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
MEDIAN_ROBtwo views0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
AVERAGE_ROBtwo views0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
PWCKtwo views0.01
3
0.02
6
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
HITNettwo views0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
Vladimir Tankovich, Christian Häne, Sean Fanello, Yinda Zhang, Shahram Izadi, Sofien Bouaziz: HITNet: Hierarchical Iterative Tile Refinement Network for Real-time Stereo Matching.
PWCDC_ROBbinarytwo views0.01
3
0.01
3
0.02
7
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
PWC_ROBbinarytwo views0.01
3
0.02
6
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
NVstereo2Dtwo views0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
0.01
3
LSM0two views0.02
8
0.02
6
0.02
7
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
LSMtwo views0.02
8
0.02
6
0.02
7
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
0.02
8
NVStereoNet_ROBtwo views0.04
10
0.04
10
0.04
10
0.04
10
0.04
10
0.04
10
0.04
10
0.04
10
0.04
10
0.04
10
0.04
10
0.04
10
0.04
10
0.04
10
0.04
10
0.04
10
0.04
10
0.04
10
0.04
10
0.04
10
0.04
10
Nikolai Smolyanskiy, Alexey Kamenev, Stan Birchfield: On the Importance of Stereo for Accurate Depth Estimation: An Efficient Semi-Supervised Deep Neural Network Approach. Arxiv
ELAScopylefttwo views0.13
11
0.16
13
0.11
14
0.15
14
0.09
11
0.18
16
0.11
14
0.18
16
0.11
14
0.17
15
0.11
14
0.18
16
0.11
14
0.18
15
0.11
14
0.14
14
0.08
11
0.14
14
0.08
11
0.14
14
0.09
11
A. Geiger, M. Roser, R. Urtasun: Efficient large-scale stereo matching. ACCV 2010
DRN-Testtwo views0.14
12
0.13
11
0.09
11
0.14
11
0.09
11
0.15
12
0.09
11
0.14
12
0.10
12
0.14
12
0.09
11
0.14
12
0.09
11
0.13
12
0.09
11
0.12
12
0.09
13
0.12
12
0.09
12
0.12
12
0.73
39
ELAS_RVCcopylefttwo views0.14
12
0.16
13
0.10
12
0.15
14
0.10
15
0.18
16
0.11
14
0.19
17
0.11
14
0.19
17
0.12
15
0.18
16
0.11
14
0.19
16
0.12
15
0.14
14
0.08
11
0.14
14
0.09
12
0.15
15
0.09
11
A. Geiger, M. Roser, R. Urtasun: Efficient large-scale stereo matching. ACCV 2010
HSMtwo views0.14
12
0.16
13
0.12
16
0.17
16
0.12
19
0.17
14
0.11
14
0.17
15
0.11
14
0.17
15
0.12
15
0.17
15
0.12
16
0.17
14
0.12
15
0.16
17
0.11
16
0.16
16
0.11
17
0.16
16
0.12
17
SGM_RVCbinarytwo views0.15
15
0.17
16
0.11
14
0.18
17
0.11
16
0.19
18
0.11
14
0.19
17
0.12
18
0.19
17
0.12
15
0.20
19
0.12
16
0.19
16
0.12
15
0.16
17
0.11
16
0.17
18
0.10
16
0.17
17
0.10
13
Heiko Hirschmueller: Stereo processing by semiglobal matching and mutual information. TPAMI 2008, Volume 30(2), pp. 328-341
StereoDRNettwo views0.15
15
0.14
12
0.10
12
0.14
11
0.09
11
0.15
12
0.09
11
0.14
12
0.09
11
0.14
12
0.10
13
0.14
12
0.09
11
0.14
13
0.09
11
0.13
13
0.10
14
0.13
13
0.09
12
0.13
13
0.82
43
LALA_ROBtwo views0.15
15
0.19
18
0.12
16
0.18
17
0.11
16
0.20
20
0.12
19
0.21
20
0.12
18
0.20
20
0.12
15
0.20
19
0.12
16
0.21
19
0.13
18
0.17
19
0.10
14
0.18
19
0.11
17
0.18
18
0.11
15
DeepPruner_ROBtwo views0.16
18
0.18
17
0.13
18
0.19
19
0.13
20
0.19
18
0.13
20
0.19
17
0.13
20
0.19
17
0.13
20
0.19
18
0.13
19
0.19
16
0.13
18
0.18
20
0.13
19
0.18
19
0.13
20
0.18
18
0.13
18
iResNetv2_ROBtwo views0.20
19
0.23
19
0.18
21
0.24
20
0.20
22
0.24
21
0.18
23
0.24
21
0.18
23
0.24
23
0.18
25
0.23
21
0.19
24
0.24
21
0.18
23
0.21
23
0.16
22
0.21
23
0.16
21
0.22
22
0.16
21
iResNettwo views0.20
19
0.23
19
0.18
21
0.24
20
0.18
21
0.24
21
0.20
24
0.24
21
0.18
23
0.23
21
0.17
21
0.23
21
0.18
23
0.23
20
0.18
23
0.22
24
0.16
22
0.21
23
0.16
21
0.21
21
0.16
21
iResNet_ROBtwo views0.28
21
0.32
23
0.24
24
0.32
25
0.25
25
0.32
26
0.24
27
0.32
27
0.24
28
0.33
27
0.24
29
0.35
28
0.24
28
0.33
26
0.24
28
0.31
28
0.24
26
0.32
30
0.24
25
0.32
25
0.24
26
DANettwo views0.30
22
0.30
21
0.30
28
0.30
24
0.30
29
0.30
25
0.30
32
0.30
26
0.30
32
0.30
26
0.30
33
0.30
26
0.30
32
0.30
25
0.30
32
0.30
26
0.30
30
0.30
28
0.30
29
0.30
24
0.30
30
DN-CSS_ROBtwo views0.31
23
0.35
24
0.28
27
0.35
27
0.28
27
0.34
27
0.27
29
0.34
29
0.25
29
0.35
29
0.27
31
0.36
31
0.26
29
0.34
28
0.28
31
0.35
30
0.28
29
0.34
31
0.28
28
0.34
26
0.27
27
MLCVtwo views0.31
23
0.35
24
0.27
26
0.35
27
0.28
27
0.35
28
0.27
29
0.35
30
0.28
31
0.36
31
0.27
31
0.35
28
0.27
31
0.35
29
0.27
30
0.34
29
0.27
28
0.34
31
0.27
27
0.34
26
0.27
27
StereoDRNet-Refinedtwo views0.33
25
0.39
28
0.26
25
0.39
31
0.26
26
0.40
32
0.26
28
0.40
34
0.26
30
0.39
34
0.26
30
0.39
34
0.26
29
0.39
32
0.26
29
0.37
33
0.25
27
0.37
34
0.25
26
0.37
29
0.37
33
Rohan Chabra, Julian Straub, Chris Sweeney, Richard Newcombe, Henry Fuchs: StereoDRNet. CVPR
ETE_ROBtwo views0.35
26
0.35
24
0.35
30
0.35
27
0.35
30
0.35
28
0.35
34
0.35
30
0.35
33
0.35
29
0.35
34
0.35
28
0.35
34
0.35
29
0.35
33
0.35
30
0.35
32
0.35
33
0.35
31
0.35
28
0.35
32
XPNet_ROBtwo views0.37
27
0.37
27
0.37
31
0.37
30
0.37
31
0.37
30
0.37
35
0.37
32
0.37
35
0.37
32
0.37
35
0.37
32
0.37
35
0.37
31
0.37
34
0.37
33
0.37
33
0.37
34
0.37
32
0.37
29
0.37
33
PASMtwo views0.39
28
3.06
52
1.36
48
1.58
48
0.09
11
0.11
11
0.11
14
0.11
11
0.11
14
0.09
11
0.09
11
0.11
11
0.09
11
0.09
11
0.09
11
0.11
11
0.11
16
0.11
11
0.09
12
0.11
11
0.11
15
AdaStereotwo views0.40
29
0.40
30
0.40
32
0.40
33
0.40
32
0.40
32
0.40
36
0.40
34
0.40
36
0.40
35
0.40
36
0.40
35
0.40
36
0.40
33
0.40
36
0.40
35
0.40
35
0.40
36
0.40
33
0.40
31
0.40
35
Xiao Song, Guorun Yang, Xinge Zhu, Hui Zhou, Zhe Wang, Jianping Shi: AdaStereo: A Simple and Efficient Approach for Adaptive Stereo Matching. ArXiv
SANettwo views0.50
30
0.50
31
0.50
33
0.50
34
0.50
33
0.50
34
0.50
38
0.50
36
0.50
37
0.50
36
0.50
37
0.50
36
0.50
37
0.50
34
0.50
38
0.50
37
0.50
36
0.50
38
0.50
34
0.50
32
0.50
36
PSMNet_ROBtwo views0.54
31
0.54
32
0.54
34
0.54
36
0.53
34
0.54
35
0.54
40
0.53
38
0.54
39
0.54
37
0.54
39
0.54
38
0.53
39
0.54
36
0.54
39
0.54
38
0.53
37
0.54
39
0.54
35
0.54
33
0.54
37
NaN_ROBtwo views0.80
32
0.80
33
0.80
36
0.80
37
0.80
37
0.80
36
0.80
42
0.80
40
0.80
41
0.80
40
0.80
41
0.80
40
0.80
41
0.80
38
0.80
41
0.80
39
0.80
39
0.80
40
0.80
38
0.80
34
0.80
40
CSANtwo views0.80
32
0.80
33
0.80
36
0.80
37
0.80
37
0.80
36
0.80
42
0.80
40
0.80
41
0.80
40
0.80
41
0.80
40
0.80
41
0.80
38
0.80
41
0.80
39
0.80
39
0.80
40
0.80
38
0.80
34
0.80
40
PDISCO_ROBtwo views0.83
34
2.80
51
3.49
57
0.14
11
0.11
16
0.17
14
0.09
11
0.15
14
0.10
12
0.15
14
0.12
15
0.16
14
0.14
20
3.30
57
0.13
18
0.14
14
2.34
56
0.16
16
0.12
19
2.62
52
0.10
13
DPSM_ROBtwo views0.92
35
0.91
35
0.91
39
0.91
39
0.92
40
0.92
39
0.92
45
0.92
43
0.91
43
0.93
44
0.91
44
0.92
42
0.90
44
0.91
40
0.92
44
0.92
41
0.91
42
0.91
43
0.92
41
0.93
37
0.93
46
DPSMtwo views0.92
35
0.91
35
0.91
39
0.91
39
0.92
40
0.92
39
0.92
45
0.92
43
0.91
43
0.93
44
0.91
44
0.92
42
0.90
44
0.91
40
0.92
44
0.92
41
0.91
42
0.91
43
0.92
41
0.93
37
0.93
46
pmcnntwo views0.92
35
0.92
37
0.92
41
0.92
41
0.92
40
0.92
39
0.92
45
0.92
43
0.92
46
0.92
43
0.92
46
0.92
42
0.92
46
0.92
42
0.92
44
0.92
41
0.92
45
0.92
45
0.92
41
0.92
36
0.92
45
DGTPSM_ROBtwo views0.93
38
1.00
38
0.92
41
0.94
42
0.96
43
0.91
38
0.92
45
0.91
42
0.96
47
0.91
42
0.96
47
0.93
45
0.92
46
0.94
43
0.92
44
0.92
41
0.92
45
0.90
42
0.92
41
0.94
39
0.91
44
DPSimNet_ROBtwo views0.97
39
1.18
43
0.81
38
1.10
46
0.91
39
1.02
45
0.82
44
1.04
49
0.91
43
1.03
49
0.86
43
1.28
50
0.82
43
1.03
47
0.89
43
1.17
49
0.81
41
1.02
49
0.82
40
1.08
43
0.81
42
CVANettwo views1.00
40
1.00
38
1.00
43
1.00
43
1.00
44
1.00
42
1.00
49
1.00
46
1.00
48
1.00
46
1.00
48
1.00
46
1.00
48
1.00
44
1.00
48
1.00
45
1.00
47
1.00
46
1.00
45
1.00
40
1.00
49
GANettwo views1.00
40
1.00
38
1.00
43
1.00
43
1.00
44
1.00
42
1.00
49
1.00
46
1.00
48
1.00
46
1.00
48
1.00
46
1.00
48
1.00
44
1.00
48
1.00
45
1.00
47
1.00
46
1.00
45
1.00
40
1.00
49
AMftwo views1.00
40
1.00
38
1.00
43
1.00
43
1.00
44
1.00
42
1.00
49
1.00
46
1.00
48
1.00
46
1.00
48
1.00
46
1.00
48
1.00
44
1.00
48
1.00
45
1.00
47
1.00
46
1.00
45
1.00
40
1.00
49
DISCOtwo views1.11
43
0.39
28
5.28
61
0.39
31
0.20
22
0.39
31
0.27
29
0.39
33
0.22
27
0.38
33
0.20
27
0.38
33
0.20
25
6.95
61
0.22
27
0.30
26
0.21
25
0.27
26
0.21
24
5.25
59
0.21
24
MSMD_ROBtwo views1.19
44
1.10
42
0.70
35
1.10
46
0.60
36
1.10
46
0.70
41
1.10
50
0.70
40
1.10
50
0.70
40
1.10
49
0.70
40
7.00
62
0.70
40
1.10
48
0.70
38
1.10
50
0.70
36
1.10
44
0.70
38
CPSMnet_ROBtwo views1.23
45
4.80
55
0.16
19
0.24
20
3.95
58
4.99
59
0.16
21
0.24
21
0.16
21
0.24
23
0.17
21
0.24
23
0.16
21
0.24
21
0.16
21
0.20
21
0.14
20
0.20
21
3.56
56
4.37
57
0.14
19
ccstwo views1.24
46
4.74
54
0.16
19
0.24
20
4.07
59
4.95
58
0.16
21
0.24
21
0.16
21
0.24
23
0.17
21
0.25
24
0.16
21
0.24
21
0.16
21
0.20
21
0.15
21
0.20
21
3.87
59
4.25
56
0.14
19
AANet_RVCtwo views1.26
47
0.31
22
4.99
60
5.93
60
0.20
22
0.25
23
0.20
24
0.25
25
0.20
25
0.23
21
0.19
26
0.25
24
0.20
25
0.25
24
0.20
25
5.66
61
4.76
62
0.26
25
0.30
29
0.24
23
0.27
27
ccs_robtwo views1.51
48
5.83
57
0.21
23
0.33
26
4.66
62
6.03
61
0.21
26
0.33
28
0.21
26
0.33
27
0.21
28
0.33
27
0.21
27
0.33
26
0.21
26
0.27
25
0.20
24
0.27
26
4.62
62
5.31
60
0.18
23
CPSMnettwo views1.55
49
6.05
59
0.34
29
0.51
35
4.32
61
5.99
60
0.34
33
0.51
37
0.35
33
0.55
38
0.17
21
0.52
37
0.31
33
0.52
35
0.38
35
0.45
36
0.33
31
0.40
36
4.02
60
4.64
58
0.34
31
GANetREF_RVCpermissivetwo views1.55
49
1.94
48
1.22
47
1.88
51
1.21
49
1.88
48
1.22
52
1.88
52
1.22
51
1.88
52
1.22
51
1.88
52
1.22
51
1.89
49
1.22
51
1.87
53
1.22
51
1.88
54
1.22
49
1.88
48
1.22
53
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
SPS-STEREOcopylefttwo views1.59
51
1.91
46
1.21
46
1.94
52
1.20
48
2.00
50
1.23
53
1.99
54
1.24
52
2.00
55
1.25
52
2.03
54
1.26
52
2.00
51
1.29
52
1.90
54
1.18
50
1.89
55
1.21
48
1.89
49
1.20
52
K. Yamaguchi, D. McAllester, R. Urtasun: Efficient Joint Segmentation, Occlusion Labeling, Stereo and Flow Estimation. ECCV 2014
DPSMNet_ROBtwo views1.60
52
1.59
44
1.70
49
1.59
49
1.59
50
1.61
47
1.61
54
1.60
51
1.60
53
1.62
51
1.59
53
1.60
51
1.60
53
1.60
48
1.59
53
1.59
51
1.60
52
1.59
52
1.59
50
1.59
46
1.59
54
FC-DCNNcopylefttwo views1.77
53
1.90
45
1.75
51
1.76
50
1.03
47
2.63
53
1.81
56
2.37
56
1.68
54
2.85
59
2.62
57
2.99
59
1.85
56
2.19
52
1.72
55
1.26
50
0.91
42
1.15
51
0.77
37
1.25
45
0.96
48
MFMNet_retwo views1.81
54
1.91
46
1.71
50
1.95
53
1.70
51
1.95
49
1.70
55
1.96
53
1.74
55
1.97
54
1.72
54
1.95
53
1.71
54
1.97
50
1.71
54
1.86
52
1.62
53
1.85
53
1.66
51
1.86
47
1.64
55
FBW_ROBtwo views2.12
55
2.46
50
1.77
52
2.49
55
1.79
52
2.38
52
1.83
57
2.46
57
1.78
56
2.48
57
1.97
55
2.40
56
1.78
55
2.42
54
1.83
56
2.31
56
1.85
54
2.38
57
1.82
52
2.35
51
1.84
56
NCCL2two views2.28
56
2.27
49
2.28
53
2.28
54
2.28
53
2.27
51
2.29
58
2.28
55
2.28
57
2.27
56
2.28
56
2.28
55
2.27
57
2.27
53
2.28
57
2.28
55
2.28
55
2.27
56
2.29
53
2.27
50
2.29
57
Anonymous Stereotwo views2.36
57
18.47
61
11.50
63
9.80
63
0.53
34
0.28
24
0.50
38
0.58
39
0.53
38
0.59
39
0.53
38
0.66
39
0.52
38
0.62
37
0.41
37
0.36
32
0.37
33
0.31
29
0.19
23
0.20
20
0.22
25
DPSNettwo views3.66
58
3.60
53
3.62
58
3.63
58
3.64
57
3.65
56
3.65
61
3.66
60
3.67
60
3.67
60
3.65
61
3.67
60
3.66
60
3.66
58
3.68
61
3.68
59
3.66
59
3.67
60
3.68
58
3.67
55
3.67
61
DispFullNettwo views4.61
59
4.84
56
3.12
56
5.00
59
3.57
56
4.75
57
3.03
60
7.75
62
4.17
61
4.91
61
3.09
60
6.20
61
3.95
61
6.71
60
3.51
60
5.30
60
3.67
60
5.86
62
3.63
57
5.61
61
3.46
60
CC-Net-ROBtwo views4.74
60
40.54
63
2.83
55
2.84
56
2.86
55
2.81
55
2.90
59
2.91
59
2.90
59
2.84
58
2.89
59
2.78
57
2.83
59
2.82
56
2.90
59
2.80
57
2.80
57
2.89
59
2.85
55
2.88
54
2.88
59
CC-Nettwo views5.04
61
50.87
64
2.78
54
2.87
57
2.71
54
2.78
54
0.47
37
2.88
58
2.86
58
1.89
53
2.67
58
2.79
58
2.79
58
2.78
55
2.83
58
2.82
58
2.82
58
2.83
58
2.77
54
2.83
53
2.76
58
SGM-Foresttwo views5.21
62
5.92
58
4.08
59
6.18
61
4.16
60
6.31
62
4.34
62
6.50
61
4.33
62
6.14
62
4.21
62
6.61
62
4.55
62
6.67
59
4.48
62
5.94
62
3.94
61
5.85
61
4.03
61
5.79
62
4.17
62
Johannes L. Schönberger, Sudipta Sinha, Marc Pollefeys: Learning to Fuse Proposals from Multiple Scanline Optimizations in Semi-Global Matching. ECCV 2018
SGM+DAISYtwo views7.06
63
9.15
60
5.38
62
8.84
62
5.18
63
8.80
63
5.31
63
8.79
63
5.28
63
8.89
63
5.20
63
8.93
63
5.33
63
8.95
63
5.36
63
8.70
63
5.21
63
8.74
63
5.20
63
8.89
63
5.15
63
WCMA_ROBtwo views31.10
64
35.43
62
27.12
64
39.51
64
23.10
64
38.78
64
25.30
64
37.49
64
25.39
64
37.29
64
27.02
64
38.52
64
26.48
64
37.80
64
26.44
64
36.28
64
22.65
64
33.90
64
22.30
64
37.10
64
24.18
64
MDST_ROBtwo views69.83
65
87.70
65
41.95
65
113.75
67
65.62
67
75.05
65
55.25
66
75.64
66
45.04
65
71.61
65
41.75
65
72.81
65
44.06
65
68.38
65
44.63
65
101.89
67
59.57
67
107.10
67
61.05
67
104.38
67
59.38
67
NOSS_ROBtwo views102.95
66
153.00
66
121.00
67
51.00
65
44.00
65
165.00
68
127.00
67
153.00
67
119.00
67
164.00
68
125.00
67
168.00
68
120.00
67
153.00
67
117.00
67
49.00
65
44.00
65
49.00
65
44.00
65
49.00
65
44.00
65
CBMVpermissivetwo views128.50
67
1422.70
72
53.10
66
79.50
66
51.30
66
77.30
66
49.70
65
74.00
65
48.20
66
77.20
66
48.80
66
73.90
66
48.00
66
73.80
66
48.80
66
70.40
66
45.10
66
68.90
66
46.10
66
68.20
66
45.00
66
Konstantinos Batsos, Changjiang Cai, Philippos Mordohai: CBMV: A Coalesced Bidirectional Matching Volume for Disparity Estimation. Computer Vision and Pattern Recognition (CVPR) 2018
MeshStereopermissivetwo views159.24
68
171.00
67
160.68
68
162.58
68
160.59
68
164.01
67
160.35
68
158.51
68
158.56
68
158.34
67
160.12
68
158.56
67
159.92
68
157.26
68
158.94
68
154.38
68
158.36
68
155.75
68
159.13
68
153.67
68
154.07
68
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
DLCB_ROBtwo views284.23
69
354.61
68
207.27
69
363.24
69
206.46
69
364.72
69
210.41
69
364.72
69
210.41
69
364.81
69
208.64
69
364.81
69
208.64
69
364.72
69
210.41
69
354.70
69
205.53
69
354.70
69
205.53
69
354.70
69
205.53
69
LE_ROBtwo views396.57
70
471.28
69
329.84
70
471.48
70
308.15
70
526.83
70
322.10
70
488.15
70
323.76
70
495.46
70
317.97
70
497.17
70
320.10
70
481.62
70
326.76
70
462.71
70
298.97
70
466.16
70
285.98
70
447.62
70
289.21
70
SGM-ForestMtwo views596.69
71
677.77
70
444.52
71
699.85
71
517.25
71
732.94
71
488.29
71
770.79
71
460.11
71
750.81
71
487.98
71
792.79
71
499.41
71
730.90
71
475.81
71
720.03
71
491.16
71
663.96
71
418.60
71
674.76
71
436.05
71
CBMV_ROBtwo views818.48
72
913.88
71
709.52
72
862.84
72
597.78
72
1073.99
72
700.52
72
1015.66
72
702.59
72
1115.65
72
760.02
72
1130.24
72
721.57
72
1037.41
72
692.65
72
814.05
72
564.29
72
843.28
72
595.31
72
915.51
72
602.92
72