
116
0.19
84
1.47
114
2.84
118
MFMNet_retwo views14.09
114
9.67
127
20.31
111
10.40
129
7.62
126
21.38
113
17.49
130
21.36
120
31.95
118
23.70
112
29.22
107
18.87
96
27.41
127
14.27
89
11.16
86
2.85
126
1.51
122
0.41
118
0.28
96
4.88
128
7.10
129
EDNetEfficienttwo views14.14
115
6.71
120
56.12
131
2.99
75
3.58
94
5.38
37
8.49
105
9.35
18
33.24
119
40.73
129
30.21
110
18.04
91
22.92
120
12.13
75
26.26
122
0.16
68
0.24
97
0.07
82
1.15
120
2.58
123
2.39
114
LSMtwo views14.85
116
7.41
123
36.29
127
7.76
127
43.77
135
11.24
84
12.14
116
16.73
95
23.72
106
23.73
113
32.92
116
18.89
97
13.31
93
16.60
103
12.37
92
0.26
83
0.86
111
0.11
88
0.43
106
1.24
109
17.26
135
SAMSARAtwo views15.67
117
3.89
107
14.48
99
13.77
132
7.92
128
38.69
129
73.90
137
21.22
118
24.66
108
17.04
96
13.86
61
30.01
119
13.07
91
18.07
109
17.48
106
0.23
78
1.46
120
0.07
82
0.40
103
0.99
104
2.15
110
SPS-STEREOcopylefttwo views15.83
118
7.38
122
14.92
101
11.63
131
11.77
133
25.30
119
7.12
98
25.20
125
17.06
85
32.24
124
30.34
111
31.63
122
22.59
117
25.67
123
21.97
112
7.71
133
4.50
133
3.35
131
1.56
129
6.97
132
7.72
130
K. Yamaguchi, D. McAllester, R. Urtasun: Efficient Joint Segmentation, Occlusion Labeling, Stereo and Flow Estimation. ECCV 2014
PVDtwo views16.41
119
3.98
109
16.04
103
5.38
115
4.59
112
19.44
109
7.28
100
28.88
130
49.27
134
31.53
123
43.72
131
30.36
121
32.17
131
26.70
125
23.19
116
0.25
80
1.48
121
0.04
75
0.59
113
0.52
88
2.87
119
AnyNet_C01two views16.99
120
12.34
131
60.47
132
4.88
114
3.11
87
31.77
125
16.79
126
18.60
104
17.96
91
18.62
97
38.50
126
24.03
110
19.00
108
30.59
128
35.18
134
0.29
87
0.56
108
0.45
119
0.48
109
2.70
124
3.40
123
MSC_U_SF_DS_RVCtwo views17.08
121
7.98
124
24.37
116
6.82
125
2.93
80
39.68
130
16.11
123
25.46
126
34.93
122
34.13
126
34.20
120
29.20
118
24.94
124
23.01
119
19.39
109
2.79
125
2.87
129
1.18
126
1.53
128
6.78
131
3.31
122
NVStereoNet_ROBtwo views17.22
122
8.76
125
16.47
105
7.52
126
8.42
129
13.75
95
11.07
111
24.24
124
26.67
112
31.00
121
46.49
133
34.88
127
27.06
126
22.72
118
23.61
118
6.83
132
4.32
132
6.32
133
10.37
135
4.39
127
9.57
132
Nikolai Smolyanskiy, Alexey Kamenev, Stan Birchfield: On the Importance of Stereo for Accurate Depth Estimation: An Efficient Semi-Supervised Deep Neural Network Approach. Arxiv
SGM+DAISYtwo views17.25
123
9.46
126
22.74
114
10.11
128
11.40
132
28.98
123
13.74
119
21.15
117
16.85
84
31.39
122
34.79
121
34.05
126
24.24
123
26.23
124
19.31
108
8.79
134
7.56
135
4.88
132
3.93
131
6.07
130
9.33
131
LE_ROBtwo views17.31
124
2.62
89
14.26
98
4.20
104
1.80
58
17.40
105
9.70
109
15.38
85
55.92
135
63.97
135
40.81
128
35.50
128
39.26
134
15.64
99
28.51
127
0.06
55
0.11
76
0.13
93
0.21
88
0.14
28
0.51
42
ELAS_RVCcopylefttwo views17.82
125
3.60
104
11.92
88
6.63
124
5.73
119
30.34
124
19.64
131
26.92
127
35.06
123
40.79
130
41.32
130
37.31
129
31.61
129
30.54
127
22.93
114
1.97
124
2.40
127
1.28
127
1.27
123
1.71
117
3.51
125
A. Geiger, M. Roser, R. Urtasun: Efficient large-scale stereo matching. ACCV 2010
ELAScopylefttwo views17.99
126
3.49
101
11.19
86
6.14
122
5.79
120
34.89
128
17.28
129
28.75
129
30.47
117
40.80
131
45.79
132
39.04
130
32.05
130
27.63
126
24.30
120
1.88
123
2.41
128
1.09
123
1.48
126
1.83
119
3.48
124
A. Geiger, M. Roser, R. Urtasun: Efficient large-scale stereo matching. ACCV 2010
DispFullNettwo views18.06
127
27.03
133
34.90
126
22.68
134
20.79
134
15.03
97
3.18
61
16.95
98
26.06
110
20.48
104
16.91
76
20.55
105
18.37
107
23.48
120
12.16
91
5.31
130
2.91
130
31.24
135
8.09
133
22.63
134
12.44
134
EDNetEfficientorigintwo views18.25
128
14.44
132
99.41
139
1.72
23
1.00
37
12.07
89
6.97
97
13.67
71
35.06
123
38.59
128
37.10
125
27.03
115
27.46
128
15.38
97
26.92
123
0.05
52
0.08
70
0.15
96
0.62
114
3.33
125
3.92
127
SGM-ForestMtwo views18.61
129
3.38
99
10.14
82
3.44
92
2.24
69
34.17
127
15.39
121
29.44
131
33.93
120
38.55
127
41.08
129
53.64
134
36.76
133
34.82
130
27.16
124
1.25
115
1.81
125
0.74
120
1.01
118
0.61
89
2.61
117
RTStwo views19.64
130
10.42
128
87.01
134
5.88
117
6.82
122
41.73
131
16.69
124
16.80
96
41.90
131
23.44
109
32.96
117
13.79
68
21.49
115
38.85
131
30.84
128
0.46
97
0.70
109
0.00
1
0.09
73
1.13
106
1.70
102
RTSAtwo views19.64
130
10.42
128
87.01
134
5.88
117
6.82
122
41.73
131
16.69
124
16.80
96
41.90
131
23.44
109
32.96
117
13.79
68
21.49
115
38.85
131
30.84
128
0.46
97
0.70
109
0.00
1
0.09
73
1.13
106
1.70
102
MANEtwo views20.98
132
3.23
96
8.81
71
5.96
119
6.96
124
33.07
126
13.20
118
35.43
133
38.86
129
44.72
132
52.41
136
53.20
133
41.84
135
34.74
129
27.45
125
1.87
122
2.27
126
1.54
129
10.18
134
0.69
94
3.11
120
BEATNet-Init1two views24.70
133
11.33
130
44.28
128
5.48
116
3.94
99
47.60
134
22.33
132
34.94
132
40.01
130
48.11
134
49.98
134
59.86
135
43.17
136
39.48
133
31.77
130
1.37
118
1.76
124
1.17
125
1.17
121
2.44
121
3.85
126
MADNet+two views27.96
134
35.32
134
91.41
136
21.04
133
8.86
130
49.25
135
47.41
134
36.98
134
37.84
127
21.00
106
30.83
112
25.41
112
34.95
132
46.51
135
51.54
135
3.04
127
3.33
131
1.48
128
1.05
119
5.85
129
6.05
128
PWCKtwo views31.62
135
45.59
135
49.67
130
30.49
135
7.80
127
42.91
133
29.73
133
45.45
135
44.44
133
47.41
133
36.92
124
47.86
132
26.79
125
41.90
134
33.36
132
22.77
135
4.56
134
29.89
134
5.13
132
29.43
135
10.36
133
DPSimNet_ROBtwo views54.45
136
65.73
136
47.00
129
54.49
136
46.37
136
55.34
136
56.47
135
56.18
136
59.04
136
64.57
136
51.76
135
64.01
136
54.49
137
47.54
136
67.21
136
63.21
136
27.86
136
58.04
136
51.28
136
46.61
136
51.87
136
MADNet++two views82.81
137
81.73
137
74.64
133
87.71
137
82.67
137
93.35
137
70.27
136
86.39
137
82.88
137
93.51
137
86.62
137
86.40
137
81.37
138
88.21
137
88.63
137
86.59
137
84.23
137
72.14
137
68.69
137
78.88
137
81.36
137
MEDIAN_ROBtwo views98.44
138
99.70
138
99.34
138
97.13
138
97.06
138
96.94
138
95.91
139
97.71
138
97.33
138
98.78
140
98.91
138
99.19
138
98.13
139
96.94
138
96.99
138
99.97
140
99.18
138
100.00
138
99.99
138
99.70
138
99.83
138
AVERAGE_ROBtwo views99.63
139
99.96
139
98.88
137
100.00
143
100.00
139
98.16
139
95.64
138
100.00
141
100.00
139
100.00
141
100.00
139
100.00
141
100.00
140
100.00
141
99.99
139
100.00
142
100.00
139
100.00
138
100.00
139
100.00
141
100.00
144
DGTPSM_ROBtwo views99.90
140
100.00
140
99.99
140
99.99
141
100.00
139
100.00
140
100.00
140
99.98
139
100.00
139
98.37
138
100.00
139
99.81
139
100.00
140
99.97
139
99.99
139
99.99
141
100.00
139
100.00
138
100.00
139
100.00
141
99.99
141
DPSMNet_ROBtwo views99.90
140
100.00
140
99.99
140
99.99
141
100.00
139
100.00
140
100.00
140
99.98
139
100.00
139
98.37
138
100.00
139
99.81
139
100.00
140
99.97
139
99.99
139
100.00
142
100.00
139
100.00
138
100.00
139
100.00
141
99.99
141
DPSM_ROBtwo views99.94
142
100.00
140
100.00
142
99.74
139
100.00
139
100.00
140
100.00
140
100.00
141
100.00
139
100.00
141
100.00
139
100.00
141
100.00
140
100.00
141
100.00
142
99.08
138
100.00
139
100.00
138
100.00
139
99.98
139
99.94
139
DPSMtwo views99.94
142
100.00
140
100.00
142
99.74
139
100.00
139
100.00
140
100.00
140
100.00
141
100.00
139
100.00
141
100.00
139
100.00
141
100.00
140
100.00
141
100.00
142
99.08
138
100.00
139
100.00
138
100.00
139
99.98
139
99.94
139
LSM0two views100.00
144
100.00
140
100.00
142
100.00
143
100.00
139
100.00
140
100.00
140
100.00
141
100.00
139
100.00
141
100.00
139
100.00
141
100.00
140
100.00
141
100.00
142
100.00
142
100.00
139
100.00
138
100.00
139
100.00
141
99.99
141
MSMDNettwo views1.28
7