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
HITNettwo views0.19
1
0.55
48
1.36
16
0.00
1
0.00
1
0.25
3
0.00
1
1.09
30
0.23
10
0.00
1
0.00
1
0.14
11
0.02
12
0.02
1
0.14
22
0.00
1
0.00
1
0.00
1
0.00
1
0.05
31
0.00
1
Vladimir Tankovich, Christian Häne, Sean Fanello, Yinda Zhang, Shahram Izadi, Sofien Bouaziz: HITNet: Hierarchical Iterative Tile Refinement Network for Real-time Stereo Matching.
AdaStereotwo views0.20
2
0.06
1
0.86
4
0.00
1
0.00
1
0.25
3
0.00
1
1.71
45
0.56
18
0.00
1
0.00
1
0.31
22
0.01
6
0.11
5
0.11
18
0.00
1
0.00
1
0.00
1
0.00
1
0.01
8
0.00
1
Xiao Song, Guorun Yang, Xinge Zhu, Hui Zhou, Zhe Wang, Jianping Shi: AdaStereo: A Simple and Efficient Approach for Adaptive Stereo Matching. ArXiv
MLCVtwo views0.23
3
0.46
36
1.15
12
0.00
1
0.00
1
0.45
11
0.00
1
0.57
16
0.15
4
0.09
38
0.00
1
0.27
18
0.03
22
1.09
10
0.25
28
0.00
1
0.00
1
0.00
1
0.00
1
0.05
31
0.00
1
DeepPruner_ROBtwo views0.23
3
0.37
30
1.26
13
0.00
1
0.00
1
1.11
50
0.03
43
0.20
4
0.07
1
0.00
1
0.00
1
0.39
29
0.03
22
1.12
11
0.02
5
0.00
1
0.00
1
0.00
1
0.00
1
0.04
28
0.00
1
StereoDRNet-Refinedtwo views0.24
5
0.08
6
0.68
2
0.00
1
0.00
1
0.83
32
0.00
1
0.66
19
0.61
35
0.00
1
0.00
1
0.50
35
0.02
12
1.28
13
0.08
15
0.00
1
0.00
1
0.00
1
0.00
1
0.10
44
0.04
41
Rohan Chabra, Julian Straub, Chris Sweeney, Richard Newcombe, Henry Fuchs: StereoDRNet. CVPR
MFMNet_retwo views0.24
5
0.74
61
2.31
45
0.53
63
0.00
1
0.17
1
0.01
31
0.02
1
0.59
33
0.01
21
0.03
39
0.12
8
0.00
1
0.07
2
0.01
1
0.00
1
0.00
1
0.00
1
0.00
1
0.19
57
0.01
17
iResNettwo views0.25
7
0.27
16
1.68
27
0.00
1
0.00
1
0.69
26
0.00
1
1.22
35
0.56
18
0.00
1
0.00
1
0.04
2
0.03
22
0.12
6
0.30
30
0.00
1
0.00
1
0.00
1
0.00
1
0.05
31
0.01
17
DN-CSS_ROBtwo views0.26
8
0.75
62
1.33
15
0.00
1
0.00
1
0.99
43
0.00
1
1.04
27
0.61
35
0.01
21
0.00
1
0.19
15
0.02
12
0.07
2
0.01
1
0.00
1
0.00
1
0.00
1
0.00
1
0.22
58
0.01
17
DLCB_ROBtwo views0.27
9
0.16
11
0.71
3
0.00
1
0.00
1
0.44
10
0.00
1
0.43
9
0.57
21
0.02
28
0.04
40
0.63
37
0.04
29
2.18
18
0.24
27
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
NCCL2two views0.29
10
0.50
39
1.52
22
0.00
1
0.00
1
0.81
30
0.00
1
0.45
10
0.22
9
0.00
1
0.00
1
0.02
1
0.03
22
2.14
17
0.06
11
0.00
1
0.00
1
0.02
61
0.00
1
0.01
8
0.03
34
iResNet_ROBtwo views0.30
11
0.66
56
1.41
18
0.00
1
0.00
1
0.70
27
0.00
1
0.79
23
0.57
21
0.04
35
0.01
31
0.08
4
0.01
6
1.55
14
0.03
7
0.00
1
0.00
1
0.00
1
0.00
1
0.06
35
0.00
1
LALA_ROBtwo views0.31
12
0.38
31
0.91
8
0.00
1
0.01
39
1.00
44
0.01
31
1.60
42
0.61
35
0.15
43
0.00
1
0.09
6
0.01
6
1.04
9
0.41
33
0.00
1
0.00
1
0.00
1
0.00
1
0.02
17
0.00
1
CFNet_RVCtwo views0.31
12
0.34
21
0.96
9
0.00
1
0.00
1
1.08
47
0.00
1
0.19
3
0.15
4
0.01
21
0.00
1
0.48
33
0.00
1
2.95
24
0.05
10
0.00
1
0.00
1
0.01
56
0.00
1
0.01
8
0.01
17
DISCOtwo views0.31
12
0.11
7
1.73
30
0.16
54
0.16
58
0.31
5
0.00
1
0.09
2
0.17
6
0.00
1
0.01
31
1.51
51
0.00
1
1.21
12
0.62
40
0.00
1
0.00
1
0.00
1
0.00
1
0.03
24
0.01
17
ETE_ROBtwo views0.31
12
0.34
21
0.89
5
0.00
1
0.00
1
0.97
39
0.01
31
1.45
40
0.57
21
0.02
28
0.00
1
0.17
12
0.02
12
1.75
15
0.07
13
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
NLCA_NET_v2_RVCtwo views0.35
16
0.49
38
2.00
37
0.00
1
0.02
43
0.96
37
0.00
1
0.55
15
0.18
8
0.03
31
0.00
1
0.35
26
0.10
37
2.37
19
0.01
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
Zhibo Rao, Mingyi He, Yuchao Dai, Zhidong Zhu, Bo Li, and Renjie He.: NLCA-Net: A non-local context attention network for stereo matching.
iResNetv2_ROBtwo views0.36
17
1.03
67
2.96
55
0.00
1
0.00
1
0.58
18
0.00
1
1.51
41
0.58
29
0.01
21
0.00
1
0.08
4
0.02
12
0.07
2
0.02
5
0.00
1
0.00
1
0.00
1
0.00
1
0.29
64
0.02
27
CC-Net-ROBtwo views0.37
18
0.50
39
2.09
41
0.00
1
0.02
43
0.98
40
0.00
1
0.46
11
0.17
6
0.08
36
0.00
1
0.33
23
0.11
38
2.54
21
0.01
1
0.00
1
0.00
1
0.00
1
0.01
57
0.01
8
0.00
1
XPNet_ROBtwo views0.37
18
0.34
21
1.01
10
0.01
38
0.00
1
0.96
37
0.00
1
1.11
32
0.46
12
0.00
1
0.00
1
0.17
12
0.02
12
2.73
22
0.54
37
0.00
1
0.00
1
0.00
1
0.00
1
0.01
8
0.01
17
PSMNet_ROBtwo views0.41
20
0.54
44
1.69
29
0.00
1
0.00
1
0.92
36
0.00
1
0.27
5
0.53
16
0.00
1
0.00
1
0.25
17
0.03
22
3.43
27
0.47
35
0.00
1
0.00
1
0.00
1
0.00
1
0.05
31
0.03
34
StereoDRNettwo views0.42
21
0.50
39
2.50
47
0.02
40
0.06
51
0.51
13
0.00
1
0.34
6
0.61
35
0.00
1
0.00
1
0.33
23
0.02
12
2.87
23
0.50
36
0.00
1
0.00
1
0.00
1
0.00
1
0.02
17
0.02
27
DRN-Testtwo views0.43
22
0.13
9
1.95
36
0.03
43
0.09
57
0.61
21
0.00
1
1.43
38
0.61
35
0.00
1
0.00
1
0.18
14
0.04
29
3.32
26
0.11
18
0.00
1
0.00
1
0.00
1
0.00
1
0.01
8
0.01
17
NVstereo2Dtwo views0.45
23
0.11
7
1.91
35
0.00
1
0.00
1
0.59
20
0.00
1
1.39
36
0.61
35
0.03
31
0.00
1
0.33
23
0.00
1
3.94
32
0.07
13
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
ccs_robtwo views0.46
24
0.54
44
1.68
27
0.00
1
0.00
1
0.98
40
0.01
31
0.70
21
0.57
21
0.00
1
0.00
1
0.07
3
0.01
6
4.35
36
0.17
23
0.00
1
0.00
1
0.00
1
0.00
1
0.11
47
0.03
34
ccstwo views0.46
24
0.56
49
1.46
20
0.00
1
0.00
1
0.81
30
0.01
31
0.64
18
0.57
21
0.01
21
0.00
1
0.09
6
0.01
6
4.76
38
0.21
26
0.00
1
0.00
1
0.00
1
0.00
1
0.08
41
0.01
17
TDLMtwo views0.49
26
0.39
32
1.02
11
0.00
1
0.00
1
1.08
47
0.01
31
0.36
7
0.58
29
0.14
41
0.00
1
0.27
18
0.19
41
5.46
49
0.20
25
0.00
1
0.00
1
0.00
1
0.00
1
0.13
51
0.01
17
PASMtwo views0.51
27
0.32
19
3.07
56
0.00
1
0.00
1
0.31
5
0.00
1
0.46
11
0.12
2
0.03
31
0.00
1
0.44
31
0.02
12
5.22
45
0.04
8
0.00
1
0.00
1
0.00
1
0.00
1
0.12
48
0.03
34
HSM-Net_RVCpermissivetwo views0.52
28
0.06
1
1.45
19
0.00
1
0.00
1
0.61
21
0.00
1
3.83
64
0.52
14
0.14
41
0.00
1
0.22
16
0.03
22
3.50
29
0.04
8
0.01
56
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
Gengshan Yang, Joshua Manela, Michael Happold, and Deva Ramanan: Hierarchical Deep Stereo Matching on High-resolution Images. CVPR 2019
CVANet_RVCtwo views0.53
29
0.32
19
1.26
13
0.00
1
0.00
1
0.98
40
0.00
1
0.72
22
0.55
17
0.35
45
0.12
42
0.36
28
0.14
40
5.53
51
0.08
15
0.00
1
0.00
1
0.00
1
0.00
1
0.16
55
0.00
1
RYNettwo views0.53
29
0.19
13
1.79
32
0.00
1
0.00
1
0.55
16
0.00
1
1.12
33
0.61
35
0.02
28
0.00
1
0.83
41
0.04
29
5.05
43
0.45
34
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.02
27
GANetREF_RVCpermissivetwo views0.54
31
0.70
59
2.06
39
0.00
1
0.00
1
1.14
51
0.00
1
1.61
43
0.58
29
0.00
1
0.01
31
0.29
20
0.04
29
4.21
34
0.06
11
0.00
1
0.00
1
0.00
1
0.00
1
0.03
24
0.02
27
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
GANettwo views0.55
32
0.35
26
1.40
17
0.00
1
0.00
1
0.77
29
0.08
44
0.36
7
0.57
21
0.01
21
0.00
1
0.68
38
0.11
38
5.52
50
1.07
44
0.00
1
0.00
1
0.02
61
0.00
1
0.09
42
0.03
34
PDISCO_ROBtwo views0.56
33
0.57
50
2.07
40
0.17
55
0.00
1
1.56
58
0.01
31
2.27
54
0.51
13
0.08
36
0.01
31
0.74
39
0.28
43
2.37
19
0.13
21
0.00
1
0.00
1
0.00
1
0.00
1
0.33
65
0.00
1
CFNettwo views0.56
33
0.50
39
1.58
23
0.00
1
0.00
1
1.06
45
0.01
31
1.08
29
0.57
21
0.00
1
0.01
31
0.12
8
0.01
6
5.87
57
0.19
24
0.00
1
0.00
1
0.00
1
0.00
1
0.13
51
0.02
27
CBMVpermissivetwo views0.57
35
0.22
15
1.61
25
0.00
1
0.00
1
0.84
33
0.78
58
2.58
58
0.61
35
0.38
47
0.00
1
1.25
49
0.63
47
2.07
16
0.38
31
0.00
1
0.00
1
0.00
1
0.00
1
0.04
28
0.08
51
Konstantinos Batsos, Changjiang Cai, Philippos Mordohai: CBMV: A Coalesced Bidirectional Matching Volume for Disparity Estimation. Computer Vision and Pattern Recognition (CVPR) 2018
PA-Nettwo views0.57
35
0.50
39
2.73
53
0.03
43
0.06
51
0.89
35
0.02
41
0.66
19
0.63
48
0.00
1
0.00
1
0.43
30
0.07
35
4.82
39
0.55
39
0.01
56
0.00
1
0.00
1
0.00
1
0.02
17
0.04
41
Zhibo Rao, Mingyi He, Yuchao Dai, Zhelun Shen: Patch Attention Network with Generative Adversarial Model for Semi-Supervised Binocular Disparity Prediction.
DANettwo views0.58
37
0.19
13
1.87
33
0.26
59
0.06
51
0.51
13
0.11
45
1.41
37
0.52
14
0.22
44
0.06
41
1.06
45
0.02
12
4.25
35
0.92
42
0.00
1
0.00
1
0.00
1
0.00
1
0.02
17
0.06
47
Anonymous Stereotwo views0.59
38
0.78
63
2.67
50
0.00
1
0.00
1
0.55
16
0.49
52
1.05
28
0.28
11
0.00
1
0.01
31
0.12
8
0.02
12
5.54
52
0.27
29
0.00
1
0.00
1
0.00
1
0.00
1
0.07
38
0.00
1
NOSS_ROBtwo views0.61
39
0.17
12
0.63
1
0.19
57
0.00
1
0.67
25
0.00
1
2.38
55
0.57
21
0.13
40
0.00
1
0.94
42
0.00
1
6.27
61
0.10
17
0.00
1
0.00
1
0.00
1
0.00
1
0.04
28
0.10
57
PWC_ROBbinarytwo views0.62
40
0.54
44
3.22
57
0.00
1
0.00
1
0.22
2
0.00
1
2.18
52
0.61
35
0.00
1
0.20
43
0.29
20
0.03
22
3.46
28
1.22
46
0.00
1
0.00
1
0.00
1
0.00
1
0.39
66
0.01
17
HSMtwo views0.63
41
0.36
29
0.89
5
0.00
1
0.00
1
0.52
15
0.01
31
0.98
26
0.56
18
0.00
1
0.01
31
2.27
54
0.65
48
6.18
59
0.11
18
0.00
1
0.00
1
0.00
1
0.00
1
0.01
8
0.00
1
AANet_RVCtwo views0.75
42
0.65
54
2.52
48
0.01
38
0.00
1
0.66
24
0.00
1
0.61
17
0.13
3
0.00
1
1.29
50
0.96
43
0.04
29
5.98
58
1.82
52
0.13
63
0.02
58
0.00
1
0.00
1
0.07
38
0.04
41
NaN_ROBtwo views0.80
43
0.47
37
1.63
26
0.17
55
0.19
62
0.58
18
0.38
50
2.11
49
0.80
52
0.53
48
0.33
44
0.48
33
1.13
51
4.98
42
1.94
53
0.00
1
0.02
58
0.01
56
0.05
62
0.06
35
0.14
62
FBW_ROBtwo views0.83
44
0.31
18
1.48
21
0.10
51
0.01
39
0.62
23
0.01
31
1.91
46
0.59
33
0.01
21
0.01
31
1.10
47
0.05
34
7.84
65
2.07
54
0.05
62
0.01
55
0.03
63
0.01
57
0.01
8
0.47
66
CBMV_ROBtwo views0.84
45
0.06
1
1.73
30
0.00
1
0.00
1
1.37
54
0.47
51
2.57
57
0.58
29
2.13
53
0.76
46
1.06
45
1.38
53
4.04
33
0.54
37
0.00
1
0.00
1
0.00
1
0.00
1
0.03
24
0.05
45
SGM-Foresttwo views0.84
45
0.07
5
0.90
7
0.02
40
0.07
54
1.10
49
0.68
55
2.12
51
0.61
35
1.14
51
0.78
47
1.36
50
0.98
49
5.55
53
1.31
49
0.00
1
0.00
1
0.00
1
0.00
1
0.02
17
0.06
47
Johannes L. Schönberger, Sudipta Sinha, Marc Pollefeys: Learning to Fuse Proposals from Multiple Scanline Optimizations in Semi-Global Matching. ECCV 2018
SPS-STEREOcopylefttwo views0.88
47
0.40
34
2.01
38
0.00
1
0.01
39
0.42
9
0.17
46
0.50
13
0.61
35
2.19
54
0.79
48
2.04
53
0.27
42
5.10
44
2.89
58
0.00
1
0.00
1
0.00
1
0.00
1
0.03
24
0.11
58
K. Yamaguchi, D. McAllester, R. Urtasun: Efficient Joint Segmentation, Occlusion Labeling, Stereo and Flow Estimation. ECCV 2014
PWCDC_ROBbinarytwo views0.94
48
0.91
65
2.52
48
0.00
1
0.00
1
1.15
52
0.00
1
2.11
49
1.18
55
0.00
1
8.99
62
0.35
26
0.09
36
0.64
7
0.40
32
0.00
1
0.00
1
0.00
1
0.00
1
0.50
67
0.03
34
CSANtwo views1.05
49
0.61
52
2.46
46
0.09
50
0.08
56
0.71
28
0.71
56
2.20
53
0.93
54
0.85
50
2.92
52
0.79
40
1.46
54
5.37
46
1.58
50
0.01
56
0.01
55
0.00
1
0.00
1
0.09
42
0.05
45
NVStereoNet_ROBtwo views1.08
50
0.39
32
2.15
44
0.08
49
0.02
43
0.45
11
0.02
41
0.89
24
0.67
49
0.10
39
5.85
56
1.90
52
1.11
50
6.20
60
1.14
45
0.20
65
0.00
1
0.03
63
0.07
63
0.12
48
0.13
60
Nikolai Smolyanskiy, Alexey Kamenev, Stan Birchfield: On the Importance of Stereo for Accurate Depth Estimation: An Efficient Semi-Supervised Deep Neural Network Approach. Arxiv
DispFullNettwo views1.20
51
0.63
53
3.58
60
1.00
65
0.82
65
0.41
8
0.00
1
1.44
39
0.61
35
2.55
55
0.64
45
1.10
47
8.95
66
0.84
8
1.22
46
0.00
1
0.02
58
0.00
1
0.00
1
0.24
60
0.02
27
DPSNettwo views1.25
52
0.28
17
5.34
65
0.28
61
0.07
54
1.50
55
0.00
1
3.53
63
0.69
50
0.03
31
0.00
1
0.45
32
2.10
56
5.67
54
4.61
62
0.02
59
0.04
61
0.00
1
0.02
61
0.23
59
0.09
54
MDST_ROBtwo views1.54
53
0.06
1
3.42
59
0.14
52
0.16
58
5.42
65
0.32
49
4.33
66
0.61
35
6.08
60
1.25
49
1.04
44
1.16
52
5.81
56
0.83
41
0.00
1
0.00
1
0.00
1
0.00
1
0.01
8
0.11
58
SGM_RVCbinarytwo views1.54
53
0.35
26
1.60
24
0.15
53
0.17
60
1.25
53
0.77
57
1.97
48
0.88
53
4.31
57
2.46
51
4.36
58
2.37
57
6.98
64
1.78
51
0.21
66
0.23
67
0.20
68
0.23
67
0.26
61
0.31
63
Heiko Hirschmueller: Stereo processing by semiglobal matching and mutual information. TPAMI 2008, Volume 30(2), pp. 328-341
SANettwo views1.99
55
0.71
60
3.37
58
0.03
43
0.03
48
2.85
60
0.85
59
2.42
56
8.33
66
1.55
52
4.13
54
4.29
57
3.23
59
5.39
47
2.47
56
0.00
1
0.00
1
0.00
1
0.00
1
0.06
35
0.06
47
WCMA_ROBtwo views2.01
56
0.34
21
1.89
34
0.04
46
0.00
1
1.07
46
0.60
53
1.09
30
1.21
56
5.12
58
9.87
65
8.09
62
4.48
61
3.74
31
2.55
57
0.00
1
0.00
1
0.01
56
0.00
1
0.02
17
0.13
60
MSMD_ROBtwo views2.04
57
0.40
34
2.13
42
0.00
1
0.00
1
4.88
64
0.22
47
0.93
25
1.82
60
5.71
59
7.79
61
7.39
61
3.72
60
4.62
37
1.22
46
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.02
27
pmcnntwo views2.10
58
0.59
51
4.52
64
0.27
60
0.86
66
0.34
7
0.29
48
1.69
44
0.75
51
0.36
46
7.50
60
14.39
67
0.56
45
4.90
40
4.69
63
0.00
1
0.00
1
0.00
1
0.00
1
0.16
55
0.04
41
SGM+DAISYtwo views2.38
59
0.98
66
4.37
63
0.48
62
0.33
63
2.24
59
1.94
62
1.16
34
1.47
58
6.62
61
5.89
57
5.77
59
5.09
62
5.76
55
4.39
61
0.16
64
0.15
66
0.06
65
0.14
65
0.26
61
0.40
65
PWCKtwo views2.44
60
1.65
68
7.31
66
0.22
58
0.02
43
3.12
61
4.74
66
3.06
62
3.48
63
0.72
49
4.42
55
3.40
55
1.67
55
9.06
66
3.36
60
0.39
68
0.01
55
0.00
1
0.00
1
2.19
68
0.08
51
MeshStereopermissivetwo views2.61
61
0.65
54
2.71
52
0.02
40
0.02
43
1.55
57
0.94
60
3.02
61
1.29
57
11.21
64
7.34
59
11.54
64
2.74
58
6.34
62
2.43
55
0.00
1
0.00
1
0.00
1
0.00
1
0.28
63
0.03
34
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
ELAScopylefttwo views2.75
62
0.35
26
2.67
50
0.05
47
0.04
49
6.72
66
1.83
61
2.84
60
3.38
62
8.25
63
9.14
63
3.95
56
6.15
63
3.73
30
5.70
65
0.00
1
0.04
61
0.01
56
0.01
57
0.10
44
0.08
51
A. Geiger, M. Roser, R. Urtasun: Efficient large-scale stereo matching. ACCV 2010
ELAS_RVCcopylefttwo views2.84
63
0.34
21
3.86
62
0.05
47
0.05
50
1.52
56
5.02
67
1.93
47
5.28
64
7.10
62
6.96
58
6.26
60
8.34
64
4.96
41
4.90
64
0.00
1
0.04
61
0.01
56
0.01
57
0.10
44
0.09
54
A. Geiger, M. Roser, R. Urtasun: Efficient large-scale stereo matching. ACCV 2010
FC-DCNNcopylefttwo views3.38
64
0.13
9
2.13
42
0.00
1
0.01
39
3.21
62
0.63
54
2.80
59
2.58
61
11.57
65
12.95
66
14.34
66
8.61
65
5.41
48
3.20
59
0.00
1
0.00
1
0.00
1
0.00
1
0.02
17
0.07
50
LSMtwo views4.58
65
0.88
64
17.79
69
3.73
69
38.79
69
0.85
34
2.37
63
0.54
14
1.64
59
2.88
56
3.09
53
0.62
36
0.60
46
3.16
25
1.06
43
0.00
1
0.00
1
0.00
1
0.00
1
0.12
48
13.55
69
DPSimNet_ROBtwo views7.90
66
5.46
69
13.55
68
3.66
68
5.16
68
7.26
67
9.60
69
6.11
67
7.46
65
13.33
66
9.50
64
11.03
63
11.89
67
10.08
67
12.35
66
4.36
69
5.55
69
5.35
69
5.34
69
4.96
69
5.98
68
SGM-ForestMtwo views8.89
67
0.68
57
3.61
61
0.57
64
0.18
61
14.89
69
5.99
68
9.62
68
10.59
67
20.43
67
17.16
68
39.22
69
20.29
68
20.63
69
13.19
67
0.04
60
0.14
65
0.00
1
0.07
63
0.15
53
0.36
64
LE_ROBtwo views9.96
68
0.54
44
9.62
67
1.40
67
1.24
67
3.29
63
4.64
65
4.31
65
43.01
69
51.54
69
13.34
67
14.15
65
24.60
70
6.66
63
20.28
69
0.04
60
0.05
64
0.12
67
0.20
66
0.07
38
0.09
54
MANEtwo views9.96
68
0.69
58
2.88
54
1.22
66
0.61
64
13.15
68
3.46
64
15.62
69
22.27
68
22.74
68
30.00
69
29.74
68
23.22
69
17.17
68
13.78
68
0.29
67
0.56
68
0.09
66
0.61
68
0.15
53
1.00
67
DGTPSM_ROBtwo views81.49
70
60.74
70
99.95
72
61.20
70
86.99
72
99.11
72
100.00
72
95.92
71
99.99
72
73.08
70
100.00
71
70.81
70
100.00
72
77.39
70
99.98
72
35.10
70
97.48
73
35.12
70
99.89
73
46.16
70
90.85
70
DPSMNet_ROBtwo views81.54
71
60.77
71
99.95
72
61.42
71
87.05
73
99.11
72
100.00
72
95.93
72
99.99
72
73.11
71
100.00
71
70.84
71
100.00
72
77.43
71
99.98
72
35.39
71
97.54
74
35.21
71
99.89
73
46.30
71
90.94
71
MEDIAN_ROBtwo views93.37
72
98.75
74
96.05
71
90.90
74
90.53
74
85.63
70
76.45
70
92.03
70
89.53
70
95.05
72
94.99
70
96.02
72
93.85
71
90.36
72
87.21
70
98.36
75
95.79
70
99.61
74
98.77
72
98.80
75
98.72
75
DPSMtwo views94.00
73
79.32
72
100.00
74
83.19
72
85.69
70
100.00
74
100.00
72
100.00
73
100.00
74
100.00
73
100.00
71
100.00
73
100.00
72
100.00
74
100.00
74
89.53
72
97.42
71
74.97
72
95.13
70
83.79
72
90.98
72
DPSM_ROBtwo views94.00
73
79.32
72
100.00
74
83.19
72
85.69
70
100.00
74
100.00
72
100.00
73
100.00
74
100.00
73
100.00
71
100.00
73
100.00
72
100.00
74
100.00
74
89.53
72
97.42
71
74.97
72
95.13
70
83.79
72
90.98
72
AVERAGE_ROBtwo views98.45
75
99.13
75
95.46
70
100.00
76
100.00
76
93.36
71
81.95
71
100.00
73
99.97
71
100.00
73
100.00
71
100.00
73
100.00
72
99.72
73
99.40
71
100.00
76
100.00
76
100.00
75
100.00
75
100.00
76
100.00
76
LSM0two views99.11
76
99.97
76
100.00
74
95.93
75
98.59
75
100.00
74
100.00
72
100.00
73
100.00
74
100.00
73
100.00
71
100.00
73
100.00
72
100.00
74
100.00
74
93.89
74
99.97
75
100.00
75
100.00
75
96.64
74
97.25
74
MSMDNettwo views0.34
44