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 views3.58
1
1.05
3
3.35
1
2.17
2
0.59
1
5.61
1
0.40
1
10.50
1
13.34
5
6.93
1
10.44
1
5.29
1
3.47
1
5.82
1
2.62
1
0.01
2
0.01
2
0.02
9
0.01
11
0.04
1
0.01
1
PMTNettwo views5.36
2
0.97
1
3.42
2
1.97
1
1.30
2
8.96
7
0.69
2
12.33
2
12.16
4
8.50
2
11.35
2
6.09
2
4.50
2
9.34
5
2.70
2
22.71
130
0.01
2
0.04
13
0.02
15
0.06
2
0.01
1
Gwc-CoAtRStwo views5.54
3
1.22
4
8.40
23
4.28
3
1.36
6
6.93
3
1.31
3
15.48
4
15.03
8
12.62
3
13.32
3
12.74
7
5.71
6
8.61
4
2.98
3
0.01
2
0.22
43
0.00
1
0.29
50
0.28
11
0.07
3
DPM-Stereotwo views6.74
4
1.93
21
10.02
33
4.28
3
1.64
8
13.54
23
1.52
5
15.16
3
6.63
1
14.30
5
26.81
19
10.90
4
5.19
3
14.78
19
7.56
9
0.00
1
0.06
23
0.05
15
0.00
1
0.17
9
0.17
14
R-Stereo Traintwo views7.04
5
1.34
8
7.89
18
6.23
23
2.30
14
7.51
4
5.54
26
23.96
16
6.94
2
16.26
15
33.57
52
12.04
5
5.61
4
7.67
2
3.74
4
0.01
2
0.05
17
0.01
5
0.00
1
0.10
3
0.12
6
RAFT-Stereopermissivetwo views7.04
5
1.34
8
7.89
18
6.23
23
2.30
14
7.51
4
5.54
26
23.96
16
6.94
2
16.26
15
33.57
52
12.04
5
5.61
4
7.67
2
3.74
4
0.01
2
0.05
17
0.01
5
0.00
1
0.10
3
0.12
6
Lahav Lipson, Zachary Teed, and Jia Deng: RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matching. 3DV
HITNettwo views7.83
7
2.93
38
8.39
22
4.76
5
1.48
7
12.75
16
4.64
18
20.97
6
14.37
7
15.14
11
22.15
6
14.57
10
10.62
11
14.38
17
8.14
13
0.04
8
0.01
2
0.48
48
0.02
15
0.57
25
0.10
5
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
FENettwo views7.94
8
1.01
2
6.72
11
5.87
16
2.07
10
11.11
11
2.03
6
21.39
8
20.81
15
17.14
22
22.65
7
14.43
9
11.54
15
12.97
13
7.89
11
0.15
20
0.24
46
0.03
11
0.02
15
0.14
6
0.62
32
DN-CSS_ROBtwo views8.17
9
2.59
33
14.14
50
6.11
21
3.22
25
6.31
2
3.53
8
22.92
15
20.59
14
16.66
19
30.05
31
9.31
3
7.95
9
13.49
14
4.78
6
0.72
58
0.00
1
0.45
46
0.00
1
0.54
23
0.09
4
DMCAtwo views8.51
10
1.90
20
9.96
32
5.16
6
4.14
34
10.29
10
4.46
17
25.52
24
19.55
12
16.33
17
20.69
5
17.59
14
11.26
13
10.42
6
11.36
32
0.50
48
0.08
29
0.35
38
0.05
25
0.36
13
0.31
20
MLCVtwo views9.32
11
2.22
30
12.55
44
5.22
8
1.32
3
13.79
24
1.35
4
21.14
7
23.65
22
21.91
39
31.37
39
17.14
12
7.57
8
16.00
29
10.18
22
0.22
25
0.01
2
0.03
11
0.02
15
0.40
14
0.29
19
BEATNet_4xtwo views9.37
12
4.85
65
12.05
41
5.48
10
1.35
5
13.15
20
6.15
36
25.06
22
16.74
9
16.18
14
24.65
11
17.29
13
12.51
18
16.91
34
10.92
27
0.38
36
0.18
40
0.81
64
0.05
25
1.95
64
0.65
35
GANet-RSSMtwo views9.48
13
1.32
7
6.45
9
6.72
31
5.51
57
12.78
17
6.03
33
26.55
30
18.96
10
14.92
6
27.95
25
25.41
41
14.69
37
13.52
15
7.60
10
0.10
15
0.10
34
0.16
28
0.01
11
0.51
21
0.39
23
PSMNet-RSSMtwo views9.71
14
1.73
16
6.09
7
7.47
39
4.27
37
10.06
8
6.00
31
29.14
41
21.74
17
15.76
12
32.88
49
22.63
23
13.18
19
12.94
12
9.12
19
0.09
14
0.02
9
0.07
20
0.10
36
0.70
31
0.24
16
ccstwo views9.74
15
1.25
6
7.47
16
5.77
13
2.16
12
7.74
6
3.54
9
33.83
69
35.91
62
13.42
4
23.80
9
22.01
20
11.97
16
11.51
7
9.09
18
0.41
40
1.75
95
0.17
29
0.17
41
0.63
27
2.12
72
CFNet-ftpermissivetwo views9.87
16
2.18
28
5.49
4
7.45
37
5.25
50
15.76
36
5.98
29
21.89
10
21.87
18
14.96
7
30.02
29
25.68
42
13.68
23
12.46
9
10.94
28
0.08
12
0.06
23
1.83
85
0.22
43
1.09
42
0.43
26
CFNet_RVCtwo views9.87
16
2.18
28
5.49
4
7.45
37
5.25
50
15.76
36
5.98
29
21.89
10
21.87
18
14.96
7
30.02
29
25.68
42
13.68
23
12.46
9
10.94
28
0.08
12
0.06
23
1.83
85
0.22
43
1.09
42
0.43
26
cf-rtwo views10.06
18
1.60
11
7.99
20
5.82
14
5.70
62
12.92
19
4.78
19
24.18
19
27.47
32
16.91
21
31.02
37
25.12
39
14.05
29
11.63
8
11.01
30
0.03
7
0.03
12
0.05
15
0.03
22
0.43
15
0.39
23
AdaStereotwo views10.22
19
3.63
49
9.14
29
9.15
56
3.24
26
15.79
38
5.39
24
31.94
56
25.42
25
18.39
25
26.47
15
19.44
17
9.50
10
16.60
33
8.59
15
0.44
45
0.05
17
0.40
44
0.00
1
0.57
25
0.16
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.
iResNettwo views10.26
20
3.10
44
15.72
59
7.35
35
2.13
11
13.86
25
7.07
45
22.80
14
28.01
35
20.10
31
30.40
33
16.79
11
11.06
12
14.46
18
10.29
25
0.39
39
0.05
17
0.00
1
0.07
34
0.73
34
0.73
38
ACVNettwo views10.36
21
1.68
13
5.49
4
6.01
18
2.75
19
12.78
17
3.56
10
21.89
10
34.81
58
14.96
7
30.78
35
24.72
36
13.88
26
15.81
25
17.20
50
0.13
17
0.01
2
0.06
18
0.06
31
0.51
21
0.15
11
ccs_robtwo views10.52
22
2.12
26
7.47
16
6.02
20
3.07
22
19.32
52
4.16
13
25.06
22
27.77
34
20.21
32
27.93
24
30.56
56
13.32
21
14.95
21
7.25
8
0.04
8
0.05
17
0.21
30
0.05
25
0.55
24
0.27
18
GwcNet-RSSMtwo views10.65
23
1.94
22
8.57
25
6.89
33
5.26
52
12.26
15
5.21
22
25.80
27
31.50
45
16.67
20
33.52
51
23.70
31
13.88
26
12.84
11
13.67
37
0.05
10
0.10
34
0.14
26
0.02
15
0.65
30
0.39
23
CFNettwo views10.67
24
2.33
31
8.90
26
6.66
29
4.09
33
16.05
39
4.26
15
30.44
48
31.01
44
18.53
26
26.48
16
23.69
30
13.68
23
15.27
23
10.59
26
0.05
10
0.02
9
0.44
45
0.05
25
0.73
34
0.23
15
DeepPruner_ROBtwo views10.77
25
4.56
60
13.13
46
6.37
26
4.28
38
10.14
9
6.86
41
36.42
85
19.59
13
17.87
23
27.37
22
23.19
26
12.26
17
19.09
44
10.23
24
1.05
72
0.48
69
0.23
32
0.15
39
0.89
39
1.17
51
HSM-Net_RVCpermissivetwo views10.88
26
1.23
5
5.46
3
5.57
11
2.63
17
20.12
56
6.06
34
27.67
34
28.35
36
20.63
35
25.51
12
34.01
71
14.38
36
16.18
31
9.29
20
0.11
16
0.07
27
0.01
5
0.00
1
0.13
5
0.14
10
Gengshan Yang, Joshua Manela, Michael Happold, and Deva Ramanan: Hierarchical Deep Stereo Matching on High-resolution Images. CVPR 2019
NOSS_ROBtwo views10.99
27
3.81
53
6.88
13
6.69
30
2.30
14
16.06
40
10.94
62
30.06
46
32.41
50
16.10
13
18.72
4
22.78
24
14.21
33
17.52
36
8.62
16
2.35
90
1.84
97
3.32
105
1.57
89
2.06
68
1.52
59
ac_64two views11.10
28
1.49
10
9.88
31
8.24
52
5.45
55
19.12
50
6.46
38
24.10
18
14.34
6
22.80
43
23.87
10
33.28
69
15.54
41
18.87
42
17.26
52
0.28
30
0.12
37
0.02
9
0.02
15
0.26
10
0.63
33
HSMtwo views11.33
29
1.80
17
7.04
15
6.13
22
3.99
32
16.16
43
6.92
43
29.69
44
19.39
11
22.95
44
26.05
14
41.08
96
15.79
44
20.25
50
8.89
17
0.02
6
0.03
12
0.01
5
0.00
1
0.15
7
0.34
22
DMCA-RVCcopylefttwo views11.41
30
3.32
45
11.68
39
10.12
71
4.81
44
11.92
14
4.11
12
24.67
20
23.66
23
20.06
30
28.51
26
26.71
46
24.96
75
15.95
28
14.82
40
0.41
40
0.34
55
0.56
53
0.16
40
0.72
33
0.69
36
RASNettwo views11.59
31
1.71
15
10.36
34
6.30
25
5.66
60
15.48
35
6.12
35
34.50
75
20.86
16
16.61
18
32.51
46
22.04
22
23.00
67
19.44
45
16.15
47
0.71
56
0.03
12
0.05
15
0.01
11
0.15
7
0.13
9
iResNet_ROBtwo views11.71
32
1.96
23
10.77
35
6.01
18
3.68
28
15.13
32
2.34
7
31.72
55
37.58
69
26.16
61
32.77
47
25.88
44
15.34
40
16.35
32
7.94
12
0.19
23
0.01
2
0.00
1
0.00
1
0.31
12
0.12
6
FADNet-RVC-Resampletwo views11.72
33
4.35
58
30.33
94
9.46
61
3.90
30
14.11
26
7.17
46
25.78
25
23.31
21
23.27
47
29.52
28
19.50
18
15.00
39
15.38
24
8.33
14
0.53
49
0.50
70
0.49
49
0.50
63
1.38
55
1.67
63
CBMV_ROBtwo views11.77
34
2.89
36
6.92
14
5.20
7
2.88
21
14.12
27
4.79
20
26.69
31
29.84
38
26.17
62
31.92
43
23.61
29
20.74
63
20.20
49
10.18
22
1.78
84
1.80
96
2.17
91
1.26
85
1.61
57
0.55
31
DSFCAtwo views11.85
35
2.01
24
12.93
45
5.40
9
6.87
73
15.36
33
10.45
59
30.68
50
31.84
47
25.99
60
23.52
8
22.03
21
14.05
29
20.28
52
12.46
33
0.57
51
0.41
60
0.28
35
0.38
57
1.02
40
0.47
29
FADNet_RVCtwo views11.94
36
5.17
69
39.74
107
6.44
27
3.46
27
11.42
12
4.17
14
29.16
42
26.89
29
17.97
24
25.96
13
13.41
8
15.67
42
15.84
27
13.54
36
0.90
65
0.75
81
0.55
52
1.20
83
3.65
88
2.85
80
iResNetv2_ROBtwo views12.00
37
2.91
37
13.46
47
5.82
14
3.73
29
13.15
20
6.45
37
34.15
70
36.02
64
25.25
57
33.81
55
29.79
54
14.24
34
13.61
16
6.10
7
0.21
24
0.02
9
0.12
23
0.00
1
0.82
37
0.25
17
hitnet-ftcopylefttwo views12.04
38
2.11
25
6.14
8
7.27
34
3.90
30
25.98
80
3.56
10
22.56
13
23.74
24
20.55
34
33.81
55
31.59
60
18.67
55
19.47
46
17.20
50
0.42
44
1.36
91
0.35
38
0.07
34
1.11
45
0.86
40
NLCA_NET_v2_RVCtwo views12.05
39
3.43
47
16.25
64
7.62
42
5.58
58
14.57
31
6.00
31
32.98
66
27.41
31
21.69
38
31.40
40
25.21
40
13.47
22
14.96
22
15.60
43
0.84
60
0.33
53
0.35
38
0.35
54
1.29
51
1.78
65
Zhibo Rao, Mingyi He, Yuchao Dai, Zhidong Zhu, Bo Li, and Renjie He.: NLCA-Net: A non-local context attention network for stereo matching.
CC-Net-ROBtwo views12.13
40
3.51
48
15.82
60
7.52
40
5.50
56
14.30
29
6.60
40
32.79
63
27.59
33
21.97
40
32.14
45
24.76
38
14.02
28
14.82
20
16.06
44
0.85
62
0.35
56
0.34
37
0.35
54
1.30
53
1.94
68
acv_fttwo views12.16
41
1.68
13
11.97
40
7.70
44
6.20
64
19.61
54
5.52
25
25.78
25
34.81
58
25.62
59
30.78
35
24.72
36
13.20
20
15.81
25
18.80
53
0.13
17
0.01
2
0.06
18
0.06
31
0.63
27
0.15
11
HGLStereotwo views12.72
42
2.66
34
12.34
42
9.93
67
7.55
82
16.16
43
7.45
47
31.48
54
27.01
30
20.01
29
35.31
60
26.98
48
18.90
56
17.87
39
18.90
54
0.33
33
0.16
39
0.07
20
0.05
25
0.48
16
0.70
37
SGM-Foresttwo views12.92
43
1.87
18
6.61
10
5.68
12
2.05
9
23.19
70
11.30
65
34.24
71
30.77
43
26.54
64
31.99
44
29.98
55
15.74
43
21.17
57
13.36
35
1.05
72
0.33
53
0.84
66
0.01
11
0.71
32
0.97
44
Johannes L. Schönberger, Sudipta Sinha, Marc Pollefeys: Learning to Fuse Proposals from Multiple Scanline Optimizations in Semi-Global Matching. ECCV 2018
STTStereotwo views12.97
44
4.88
66
26.57
88
7.54
41
5.41
54
13.43
22
6.49
39
30.53
49
26.35
28
25.14
56
31.43
41
32.00
61
14.12
31
16.01
30
14.83
41
0.29
31
0.31
51
0.67
58
1.08
79
1.21
48
1.11
48
TDLMtwo views12.98
45
3.78
52
9.77
30
10.75
76
4.94
46
16.10
41
14.08
76
35.05
78
31.87
48
22.36
41
26.51
18
26.09
45
14.97
38
24.73
69
13.11
34
1.05
72
0.05
17
1.21
80
0.27
48
1.83
60
1.00
46
AANet_RVCtwo views13.16
46
5.34
71
10.83
37
8.20
51
4.44
39
14.13
28
9.46
52
28.78
40
37.67
70
23.44
49
37.47
67
23.48
27
15.83
45
22.39
59
16.63
48
1.67
82
0.86
85
0.24
33
0.02
15
0.63
27
1.60
61
CVANet_RVCtwo views13.24
47
3.39
46
8.43
24
8.42
53
5.04
48
18.52
48
11.72
67
32.03
58
33.85
53
23.17
46
32.77
47
29.78
53
16.51
46
23.20
61
11.28
31
0.94
66
0.04
15
1.19
79
0.47
61
3.13
81
1.01
47
FADNet-RVCtwo views13.27
48
11.57
105
39.71
106
7.94
49
4.50
40
15.41
34
6.95
44
27.96
35
22.98
20
20.37
33
30.73
34
26.86
47
11.31
14
20.50
55
14.28
39
0.14
19
0.14
38
0.13
25
0.18
42
2.83
75
0.86
40
DLCB_ROBtwo views13.35
49
3.00
41
9.12
27
9.43
59
5.68
61
21.80
64
10.12
56
29.19
43
29.92
39
27.71
68
31.45
42
32.37
62
19.50
57
19.02
43
16.73
49
0.22
25
0.04
15
0.38
43
0.06
31
0.49
19
0.85
39
StereoDRNet-Refinedtwo views13.36
50
2.80
35
10.82
36
7.94
49
3.10
23
18.95
49
4.45
16
28.47
39
29.65
37
29.11
72
41.47
86
24.44
34
17.64
51
23.34
62
21.66
60
0.16
21
0.07
27
0.53
51
0.32
52
0.77
36
1.64
62
Rohan Chabra, Julian Straub, Chris Sweeney, Richard Newcombe, Henry Fuchs: StereoDRNet. CVPR
CBMVpermissivetwo views13.69
51
3.63
49
8.02
21
5.93
17
2.69
18
22.57
67
12.44
69
29.81
45
31.57
46
31.20
81
33.79
54
31.04
57
17.41
50
25.28
72
14.11
38
0.71
56
0.60
74
0.60
55
0.11
37
1.11
45
1.13
50
Konstantinos Batsos, Changjiang Cai, Philippos Mordohai: CBMV: A Coalesced Bidirectional Matching Volume for Disparity Estimation. Computer Vision and Pattern Recognition (CVPR) 2018
FADNettwo views14.18
52
10.92
102
37.56
102
7.79
47
6.64
70
17.48
45
6.91
42
34.73
76
34.03
54
18.85
27
27.02
20
28.26
49
14.24
34
17.77
38
10.09
21
0.41
40
0.67
77
0.37
41
0.37
56
8.13
111
1.34
53
pmcnntwo views14.59
53
3.68
51
19.78
77
6.83
32
4.51
41
21.27
61
14.67
79
27.37
32
30.54
40
27.09
67
40.19
81
38.71
87
18.53
54
17.70
37
19.27
55
0.84
60
0.09
33
0.00
1
0.00
1
0.48
16
0.31
20
NVstereo2Dtwo views15.15
54
2.97
39
16.28
65
9.23
57
7.56
83
30.03
90
9.53
53
42.80
111
42.72
84
15.11
10
27.03
21
23.16
25
16.93
48
24.28
66
15.45
42
3.07
99
0.43
65
1.44
81
0.61
66
6.74
103
7.60
113
StereoDRNettwo views15.30
55
4.47
59
14.35
54
10.71
75
8.69
92
25.02
77
11.03
63
39.66
100
35.62
61
29.17
73
28.81
27
34.05
72
17.80
52
18.80
41
23.77
69
0.41
40
0.30
50
0.15
27
0.14
38
1.64
58
1.42
56
DRN-Testtwo views15.88
56
3.83
54
14.70
55
9.96
69
8.02
89
28.93
89
14.65
78
42.34
107
38.25
73
28.18
71
34.31
59
29.22
52
18.35
53
20.01
48
22.68
65
0.37
35
0.41
60
0.46
47
0.28
49
1.28
50
1.34
53
PA-Nettwo views16.29
57
6.88
83
27.34
89
9.88
66
11.12
105
21.77
63
28.15
119
32.06
59
39.49
74
19.99
28
27.91
23
23.50
28
20.59
62
20.25
50
24.04
72
0.24
27
2.26
105
0.22
31
4.59
113
1.13
47
4.48
98
Zhibo Rao, Mingyi He, Yuchao Dai, Zhelun Shen: Patch Attention Network with Generative Adversarial Model for Semi-Supervised Binocular Disparity Prediction.
DISCOtwo views16.36
58
1.88
19
16.15
62
7.78
46
4.19
35
26.92
82
10.21
58
30.31
47
44.26
91
21.35
37
33.91
57
35.80
80
22.40
66
35.65
106
33.87
104
0.18
22
0.08
29
0.04
13
0.04
24
1.71
59
0.54
30
GwcNetcopylefttwo views16.48
59
5.23
70
20.03
78
10.24
73
8.48
90
28.79
87
10.09
54
38.60
98
44.51
92
23.29
48
36.95
65
31.29
58
19.79
60
20.42
54
25.70
86
0.54
50
0.72
78
0.57
54
0.39
58
1.37
54
2.64
78
NaN_ROBtwo views16.51
60
6.51
79
16.22
63
10.53
74
4.64
43
31.76
96
17.78
94
37.00
89
43.37
87
29.68
77
31.29
38
32.67
66
20.32
61
28.00
82
16.06
44
0.38
36
0.41
60
0.32
36
0.41
59
0.87
38
1.94
68
FAT-Stereotwo views16.81
61
2.97
39
17.56
68
9.80
65
6.21
65
16.12
42
13.23
73
38.18
97
39.67
76
27.00
66
39.73
80
37.55
84
31.30
97
20.39
53
21.52
59
3.44
101
2.05
99
0.81
64
1.10
80
1.89
63
5.72
102
S-Stereotwo views16.85
62
3.95
55
20.67
80
12.08
85
8.91
94
21.68
62
12.61
70
32.20
61
44.61
93
22.37
42
38.42
74
29.15
51
28.02
86
18.58
40
22.36
64
2.88
98
3.11
113
0.87
69
1.82
93
3.58
87
9.08
119
DANettwo views16.85
62
8.00
86
26.04
87
14.56
97
7.25
79
20.59
57
5.37
23
26.22
29
26.02
27
32.17
86
36.00
61
37.20
83
29.77
91
30.75
86
24.62
77
1.14
76
1.34
89
2.21
92
0.51
64
3.18
82
4.07
93
PSMNet_ROBtwo views16.90
64
5.45
72
14.16
51
13.77
91
7.25
79
27.65
84
32.74
123
43.21
114
37.88
71
24.00
50
32.97
50
34.55
76
16.69
47
17.41
35
23.96
71
0.38
36
0.22
43
0.93
70
2.05
97
1.83
60
0.96
43
GANettwo views16.92
65
4.32
57
12.46
43
10.84
78
4.24
36
23.04
69
15.36
84
39.91
101
34.47
57
32.42
87
45.10
101
40.04
94
27.54
82
23.61
64
20.20
56
1.02
71
0.08
29
0.67
58
0.22
43
1.87
62
0.97
44
NCCL2two views17.23
66
6.37
78
14.30
53
23.45
120
7.18
78
24.16
73
17.05
92
34.39
73
25.88
26
31.55
82
38.01
70
42.09
101
26.81
78
20.94
56
22.69
66
0.45
46
0.23
45
2.63
97
1.80
92
2.02
67
2.57
76
ADCReftwo views17.30
67
6.73
81
42.01
110
10.11
70
8.78
93
26.61
81
10.45
59
31.29
52
32.75
52
31.19
80
41.62
88
18.47
16
19.78
59
24.29
67
34.06
106
0.76
59
0.42
64
2.16
90
1.79
91
1.23
49
1.56
60
XPNet_ROBtwo views17.33
68
4.84
64
15.44
58
11.15
81
6.64
70
20.88
58
16.10
86
36.26
84
39.54
75
31.80
85
41.32
85
41.30
98
23.91
68
24.67
68
27.40
92
1.07
75
0.74
79
0.79
62
0.25
46
1.10
44
1.31
52
RPtwo views17.58
69
4.92
67
15.27
57
15.44
102
11.95
111
21.05
59
11.75
68
27.64
33
45.09
96
22.99
45
42.05
90
39.39
90
27.74
84
25.54
73
22.33
63
5.45
114
0.59
73
4.33
108
1.62
90
3.53
86
2.86
81
SuperBtwo views17.60
70
6.22
77
56.40
123
7.91
48
5.07
49
19.70
55
9.03
50
24.87
21
50.15
108
26.25
63
39.13
75
17.61
15
20.84
64
22.28
58
25.31
82
0.70
55
0.31
51
1.10
75
0.71
69
16.58
127
1.86
66
ETE_ROBtwo views17.67
71
9.48
93
17.66
69
13.60
89
4.57
42
23.03
68
21.09
104
32.49
62
35.24
60
30.49
78
38.06
71
46.18
112
24.49
70
26.40
76
24.21
74
0.46
47
0.24
46
1.14
76
0.84
73
1.52
56
2.13
73
PWCDC_ROBbinarytwo views17.80
72
9.62
94
18.38
73
17.95
108
7.74
84
24.61
75
5.56
28
34.41
74
52.79
112
29.66
76
50.20
110
20.63
19
19.69
58
29.85
85
20.87
57
6.18
117
0.26
49
0.10
22
0.05
25
5.36
101
2.00
70
MDST_ROBtwo views17.92
73
1.60
11
13.68
48
13.88
92
6.44
68
43.05
116
14.86
81
42.74
110
41.66
81
43.25
109
42.85
93
28.43
50
17.28
49
29.35
84
16.06
44
0.88
63
0.11
36
0.63
57
0.47
61
0.50
20
0.64
34
Nwc_Nettwo views17.98
74
4.67
62
17.79
72
14.26
93
11.65
110
25.14
78
14.88
82
40.85
104
39.83
77
21.07
36
43.24
96
34.50
75
27.64
83
26.44
77
25.03
78
3.22
100
0.18
40
2.52
96
3.18
104
1.98
65
1.51
58
ADCP+two views18.16
75
4.61
61
32.94
97
9.31
58
10.23
101
28.87
88
11.69
66
33.43
67
36.49
65
29.28
75
40.49
82
24.18
32
24.61
72
33.89
96
35.33
107
0.25
28
0.40
59
2.00
87
1.17
82
2.16
69
1.87
67
PWC_ROBbinarytwo views18.37
76
10.55
99
25.19
86
10.21
72
6.29
67
23.75
71
4.79
20
35.93
83
48.75
101
32.56
88
44.46
100
34.68
77
24.78
73
30.99
87
25.72
87
1.38
79
0.08
29
1.57
82
0.26
47
2.18
70
3.17
85
AF-Nettwo views18.38
77
5.60
73
14.21
52
15.95
103
10.78
103
22.26
65
10.45
59
35.39
80
50.22
109
24.71
55
37.77
68
41.55
99
29.24
90
29.11
83
26.67
88
4.35
109
0.06
23
4.56
109
0.90
75
2.42
71
1.38
55
Anonymous Stereotwo views18.60
78
11.75
106
49.81
116
14.43
95
12.02
113
14.33
30
23.23
113
32.16
60
43.08
86
24.32
53
34.16
58
24.24
33
14.12
31
31.67
90
30.84
101
0.89
64
0.91
86
1.74
84
1.92
96
3.23
83
3.23
86
stereogantwo views18.62
79
3.07
42
16.31
66
13.04
88
9.99
99
35.74
106
9.09
51
38.17
96
45.27
97
24.39
54
41.24
84
39.93
92
25.46
76
31.29
89
24.33
76
1.80
85
0.91
86
2.71
99
0.75
72
5.15
100
3.70
87
edge stereotwo views18.75
80
5.12
68
17.66
69
10.77
77
7.84
86
22.41
66
11.03
63
35.23
79
42.07
83
34.62
92
42.74
92
41.28
97
35.50
105
25.14
71
24.17
73
2.42
94
2.42
107
6.35
113
1.45
87
2.89
78
3.80
88
RYNettwo views18.92
81
4.69
63
16.88
67
10.99
79
15.84
120
46.72
118
16.57
88
37.84
93
48.03
99
26.92
65
26.48
16
37.84
86
24.92
74
19.80
47
31.04
102
0.25
28
0.25
48
0.62
56
0.03
22
6.34
102
6.39
105
LALA_ROBtwo views19.19
82
7.15
85
15.93
61
12.96
87
5.30
53
28.30
86
23.21
112
42.51
108
36.69
66
33.45
91
40.81
83
50.96
118
24.53
71
27.92
79
25.66
85
0.60
52
0.44
66
2.04
88
1.22
84
2.54
73
1.47
57
aanetorigintwo views19.21
83
12.59
107
53.50
119
9.67
63
7.83
85
11.85
13
13.87
74
17.05
5
34.15
56
49.18
118
48.89
108
34.05
72
30.22
93
22.86
60
28.00
99
1.01
70
0.75
81
0.67
58
0.89
74
3.07
79
4.14
94
SGM_RVCbinarytwo views19.52
84
2.12
26
6.79
12
6.59
28
1.33
4
38.20
111
15.74
85
38.13
95
34.07
55
45.71
114
41.91
89
51.41
119
38.10
110
38.77
115
27.80
96
0.61
53
0.36
57
0.51
50
0.33
53
1.03
41
0.90
42
Heiko Hirschmueller: Stereo processing by semiglobal matching and mutual information. TPAMI 2008, Volume 30(2), pp. 328-341
RTSCtwo views19.64
85
9.36
90
31.06
95
10.99
79
6.24
66
28.10
85
10.14
57
37.50
92
58.34
125
31.74
84
36.70
64
32.50
63
21.03
65
36.00
109
37.24
112
1.23
77
0.44
66
0.12
23
0.31
51
1.98
65
1.69
64
Abc-Nettwo views19.75
86
6.08
74
23.96
83
15.05
98
16.17
121
18.42
46
21.45
105
43.31
115
41.00
79
24.30
51
38.12
72
34.98
78
26.92
79
35.46
103
25.65
83
4.29
107
2.07
100
3.24
103
8.90
122
2.86
76
2.86
81
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 views19.75
86
6.08
74
23.96
83
15.05
98
16.17
121
18.42
46
21.45
105
43.31
115
41.00
79
24.30
51
38.12
72
34.98
78
26.92
79
35.46
103
25.65
83
4.29
107
2.07
100
3.24
103
8.90
122
2.86
76
2.86
81
RGCtwo views19.90
88
13.26
108
20.24
79
18.19
109
14.06
118
21.21
60
14.33
77
34.97
77
43.60
89
27.73
69
41.49
87
39.76
91
27.98
85
34.52
99
21.37
58
3.66
104
0.54
72
8.67
121
4.16
111
4.19
92
4.02
91
DeepPrunerFtwo views19.92
89
11.02
104
44.29
111
20.74
113
17.75
124
19.19
51
22.57
107
36.86
88
49.93
104
25.48
58
36.62
63
24.69
35
23.97
69
23.44
63
21.79
61
2.68
97
1.63
94
5.70
111
3.95
110
3.38
85
2.67
79
WCMA_ROBtwo views20.02
90
4.30
56
19.72
76
9.47
62
7.00
76
32.71
101
13.99
75
31.97
57
32.48
51
41.51
105
52.00
113
44.09
105
36.14
107
32.43
92
24.29
75
6.19
118
2.79
111
1.09
74
1.10
80
3.95
89
3.13
84
psmorigintwo views20.95
91
10.24
98
35.84
99
9.75
64
7.05
77
24.59
74
10.11
55
32.85
65
32.17
49
43.09
108
54.57
118
42.79
102
36.62
108
35.99
108
27.41
93
1.58
81
2.72
110
0.79
62
1.05
78
3.26
84
6.50
106
SANettwo views20.96
92
6.60
80
29.81
91
8.83
54
3.19
24
31.27
95
20.56
103
41.86
106
56.09
121
39.30
101
43.62
97
44.95
107
31.93
99
27.83
78
23.67
68
0.94
66
0.52
71
0.99
71
0.42
60
4.72
94
2.03
71
FBW_ROBtwo views21.00
93
10.23
97
22.72
81
13.71
90
6.77
72
30.49
94
16.46
87
44.58
117
43.57
88
44.25
112
39.59
78
43.27
104
26.12
77
33.96
97
21.80
62
2.58
96
1.84
97
7.14
120
2.78
103
4.00
90
4.06
92
STTStereo_v2two views21.25
94
8.24
87
37.97
104
15.14
100
9.26
96
50.87
121
16.80
90
27.96
35
30.54
40
38.92
99
42.93
94
33.27
67
32.17
100
27.99
80
25.25
80
4.40
110
3.83
119
3.60
106
0.97
76
7.91
109
6.95
109
G-Nettwo views21.25
94
8.24
87
37.97
104
15.14
100
9.26
96
50.87
121
16.80
90
27.96
35
30.54
40
38.92
99
42.93
94
33.27
67
32.17
100
27.99
80
25.25
80
4.40
110
3.83
119
3.60
106
0.97
76
7.91
109
6.95
109
SHDtwo views21.32
96
10.03
96
30.16
93
14.30
94
8.68
91
24.15
72
8.93
48
40.82
103
61.17
127
35.79
95
44.15
99
38.73
88
30.20
92
33.88
95
31.09
103
2.00
86
0.81
83
1.17
77
1.55
88
4.08
91
4.78
99
CSANtwo views21.34
97
9.45
92
23.34
82
20.99
115
4.95
47
32.57
99
34.26
124
38.83
99
49.95
106
36.97
97
39.72
79
44.97
108
31.73
98
26.05
74
23.94
70
1.52
80
0.47
68
0.85
67
1.43
86
2.73
74
2.14
75
ADCLtwo views21.64
98
6.20
76
47.32
114
9.93
67
6.91
75
38.69
113
19.97
101
31.26
51
54.04
113
27.89
70
47.97
106
31.35
59
30.66
95
33.18
94
36.62
109
0.99
68
0.92
88
2.63
97
1.82
93
2.45
72
2.13
73
ADCPNettwo views21.93
99
9.38
91
57.92
124
11.76
84
6.88
74
36.03
107
18.44
96
32.80
64
35.93
63
32.91
90
43.80
98
39.21
89
26.99
81
31.14
88
36.94
111
2.05
87
2.67
109
2.30
94
3.45
107
4.22
93
3.82
89
MeshStereopermissivetwo views22.27
100
6.87
82
11.15
38
7.69
43
4.87
45
37.70
110
13.06
71
41.64
105
36.95
67
50.92
120
53.41
115
58.12
123
41.93
118
37.73
112
27.62
95
2.52
95
2.37
106
2.99
101
2.07
98
3.09
80
2.60
77
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
ADCMidtwo views22.76
101
10.75
100
41.73
109
11.23
83
7.97
88
27.26
83
19.18
100
35.44
81
38.02
72
40.23
102
46.38
104
36.56
82
42.70
119
38.20
113
40.75
119
1.77
83
1.43
92
3.05
102
3.81
109
4.96
96
3.82
89
MSMD_ROBtwo views22.90
102
9.64
95
14.76
56
17.46
107
9.88
98
34.94
104
18.83
97
33.67
68
37.47
68
40.97
104
59.80
123
45.93
110
38.34
111
31.74
91
23.24
67
6.22
119
3.58
117
8.78
122
11.03
124
6.76
104
5.06
101
LE_ROBtwo views22.93
103
3.07
42
14.02
49
7.70
44
2.86
20
31.99
98
17.35
93
28.10
38
67.19
134
70.51
134
55.61
121
49.07
115
47.62
123
24.76
70
36.49
108
0.35
34
0.20
42
0.27
34
0.55
65
0.48
16
0.44
28
AnyNet_C32two views23.37
104
13.28
109
40.29
108
12.36
86
11.44
107
30.39
91
29.15
120
31.36
53
44.90
95
35.36
93
48.86
107
33.61
70
34.20
103
43.15
119
41.62
122
1.27
78
1.35
90
1.18
78
2.58
101
4.77
95
6.27
104
SGM-ForestMtwo views23.72
105
2.44
32
9.13
28
7.43
36
2.25
13
44.80
117
19.11
99
44.90
118
49.97
107
50.74
119
51.18
111
62.06
130
45.83
122
44.96
120
33.88
105
1.00
69
0.84
84
0.78
61
0.62
67
1.29
51
1.11
48
MFN_U_SF_DS_RVCtwo views23.92
106
13.60
110
30.09
92
25.46
122
10.02
100
38.51
112
20.40
102
35.77
82
43.76
90
42.15
106
47.10
105
39.99
93
28.20
88
35.96
107
27.06
90
3.64
103
3.85
121
6.66
114
16.30
128
4.97
97
4.98
100
XQCtwo views24.10
107
16.76
114
50.68
117
21.37
118
11.01
104
35.24
105
18.84
98
37.28
91
55.11
117
31.64
83
30.06
32
37.71
85
30.31
94
37.17
110
39.66
116
4.20
106
0.41
60
2.76
100
1.89
95
11.46
122
8.42
117
FC-DCNNcopylefttwo views24.37
108
10.99
103
19.02
75
18.44
110
9.16
95
36.98
108
23.07
110
40.44
102
43.05
85
46.25
116
53.58
117
50.44
116
37.86
109
35.45
102
27.57
94
6.87
120
3.37
114
6.01
112
4.88
114
7.86
108
6.05
103
DPSNettwo views24.40
109
6.97
84
33.14
98
11.16
82
6.54
69
53.33
124
43.32
130
51.28
125
59.37
126
30.89
79
39.36
76
40.35
95
34.30
104
32.57
93
25.09
79
4.65
113
1.62
93
0.37
41
0.66
68
8.51
113
4.44
97
PDISCO_ROBtwo views24.45
110
9.24
89
29.28
90
28.68
126
19.96
125
37.14
109
15.33
83
45.04
119
54.69
115
29.24
74
42.67
91
46.07
111
28.17
87
35.35
101
30.30
100
9.92
124
2.13
102
6.90
115
3.20
105
8.74
114
6.87
108
EDNetEfficienttwo views25.16
111
22.23
122
77.22
131
9.09
55
7.38
81
19.38
53
16.78
89
21.41
9
55.74
120
61.52
130
53.51
116
36.22
81
40.65
117
24.22
65
38.27
113
0.62
54
0.74
79
0.86
68
2.60
102
7.70
107
7.09
111
GANetREF_RVCpermissivetwo views25.41
112
29.28
126
24.63
85
25.18
121
6.04
63
30.44
92
35.39
126
47.84
120
50.60
110
36.88
96
36.04
62
34.11
74
31.23
96
37.46
111
27.96
98
8.58
121
2.66
108
16.29
127
4.95
115
14.37
126
8.32
116
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
FCDSN-DCtwo views25.59
113
14.91
113
18.83
74
28.55
125
15.68
119
39.29
114
18.16
95
42.83
113
41.69
82
44.43
113
51.94
112
50.82
117
39.22
114
34.59
100
27.29
91
6.05
116
3.63
118
7.01
118
6.41
116
10.35
118
10.17
123
Dominik Hirner, Friedrich Fraundorfer: FCDSN-DC: An accurate but lightweight end-to-end trainable neural network for stereo estimation with depth completion.
ADCStwo views26.20
114
13.71
111
46.90
113
14.51
96
10.34
102
30.46
93
22.72
109
42.81
112
57.61
124
42.58
107
49.37
109
41.91
100
40.43
115
41.90
118
44.58
127
2.37
91
2.19
104
2.37
95
3.29
106
7.32
105
6.52
107
MFN_U_SF_RVCtwo views28.16
115
14.14
112
53.23
118
21.11
116
7.91
87
39.97
115
13.13
72
51.65
127
46.34
98
47.41
117
53.38
114
55.32
122
40.53
116
38.47
114
39.74
117
5.67
115
4.46
122
6.98
117
8.73
121
7.47
106
7.44
112
PASMtwo views28.19
116
17.35
115
48.18
115
22.49
119
22.20
126
25.92
79
25.84
116
34.35
72
49.45
102
35.37
94
39.48
77
46.72
113
33.36
102
34.28
98
36.77
110
11.73
127
14.94
129
17.43
129
22.18
130
13.45
123
12.32
126
EDNetEfficientorigintwo views28.98
117
35.45
130
99.82
139
9.43
59
5.65
59
24.97
76
14.79
80
26.06
28
54.82
116
60.87
126
57.59
122
45.42
109
50.27
124
26.11
75
43.62
124
0.30
32
0.37
58
1.61
83
2.35
99
10.17
117
9.99
122
AnyNet_C01two views29.76
118
25.78
125
75.37
130
17.03
106
11.34
106
49.00
120
27.12
118
37.99
94
40.05
78
40.73
103
65.13
132
46.98
114
38.86
113
46.63
121
48.19
129
2.14
88
2.13
102
2.13
89
2.56
100
8.14
112
7.98
115
LSMtwo views29.98
119
19.52
120
55.11
122
19.06
111
53.75
133
32.57
99
23.17
111
42.61
109
48.57
100
45.97
115
55.04
120
42.93
103
36.02
106
35.63
105
27.03
89
2.33
89
9.19
128
4.87
110
7.71
120
10.65
120
27.91
131
RTStwo views30.15
120
23.12
123
96.33
134
16.71
104
11.63
108
58.66
128
24.28
114
36.80
86
62.63
129
43.79
110
46.10
102
32.56
64
43.90
120
51.29
125
40.92
120
2.38
92
0.66
75
1.06
72
0.74
70
5.11
98
4.34
95
RTSAtwo views30.15
120
23.12
123
96.33
134
16.71
104
11.63
108
58.66
128
24.28
114
36.80
86
62.63
129
43.79
110
46.10
102
32.56
64
43.90
120
51.29
125
40.92
120
2.38
92
0.66
75
1.06
72
0.74
70
5.11
98
4.34
95
DispFullNettwo views30.27
122
31.74
127
58.63
125
30.13
127
28.18
128
31.79
97
8.94
49
37.16
90
54.68
114
32.60
89
37.99
69
44.62
106
28.80
89
41.55
117
27.93
97
11.94
128
3.46
115
35.53
131
12.34
125
30.08
130
17.26
127
MANEtwo views31.97
123
10.75
100
17.71
71
20.94
114
13.77
116
47.80
119
26.85
117
53.00
130
55.20
118
59.29
123
64.30
130
64.93
132
56.31
131
51.83
128
38.69
115
9.61
123
4.72
123
6.91
116
19.93
129
9.20
115
7.64
114
PVDtwo views33.19
124
18.59
117
37.64
103
26.09
123
16.99
123
33.25
102
42.25
129
51.42
126
66.03
132
52.65
122
64.31
131
51.93
120
55.13
128
51.61
127
40.40
118
4.00
105
3.51
116
9.53
123
4.26
112
10.62
119
23.56
128
ELAScopylefttwo views33.68
125
18.95
118
36.71
101
20.64
112
13.61
115
56.43
125
35.14
125
50.03
124
49.90
103
60.58
125
63.83
129
58.64
126
53.36
125
50.73
123
44.44
126
11.03
125
6.58
126
10.02
124
7.43
119
13.58
125
11.94
125
A. Geiger, M. Roser, R. Urtasun: Efficient large-scale stereo matching. ACCV 2010
ELAS_RVCcopylefttwo views33.79
126
19.15
119
35.97
100
21.28
117
13.79
117
52.58
123
36.11
127
48.96
123
55.66
119
61.47
129
62.91
125
58.35
125
53.55
126
53.37
130
42.23
123
11.25
126
6.64
127
10.32
125
7.05
118
13.56
124
11.58
124
A. Geiger, M. Roser, R. Urtasun: Efficient large-scale stereo matching. ACCV 2010
SAMSARAtwo views34.43
127
18.53
116
31.91
96
55.34
133
35.34
130
75.99
134
94.71
137
47.97
121
50.84
111
37.47
98
36.99
66
53.12
121
38.55
112
39.56
116
38.65
114
3.59
102
5.40
125
2.22
93
3.72
108
9.43
116
9.29
121
BEATNet-Init1two views35.08
128
20.64
121
54.70
121
27.43
124
12.00
112
58.29
127
31.01
121
52.58
129
57.49
123
60.36
124
62.98
126
69.98
134
55.54
129
53.00
129
44.40
125
4.61
112
2.84
112
7.05
119
6.89
117
10.97
121
8.94
118
MSC_U_SF_DS_RVCtwo views37.16
129
34.69
129
53.68
120
37.47
129
12.55
114
56.80
126
22.63
108
52.02
128
63.29
131
66.20
132
63.82
128
58.12
123
53.73
127
51.21
124
46.48
128
8.79
122
5.31
124
14.69
126
15.22
127
17.27
128
9.21
120
NVStereoNet_ROBtwo views41.93
130
32.40
128
44.55
112
36.60
128
27.78
127
34.52
103
32.38
122
48.84
122
57.18
122
69.34
133
75.62
136
66.07
133
64.86
135
47.34
122
60.78
132
15.72
129
23.99
130
23.40
130
32.06
131
21.34
129
23.86
129
Nikolai Smolyanskiy, Alexey Kamenev, Stan Birchfield: On the Importance of Stereo for Accurate Depth Estimation: An Efficient Semi-Supervised Deep Neural Network Approach. Arxiv
MADNet+two views53.31
131
68.61
133
98.75
136
67.03
134
37.58
131
79.83
136
81.74
135
60.38
133
61.61
128
51.43
121
54.92
119
61.11
129
59.29
133
69.66
134
76.71
135
23.88
131
24.03
131
17.08
128
12.35
126
32.47
131
27.69
130
SGM+DAISYtwo views54.67
132
56.61
131
62.73
127
40.21
130
53.22
132
61.95
131
48.59
132
55.05
131
44.62
94
61.03
128
63.68
127
60.46
128
55.62
130
60.68
132
55.34
130
56.34
132
48.56
134
50.16
132
49.63
134
52.25
132
56.76
134
SPS-STEREOcopylefttwo views55.62
133
59.14
132
64.16
128
45.05
131
53.84
134
59.88
130
44.00
131
59.53
132
49.94
105
62.33
131
61.09
124
59.80
127
56.82
132
60.06
131
57.85
131
58.41
133
48.34
133
51.07
133
49.21
133
55.41
133
56.48
133
K. Yamaguchi, D. McAllester, R. Urtasun: Efficient Joint Segmentation, Occlusion Labeling, Stereo and Flow Estimation. ECCV 2014
PWCKtwo views65.90
134
88.63
136
79.40
132
80.32
136
29.96
129
63.85
132
39.45
128
72.07
135
71.26
135
78.74
135
68.99
133
77.82
135
61.77
134
82.81
136
62.64
134
80.17
134
39.88
132
81.90
136
47.37
132
73.54
135
37.51
132
MFMNet_retwo views72.05
135
79.64
134
64.23
129
52.73
132
84.71
136
77.93
135
68.35
133
63.96
134
66.93
133
61.00
127
71.29
134
63.84
131
68.91
136
72.42
135
61.11
133
82.03
136
74.22
136
80.34
135
74.44
135
83.62
136
89.34
136
DPSimNet_ROBtwo views74.29
136
83.27
135
59.30
126
77.19
135
65.13
135
75.14
133
74.95
134
73.99
136
76.94
136
81.99
136
75.48
135
80.18
136
71.76
137
67.39
133
84.00
136
81.87
135
51.14
135
79.99
134
76.70
136
70.17
134
79.21
135
MADNet++two views94.16
137
94.02
138
90.25
133
96.28
137
96.84
137
97.17
137
90.16
136
94.71
137
91.69
137
97.07
137
93.71
137
94.62
137
92.63
138
96.15
137
95.18
137
95.42
137
95.89
137
92.80
137
92.23
137
92.10
137
94.30
137
MEDIAN_ROBtwo views99.19
138
99.84
139
99.62
138
98.49
138
98.51
138
98.58
138
97.81
139
98.80
138
98.56
138
99.36
140
99.49
138
99.56
138
99.06
139
98.35
138
98.31
138
99.99
138
99.63
138
100.00
138
100.00
138
99.81
138
99.95
138
AVERAGE_ROBtwo views99.80
139
99.99
140
99.40
137
100.00
139
100.00
139
98.99
139
97.72
138
100.00
139
100.00
139
100.00
141
100.00
139
100.00
141
100.00
140
100.00
141
100.00
141
100.00
139
100.00
139
100.00
138
100.00
138
100.00
139
100.00
139
DPSMNet_ROBtwo views99.95
140
100.00
141
100.00
140
100.00
139
100.00
139
100.00
140
100.00
140
100.00
139
100.00
139
99.15
139
100.00
139
99.96
139
100.00
140
99.99
139
99.99
139
100.00
139
100.00
139
100.00
138
100.00
138
100.00
139
100.00
139
DGTPSM_ROBtwo views99.95
140
100.00
141
100.00
140
100.00
139
100.00
139
100.00
140
100.00
140
100.00
139
100.00
139
99.14
138
100.00
139
99.96
139
100.00
140
99.99
139
99.99
139
100.00
139
100.00
139
100.00
138
100.00
138
100.00
139
100.00
139
DPSMtwo views100.00
142
100.00
141
100.00
140
100.00
139
100.00
139
100.00
140
100.00
140
100.00
139
100.00
139
100.00
141
100.00
139
100.00
141
100.00
140
100.00
141
100.00
141
100.00
139
100.00
139
100.00
138
100.00
138
100.00
139
100.00
139
LSM0two views100.00
142
100.00
141
100.00
140
100.00
139
100.00
139
100.00
140
100.00
140
100.00
139
100.00
139
100.00
141
100.00
139
100.00
141
100.00
140
100.00
141
100.00
141
100.00
139
100.00
139
100.00
138
100.00
138
100.00
139
100.00
139
DPSM_ROBtwo views100.00
142
100.00
141
100.00
140
100.00
139
100.00
139
100.00
140
100.00
140
100.00
139
100.00
139
100.00
141
100.00
139
100.00
141
100.00
140
100.00
141
100.00
141
100.00
139
100.00
139
100.00
138
100.00
138
100.00
139
100.00
139
MSMDNettwo views7.19
7
ASD4two views90.49
137