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 views0.98
1
0.42
11
1.86
1
0.07
1
0.22
7
1.65
1
0.02
1
4.07
2
6.33
23
0.72
1
1.54
2
0.43
2
0.64
1
1.22
1
0.44
1
0.00
1
0.00
1
0.00
1
0.00
1
0.01
2
0.00
1
Jiankun Li, Peisen Wang, Pengfei Xiong, Tao Cai, Ziwei Yan, Lei Yang, Jiangyu Liu, Haoqiang Fan, Shuaicheng Liu: Practical Stereo Matching via Cascaded Recurrent Network with Adaptive Correlation. CVPR 2022
PMTNettwo views1.41
2
0.37
9
2.06
4
0.26
3
0.15
4
2.94
13
0.34
10
3.39
1
5.33
12
1.00
2
1.27
1
0.28
1
0.93
6
4.04
10
0.74
4
5.00
140
0.00
1
0.00
1
0.00
1
0.03
15
0.01
2
DIP-Stereotwo views1.97
3
0.64
30
2.95
15
0.17
2
0.10
1
4.83
38
0.13
3
8.60
18
4.06
4
6.42
29
4.92
7
0.44
3
0.72
2
3.57
9
1.80
9
0.00
1
0.01
40
0.00
1
0.00
1
0.05
24
0.04
28
Zihua Zheng, Ni Nie, Zhi Ling, Pengfei Xiong, Jiangyu Liu, Hao Wang, Jiankun Li: DIP: Deep Inverse Patchmatch for High-Resolution Optical Flow. cvpr2022
Gwc-CoAtRStwo views2.02
4
0.33
6
2.61
5
1.04
11
0.30
10
3.56
18
0.40
13
5.89
4
5.94
18
5.08
14
5.01
8
5.55
18
1.72
11
1.60
2
0.91
6
0.01
31
0.09
90
0.00
1
0.25
107
0.01
2
0.02
7
CFNet-RSSMtwo views2.11
5
0.33
6
2.71
10
1.04
11
0.30
10
3.21
16
0.41
14
6.67
8
6.96
29
5.62
19
4.58
6
5.40
16
1.71
10
1.86
5
0.87
5
0.00
1
0.08
85
0.00
1
0.32
113
0.02
7
0.02
7
FENettwo views2.19
6
0.34
8
3.04
17
2.05
55
0.26
9
2.82
10
0.13
3
7.02
9
5.55
14
4.97
13
1.94
3
4.91
13
3.39
25
4.71
13
2.50
15
0.02
41
0.03
60
0.00
1
0.00
1
0.00
1
0.21
68
EAI-Stereotwo views2.31
7
0.44
12
2.72
11
0.68
5
0.56
21
3.51
17
2.62
61
11.42
59
4.14
5
2.09
3
7.23
20
1.21
4
0.91
5
5.67
16
1.51
8
0.01
31
0.25
116
0.00
1
0.88
132
0.04
19
0.26
80
R-Stereo Traintwo views2.44
8
0.32
1
1.93
2
0.94
8
0.16
5
3.67
20
0.61
25
6.37
6
3.08
1
9.14
59
17.44
91
1.80
5
0.77
3
1.76
3
0.70
2
0.00
1
0.01
40
0.00
1
0.00
1
0.01
2
0.03
18
RAFT-Stereopermissivetwo views2.44
8
0.32
1
1.93
2
0.94
8
0.16
5
3.67
20
0.61
25
6.37
6
3.08
1
9.14
59
17.44
91
1.80
5
0.77
3
1.76
3
0.70
2
0.00
1
0.01
40
0.00
1
0.00
1
0.01
2
0.03
18
Lahav Lipson, Zachary Teed, and Jia Deng: RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matching. 3DV
ACVNettwo views2.58
10
0.53
20
2.69
7
0.84
6
0.59
22
2.61
6
0.55
20
8.35
13
5.70
15
4.45
8
9.26
36
5.77
20
3.50
27
2.24
6
4.41
41
0.00
1
0.00
1
0.00
1
0.00
1
0.17
58
0.01
2
DN-CSS_ROBtwo views2.69
11
1.40
81
5.34
53
2.31
72
0.75
29
3.14
15
0.06
2
6.11
5
3.87
3
5.34
17
12.18
60
2.34
7
1.22
7
7.84
39
1.48
7
0.03
52
0.00
1
0.00
1
0.00
1
0.35
88
0.03
18
HITNettwo views2.79
12
0.77
35
4.02
33
2.03
54
0.11
3
5.58
46
0.59
24
9.24
25
5.15
10
6.42
29
7.26
21
3.66
8
2.92
22
4.07
11
3.87
38
0.00
1
0.00
1
0.00
1
0.00
1
0.06
29
0.02
7
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
DMCAtwo views2.82
13
0.64
30
4.45
40
1.61
31
0.89
36
3.86
24
0.57
23
9.18
23
5.10
9
6.24
25
6.49
14
7.17
34
2.13
14
3.31
8
4.73
45
0.00
1
0.01
40
0.00
1
0.00
1
0.04
19
0.02
7
ccstwo views3.04
14
0.39
10
3.08
19
1.78
39
0.52
20
2.04
2
0.50
18
13.09
79
13.71
85
3.54
6
5.36
11
5.50
17
2.45
16
4.81
14
2.88
19
0.09
75
0.08
85
0.12
106
0.10
95
0.20
64
0.50
100
AdaStereotwo views3.09
15
0.58
24
3.04
17
2.84
91
0.48
19
4.08
27
1.29
47
12.16
70
7.77
44
6.03
22
9.62
38
5.79
22
1.53
9
4.56
12
1.93
11
0.00
1
0.00
1
0.00
1
0.00
1
0.02
7
0.02
7
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 views3.17
16
1.26
74
6.39
72
2.10
58
1.51
55
3.02
14
0.56
22
7.84
12
5.06
8
9.23
62
5.19
9
8.46
44
2.72
20
6.31
21
3.22
26
0.06
71
0.11
93
0.09
102
0.02
63
0.16
56
0.08
37
BEATNet_4xtwo views3.24
17
1.27
75
5.89
63
1.56
28
0.10
1
5.26
42
1.07
42
10.08
35
5.50
13
6.89
34
7.73
25
4.53
12
4.13
39
5.05
15
5.27
49
0.04
60
0.05
70
0.00
1
0.00
1
0.23
78
0.23
71
GANet-RSSMtwo views3.27
18
0.60
25
3.38
23
2.94
98
2.00
72
2.61
6
0.48
16
10.93
56
6.11
21
5.47
18
10.20
42
6.57
26
5.44
60
6.25
20
2.23
13
0.00
1
0.00
1
0.08
99
0.00
1
0.17
58
0.03
18
NOSS_ROBtwo views3.30
19
0.46
13
2.62
6
2.08
56
1.01
44
5.60
47
0.74
37
10.37
43
11.48
74
5.15
15
8.43
31
5.67
19
1.73
12
7.97
41
2.34
14
0.02
41
0.06
76
0.00
1
0.00
1
0.07
31
0.14
58
CFNet_RVCtwo views3.31
20
0.94
52
2.69
7
1.50
25
2.38
80
2.81
8
0.68
31
8.35
13
7.43
38
4.45
8
9.94
39
10.20
56
4.60
45
6.49
23
3.41
30
0.00
1
0.00
1
0.03
83
0.00
1
0.22
73
0.03
18
CFNet-ftpermissivetwo views3.31
20
0.94
52
2.69
7
1.50
25
2.38
80
2.81
8
0.68
31
8.35
13
7.43
38
4.45
8
9.94
39
10.20
56
4.60
45
6.50
24
3.41
30
0.00
1
0.00
1
0.03
83
0.00
1
0.22
73
0.03
18
PSMNet-RSSMtwo views3.33
22
0.77
35
3.22
22
2.12
59
1.90
71
2.33
4
0.71
34
11.17
57
5.83
17
5.90
20
11.43
54
7.04
30
4.83
54
5.96
18
3.17
24
0.00
1
0.00
1
0.02
75
0.00
1
0.08
33
0.02
7
cf-rtwo views3.42
23
0.81
39
3.97
32
2.41
76
2.78
90
2.85
11
0.64
30
8.44
16
6.32
22
6.25
26
11.57
55
8.97
49
3.54
29
5.93
17
3.83
37
0.00
1
0.00
1
0.00
1
0.00
1
0.09
37
0.07
34
MLCVtwo views3.44
24
0.88
44
5.60
58
1.39
20
0.25
8
4.36
31
0.33
9
7.25
10
7.28
35
9.17
61
12.24
62
5.09
14
2.47
17
9.15
64
3.23
27
0.00
1
0.00
1
0.00
1
0.00
1
0.10
38
0.02
7
DeepPruner_ROBtwo views3.52
25
1.14
67
4.06
34
1.12
14
1.65
59
3.65
19
0.83
39
13.96
88
4.47
7
7.80
43
10.84
47
7.05
32
2.16
15
8.14
48
3.08
23
0.07
73
0.03
60
0.00
1
0.01
59
0.32
84
0.06
32
STTStereotwo views3.60
26
0.93
51
6.34
70
2.71
88
2.23
78
3.68
22
0.63
28
9.42
26
6.73
27
9.87
70
6.97
18
8.84
48
3.65
31
6.85
25
3.04
21
0.00
1
0.02
52
0.01
63
0.00
1
0.02
7
0.02
7
ccs_robtwo views3.63
27
1.12
66
4.42
38
2.52
80
0.91
39
5.50
45
0.21
6
10.11
38
9.11
57
6.55
32
11.28
51
8.32
43
2.55
18
7.66
35
2.01
12
0.00
1
0.00
1
0.00
1
0.00
1
0.20
64
0.08
37
hitnet-ftcopylefttwo views3.66
28
0.65
32
2.79
13
0.89
7
0.72
28
4.25
29
0.55
20
9.11
22
7.52
40
7.14
36
10.19
41
12.49
75
4.70
49
7.44
31
4.41
41
0.03
52
0.00
1
0.06
95
0.00
1
0.20
64
0.04
28
iResNettwo views3.68
29
0.91
47
7.94
89
2.97
99
0.34
13
4.44
35
0.48
16
7.70
11
9.74
61
7.72
42
12.74
66
4.03
9
2.87
21
8.05
44
3.37
29
0.02
41
0.01
40
0.00
1
0.00
1
0.10
38
0.09
41
acv_fttwo views3.69
30
0.53
20
4.58
43
2.31
72
4.71
127
7.44
68
0.86
40
8.54
17
5.70
15
9.29
63
9.26
36
5.77
20
4.15
40
2.24
6
8.13
79
0.00
1
0.00
1
0.00
1
0.00
1
0.29
80
0.01
2
ac_64two views3.70
31
0.47
14
3.50
26
2.87
92
3.50
104
5.91
50
0.72
35
10.17
40
4.34
6
8.16
49
5.80
12
10.82
64
5.64
62
7.08
28
4.73
45
0.03
52
0.00
1
0.00
1
0.00
1
0.08
33
0.14
58
CFNettwo views3.72
32
1.10
64
5.03
47
2.49
79
1.59
56
4.90
40
0.22
7
11.38
58
9.88
63
4.80
11
11.25
50
6.44
25
3.68
32
8.33
50
3.00
20
0.00
1
0.00
1
0.00
1
0.00
1
0.22
73
0.07
34
GwcNet-RSSMtwo views3.77
33
1.05
58
5.06
48
2.21
65
1.87
69
2.36
5
1.52
50
8.98
21
7.96
46
5.91
21
14.59
76
7.94
41
3.50
27
6.43
22
5.70
54
0.00
1
0.00
1
0.04
90
0.00
1
0.20
64
0.08
37
FADNet-RVC-Resampletwo views3.79
34
1.62
93
12.06
107
1.43
23
0.66
24
5.94
51
2.41
59
10.18
41
8.58
54
6.28
27
4.22
5
5.33
15
4.80
53
7.71
36
3.19
25
0.17
89
0.21
110
0.17
113
0.12
97
0.41
97
0.29
85
UPFNettwo views3.82
35
0.47
14
4.34
37
2.21
65
3.38
100
7.26
65
2.69
62
9.89
31
5.96
19
7.98
46
7.53
23
7.89
40
3.15
24
7.72
37
5.88
56
0.01
31
0.02
52
0.00
1
0.00
1
0.04
19
0.06
32
NLCA_NET_v2_RVCtwo views3.84
36
1.06
59
5.23
51
2.72
89
3.27
96
4.36
31
0.61
25
10.71
50
7.56
41
8.75
54
7.89
26
9.86
54
3.90
36
7.15
29
3.44
32
0.14
83
0.02
52
0.02
75
0.03
72
0.04
19
0.03
18
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 views3.84
36
1.07
60
5.23
51
2.65
85
2.96
93
4.22
28
0.69
33
10.43
45
7.72
42
8.78
55
8.29
29
9.61
52
4.02
38
7.16
30
3.65
35
0.13
82
0.03
60
0.02
75
0.03
72
0.05
24
0.03
18
psm_uptwo views3.85
38
0.82
40
3.01
16
2.00
53
2.58
88
5.28
43
3.34
70
10.85
54
6.33
23
7.27
38
7.32
22
10.15
55
9.02
89
6.19
19
2.51
16
0.22
94
0.01
40
0.00
1
0.00
1
0.12
45
0.03
18
FADNet_RVCtwo views3.91
39
1.67
96
12.95
113
0.96
10
0.75
29
5.71
49
0.54
19
10.83
53
6.60
26
3.46
4
8.09
27
4.10
10
3.40
26
9.43
70
6.33
59
0.36
107
0.44
126
0.17
113
0.46
125
0.91
116
0.95
119
UNettwo views3.97
40
0.61
27
6.50
75
1.68
36
3.22
95
9.82
83
0.31
8
9.50
27
5.32
11
7.20
37
6.35
13
10.59
61
5.91
66
6.93
27
5.49
52
0.00
1
0.00
1
0.00
1
0.00
1
0.02
7
0.01
2
FADNet-RVCtwo views3.98
41
1.84
103
12.48
110
1.69
37
0.44
17
4.33
30
1.31
48
11.84
64
7.15
33
3.53
5
3.50
4
10.63
62
4.43
44
9.12
62
6.25
58
0.03
52
0.10
91
0.00
1
0.03
72
0.60
105
0.25
77
HSMtwo views4.00
42
0.79
38
3.16
21
1.59
30
2.17
75
6.77
61
1.11
43
12.28
71
6.35
25
6.75
33
8.11
28
13.90
81
5.37
59
8.85
58
2.71
18
0.00
1
0.00
1
0.00
1
0.00
1
0.02
7
0.02
7
TDLMtwo views4.11
43
1.11
65
3.54
27
1.62
32
1.04
45
3.91
25
7.41
115
10.60
48
10.67
68
6.38
28
12.59
65
5.95
23
4.77
51
8.79
57
3.04
21
0.58
120
0.00
1
0.01
63
0.00
1
0.19
63
0.12
53
CBMV_ROBtwo views4.14
44
0.52
17
3.14
20
1.30
17
0.77
32
6.92
62
1.97
57
10.11
38
9.58
59
8.92
58
14.20
75
7.12
33
5.90
65
8.65
54
3.50
34
0.01
31
0.05
70
0.00
1
0.00
1
0.04
19
0.09
41
CVANet_RVCtwo views4.16
45
1.16
69
3.60
28
1.94
49
1.46
53
3.92
26
4.68
92
10.89
55
8.34
51
7.58
41
10.84
47
10.27
58
6.62
72
8.56
53
2.69
17
0.39
109
0.00
1
0.00
1
0.01
59
0.21
71
0.09
41
MMNettwo views4.16
45
0.77
35
5.47
56
1.40
21
1.67
61
10.78
92
0.63
28
8.64
20
8.47
53
8.80
56
8.33
30
7.51
36
6.61
71
7.53
33
6.54
62
0.00
1
0.00
1
0.00
1
0.00
1
0.03
15
0.05
31
HSM-Net_RVCpermissivetwo views4.20
47
0.32
1
2.76
12
0.63
4
0.69
26
6.95
63
1.69
53
11.96
65
8.36
52
8.83
57
12.17
59
15.18
89
4.21
42
6.91
26
3.30
28
0.02
41
0.02
52
0.00
1
0.00
1
0.01
2
0.01
2
Gengshan Yang, Joshua Manela, Michael Happold, and Deva Ramanan: Hierarchical Deep Stereo Matching on High-resolution Images. CVPR 2019
DSFCAtwo views4.21
48
0.50
16
5.45
55
1.34
18
1.68
63
7.04
64
4.51
89
10.73
51
7.00
30
10.78
78
6.80
15
8.48
46
4.63
47
9.91
74
5.12
48
0.02
41
0.06
76
0.02
75
0.03
72
0.08
33
0.03
18
FADNettwo views4.23
49
1.65
95
11.75
106
1.64
34
0.80
34
4.80
37
0.77
38
13.76
87
11.65
76
3.97
7
5.24
10
9.62
53
5.14
56
8.40
51
3.78
36
0.21
93
0.04
65
0.07
98
0.05
84
1.14
121
0.10
49
iResNet_ROBtwo views4.23
49
1.02
56
4.90
46
2.18
62
0.93
42
2.92
12
0.37
12
15.10
99
16.91
100
7.89
45
10.51
45
7.03
29
3.07
23
8.16
49
3.46
33
0.01
31
0.00
1
0.00
1
0.00
1
0.10
38
0.02
7
HGLStereotwo views4.25
51
0.52
17
4.65
44
3.09
101
5.18
129
6.51
59
0.73
36
10.37
43
6.74
28
7.32
39
12.44
64
7.00
28
7.21
76
7.46
32
5.51
53
0.00
1
0.02
52
0.00
1
0.00
1
0.06
29
0.10
49
delettwo views4.27
52
0.72
34
4.50
41
1.07
13
3.75
111
10.79
93
4.04
79
9.95
32
7.89
45
8.09
47
8.89
34
8.46
44
3.81
34
7.56
34
5.80
55
0.00
1
0.00
1
0.00
1
0.00
1
0.11
42
0.02
7
iResNetv2_ROBtwo views4.28
53
1.43
82
7.17
84
2.91
93
1.26
49
4.36
31
1.62
51
13.64
86
10.25
67
9.83
69
11.41
52
7.68
37
4.00
37
7.75
38
1.85
10
0.00
1
0.00
1
0.00
1
0.00
1
0.37
90
0.09
41
StereoDRNet-Refinedtwo views4.46
54
0.62
29
3.80
31
1.92
46
0.40
15
9.35
77
0.15
5
10.02
33
8.83
56
12.69
90
11.62
56
9.34
50
3.87
35
8.06
45
8.02
75
0.00
1
0.00
1
0.01
63
0.05
84
0.20
64
0.26
80
Rohan Chabra, Julian Straub, Chris Sweeney, Richard Newcombe, Henry Fuchs: StereoDRNet. CVPR
DLCB_ROBtwo views4.51
55
0.91
47
3.78
30
2.19
63
1.07
46
6.28
52
3.09
65
9.78
30
7.72
42
10.65
76
12.97
67
13.91
82
3.71
33
8.72
55
5.30
50
0.00
1
0.00
1
0.00
1
0.00
1
0.03
15
0.10
49
NVstereo2Dtwo views4.51
55
0.82
40
6.86
82
3.28
108
3.38
100
8.16
71
3.13
66
10.51
46
15.15
90
4.90
12
6.89
17
7.87
38
4.78
52
9.88
73
3.91
39
0.01
31
0.00
1
0.00
1
0.06
86
0.02
7
0.58
106
RASNettwo views4.52
57
0.61
27
4.42
38
3.42
111
4.68
126
4.58
36
0.99
41
9.54
29
8.01
47
5.28
16
11.42
53
10.34
60
8.88
87
9.28
67
8.68
84
0.15
85
0.00
1
0.00
1
0.00
1
0.03
15
0.04
28
SGM-Foresttwo views4.96
58
0.32
1
2.84
14
1.21
15
0.64
23
10.23
88
6.64
109
11.55
60
10.98
69
10.94
80
13.59
71
11.65
70
4.30
43
8.94
60
4.63
44
0.11
78
0.04
65
0.00
1
0.00
1
0.05
24
0.46
96
Johannes L. Schönberger, Sudipta Sinha, Marc Pollefeys: Learning to Fuse Proposals from Multiple Scanline Optimizations in Semi-Global Matching. ECCV 2018
PA-Nettwo views4.98
59
1.47
84
7.42
86
2.40
75
2.14
74
8.73
74
3.64
77
12.42
72
13.11
81
7.03
35
7.57
24
7.88
39
6.52
70
10.16
76
7.82
72
0.02
41
0.03
60
0.00
1
0.00
1
0.11
42
1.07
122
Zhibo Rao, Mingyi He, Yuchao Dai, Zhelun Shen: Patch Attention Network with Generative Adversarial Model for Semi-Supervised Binocular Disparity Prediction.
AANet_RVCtwo views5.01
60
1.74
98
6.38
71
1.96
51
1.29
51
2.26
3
1.69
53
10.07
34
18.53
103
7.88
44
18.15
93
8.49
47
2.70
19
10.59
80
7.04
65
0.96
130
0.15
102
0.02
75
0.00
1
0.13
49
0.12
53
PSMNet_ROBtwo views5.02
61
1.63
94
6.03
65
1.90
45
1.83
68
9.57
81
6.35
106
15.58
104
7.23
34
6.15
24
10.48
44
12.22
73
4.16
41
8.02
42
8.71
85
0.02
41
0.01
40
0.01
63
0.10
95
0.20
64
0.12
53
FINETtwo views5.32
62
1.14
67
10.04
98
2.17
61
3.67
110
6.49
57
6.76
110
11.68
62
22.08
116
7.42
40
12.15
58
4.24
11
3.61
30
9.25
66
5.35
51
0.04
60
0.04
65
0.00
1
0.00
1
0.17
58
0.09
41
CBMVpermissivetwo views5.35
63
0.91
47
3.67
29
1.62
32
0.44
17
10.09
86
7.19
114
12.49
73
12.33
80
12.22
86
14.69
77
10.93
66
6.48
69
8.51
52
4.96
47
0.02
41
0.15
102
0.00
1
0.00
1
0.17
58
0.17
63
Konstantinos Batsos, Changjiang Cai, Philippos Mordohai: CBMV: A Coalesced Bidirectional Matching Volume for Disparity Estimation. Computer Vision and Pattern Recognition (CVPR) 2018
StereoDRNettwo views5.59
64
1.75
99
6.80
80
3.12
103
4.45
121
10.61
91
4.35
85
18.80
118
9.73
60
12.22
86
6.87
16
11.44
69
4.65
48
8.09
47
8.26
80
0.02
41
0.11
93
0.00
1
0.03
72
0.20
64
0.28
84
ETE_ROBtwo views5.80
65
1.77
100
6.33
69
1.44
24
0.78
33
6.43
56
6.90
111
12.53
74
8.08
48
12.93
95
14.89
78
21.13
118
5.87
64
9.83
72
6.57
63
0.04
60
0.01
40
0.00
1
0.02
63
0.08
33
0.33
86
DRN-Testtwo views5.87
66
0.98
54
5.89
63
2.69
87
3.65
109
12.37
102
3.35
71
20.07
130
10.20
66
11.93
85
12.31
63
11.06
68
5.31
58
7.89
40
9.05
88
0.04
60
0.05
70
0.04
90
0.04
82
0.18
62
0.25
77
NCCL2two views5.88
67
1.59
91
5.44
54
1.87
42
0.92
40
9.55
80
11.55
130
12.11
67
9.94
64
9.67
68
8.85
33
22.28
120
7.41
77
8.78
56
7.17
66
0.01
31
0.00
1
0.03
83
0.00
1
0.13
49
0.23
71
APVNettwo views5.98
68
1.72
97
8.72
92
3.21
106
4.04
115
12.31
101
1.81
56
19.09
120
6.10
20
6.13
23
13.41
70
10.69
63
6.30
67
11.26
87
13.90
113
0.05
68
0.02
52
0.11
104
0.03
72
0.53
100
0.20
67
NaN_ROBtwo views6.00
69
1.24
72
6.29
68
1.34
18
1.68
63
9.60
82
10.31
126
15.09
97
15.79
93
12.62
89
8.95
35
11.67
71
5.83
63
11.78
90
6.41
61
0.05
68
0.13
99
0.08
99
0.20
103
0.22
73
0.79
114
DANettwo views6.02
70
1.23
71
8.45
91
3.86
120
3.94
114
7.64
70
1.34
49
9.51
28
7.00
30
13.39
97
15.53
82
15.99
93
7.02
75
12.14
91
12.37
108
0.19
91
0.12
98
0.02
75
0.03
72
0.13
49
0.56
105
Z Ling, K Yang, J Li, Y Zhang, X Gao, L Luo, L Xie: Domain-adaptive modules for stereo matching network. Neurocomputing 2021
XPNet_ROBtwo views6.03
71
1.22
70
5.61
59
2.56
83
0.90
38
6.32
53
7.07
112
12.92
77
8.30
50
14.76
102
15.13
80
19.84
112
6.66
74
10.36
77
8.58
83
0.02
41
0.04
65
0.00
1
0.03
72
0.11
42
0.24
74
Anonymous Stereotwo views6.16
72
3.15
122
23.75
131
2.97
99
2.48
85
4.39
34
13.30
132
9.21
24
9.86
62
9.56
67
8.76
32
6.79
27
1.99
13
13.50
102
13.04
111
0.01
31
0.05
70
0.00
1
0.06
86
0.22
73
0.19
65
GANettwo views6.22
73
1.07
60
4.07
35
2.27
68
0.89
36
9.19
76
9.52
121
12.02
66
8.13
49
10.72
77
29.09
120
13.86
80
7.52
79
11.00
84
4.39
40
0.36
107
0.00
1
0.02
75
0.02
63
0.12
45
0.08
37
DISCOtwo views6.28
74
0.57
23
5.78
61
3.43
112
1.17
47
11.22
95
3.39
72
12.14
69
16.16
95
6.52
31
11.22
49
16.96
97
6.32
68
19.51
125
10.74
101
0.00
1
0.00
1
0.00
1
0.00
1
0.35
88
0.11
52
Syn2CoExtwo views6.32
75
2.72
115
9.33
95
2.45
78
1.67
61
8.59
72
3.32
69
18.08
114
12.27
79
8.53
53
21.08
99
10.30
59
9.80
93
9.30
68
7.86
73
0.01
31
0.00
1
0.00
1
0.00
1
0.63
107
0.43
93
RYNettwo views6.34
76
0.89
46
5.88
62
1.41
22
4.48
123
15.97
113
4.18
81
13.41
82
16.49
96
10.81
79
7.00
19
14.33
84
8.72
86
9.43
70
13.71
112
0.00
1
0.01
40
0.00
1
0.00
1
0.02
7
0.07
34
GwcNetcopylefttwo views6.42
77
1.97
106
10.92
101
2.59
84
5.58
132
11.55
97
2.21
58
14.10
90
16.52
99
10.04
72
17.19
90
10.86
65
5.61
61
9.23
65
8.84
87
0.01
31
0.06
76
0.03
83
0.08
90
0.56
101
0.40
90
GANetREF_RVCpermissivetwo views6.56
78
2.89
117
7.58
88
3.41
110
0.40
15
12.96
105
9.58
122
15.09
97
17.25
102
10.33
73
10.62
46
12.27
74
8.16
81
12.21
92
4.53
43
0.41
111
0.00
1
0.00
1
0.02
63
3.12
136
0.39
89
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
LALA_ROBtwo views6.58
79
1.80
102
6.25
67
1.26
16
0.94
43
10.08
85
9.02
118
16.00
105
11.51
75
12.74
91
13.02
68
24.77
122
5.25
57
10.56
79
8.02
75
0.04
60
0.05
70
0.00
1
0.02
63
0.10
38
0.25
77
S-Stereotwo views6.63
80
0.84
42
9.67
97
3.15
104
3.48
103
6.49
57
6.22
101
12.99
78
22.84
117
9.48
64
15.51
81
12.00
72
8.43
83
8.04
43
10.70
100
0.12
81
0.17
106
0.00
1
0.38
118
0.13
49
1.92
131
DeepPrunerFtwo views6.75
81
2.69
114
23.31
130
3.68
115
7.16
137
3.78
23
4.29
83
13.42
83
20.13
113
8.13
48
10.46
43
7.18
35
8.06
80
11.10
85
9.44
90
0.24
96
0.15
102
0.29
123
0.42
121
0.66
109
0.45
94
edge stereotwo views6.76
82
1.01
55
6.76
79
2.20
64
2.45
84
6.41
55
2.45
60
14.84
96
11.98
78
15.29
103
18.31
94
22.02
119
12.56
104
10.82
81
7.49
68
0.03
52
0.06
76
0.11
104
0.03
72
0.30
81
0.14
58
NCC-stereotwo views6.77
83
1.49
85
6.48
73
2.92
95
4.40
117
7.43
66
3.61
75
19.52
126
13.29
82
8.39
51
16.91
85
15.96
91
12.13
102
12.85
96
7.70
70
1.47
134
0.11
93
0.01
63
0.42
121
0.14
54
0.24
74
Abc-Nettwo views6.77
83
1.49
85
6.48
73
2.92
95
4.40
117
7.43
66
3.61
75
19.52
126
13.29
82
8.39
51
16.91
85
15.96
91
12.13
102
12.85
96
7.70
70
1.47
134
0.11
93
0.01
63
0.42
121
0.14
54
0.24
74
Xing Li, Yangyu Fan, Guoyun Lv, and Haoyue Ma: Area-based Correlation and Non-local Attention Network for Stereo Matching. The Visual Computer
FAT-Stereotwo views6.78
85
0.68
33
6.80
80
2.30
70
1.77
67
5.63
48
4.20
82
18.79
117
18.62
104
10.53
75
17.15
88
18.52
106
13.74
112
8.86
59
7.38
67
0.03
52
0.15
102
0.01
63
0.07
88
0.12
45
0.26
80
RPtwo views6.84
86
1.29
79
5.53
57
3.92
121
5.18
129
6.32
53
3.53
73
11.73
63
15.31
91
9.54
66
22.38
103
18.25
104
14.47
113
10.11
75
7.49
68
0.91
129
0.01
40
0.12
106
0.15
99
0.33
85
0.19
65
ccnettwo views6.87
87
0.92
50
5.15
50
3.09
101
2.18
77
11.78
99
4.90
93
17.97
113
8.79
55
12.79
93
20.20
97
16.00
94
8.35
82
9.12
62
14.17
115
0.42
114
0.33
120
0.25
122
0.09
93
0.40
96
0.48
97
RGCtwo views6.88
88
2.23
110
6.13
66
4.05
122
4.73
128
8.94
75
2.78
64
15.19
101
11.74
77
11.13
81
19.34
96
17.86
101
10.42
96
13.02
99
8.03
77
0.73
123
0.01
40
0.24
121
0.41
120
0.31
83
0.38
88
Nwc_Nettwo views6.97
89
1.25
73
6.63
77
3.82
119
3.37
99
10.83
94
1.67
52
19.56
128
11.35
72
8.36
50
23.62
106
17.19
99
11.44
101
11.21
86
8.08
78
0.80
125
0.00
1
0.00
1
0.02
63
0.13
49
0.09
41
ADCReftwo views7.27
90
1.38
80
16.37
119
2.52
80
3.30
98
11.63
98
3.16
67
10.80
52
9.35
58
13.03
96
25.27
114
8.17
42
8.92
88
8.06
45
21.81
129
0.15
85
0.08
85
0.16
112
0.34
116
0.38
91
0.58
106
psmorigintwo views7.34
91
1.58
90
18.31
123
2.35
74
0.87
35
6.72
60
1.70
55
10.63
49
7.14
32
19.77
115
22.71
104
20.13
114
11.34
100
12.96
98
9.49
91
0.04
60
0.17
106
0.06
95
0.17
100
0.21
71
0.50
100
CSANtwo views7.62
92
1.60
92
6.56
76
1.83
40
0.66
24
12.40
103
10.52
128
14.45
93
21.32
115
14.19
99
15.98
84
17.84
100
13.02
109
12.32
93
8.38
81
0.09
75
0.07
83
0.03
83
0.04
82
0.33
85
0.67
112
stereogantwo views7.69
93
0.88
44
7.08
83
3.49
113
3.93
113
18.98
121
3.23
68
16.52
107
19.58
110
9.93
71
18.92
95
20.50
116
9.04
90
14.07
106
6.14
57
0.26
99
0.04
65
0.21
118
0.03
72
0.63
107
0.33
86
pmcnntwo views7.72
94
1.27
75
9.42
96
2.91
93
3.14
94
9.44
78
6.23
102
12.56
75
16.51
97
14.53
100
24.08
108
27.44
127
8.49
84
9.32
69
8.44
82
0.06
71
0.08
85
0.00
1
0.00
1
0.30
81
0.15
61
AF-Nettwo views7.78
95
1.44
83
6.68
78
3.37
109
4.50
124
8.61
73
2.69
62
17.07
110
20.17
114
9.52
65
24.02
107
20.31
115
14.59
114
11.58
89
9.84
96
0.61
121
0.00
1
0.12
106
0.00
1
0.38
91
0.12
53
PASMtwo views7.90
96
4.22
128
21.97
128
3.25
107
3.29
97
5.39
44
6.57
108
10.57
47
19.09
106
12.77
92
13.92
73
18.11
103
9.51
92
13.79
105
10.77
103
0.19
91
0.45
127
0.29
123
1.08
135
1.49
128
1.19
124
PWCDC_ROBbinarytwo views7.92
97
3.17
124
7.48
87
5.73
133
4.40
117
10.45
89
0.35
11
14.52
94
28.19
125
10.36
74
31.27
124
7.04
30
9.14
91
13.22
101
8.78
86
2.74
139
0.02
52
0.00
1
0.00
1
1.31
126
0.17
63
ADCP+two views8.09
98
1.79
101
14.50
117
1.54
27
4.28
116
16.57
115
5.20
95
12.80
76
11.20
71
12.83
94
17.07
87
11.02
67
10.80
98
17.59
121
23.18
132
0.03
52
0.05
70
0.01
63
0.18
101
0.39
95
0.81
115
SuperBtwo views8.10
99
3.15
122
24.67
132
2.65
85
1.23
48
9.88
84
4.29
83
10.18
41
30.07
128
11.53
83
12.18
60
6.12
24
6.65
73
10.50
78
14.47
116
0.14
83
0.11
93
0.35
127
0.25
107
13.06
144
0.48
97
PWC_ROBbinarytwo views8.24
100
3.13
120
12.74
111
2.43
77
4.43
120
7.51
69
1.22
44
16.63
108
19.24
108
16.08
105
28.29
117
13.99
83
10.16
95
13.63
104
14.06
114
0.42
114
0.00
1
0.05
93
0.00
1
0.59
104
0.27
83
MDST_ROBtwo views8.37
101
0.32
1
9.03
93
4.18
125
2.42
83
26.86
134
6.14
99
19.36
123
13.52
84
27.09
129
22.75
105
9.47
51
4.74
50
15.06
112
6.34
60
0.02
41
0.02
52
0.00
1
0.00
1
0.02
7
0.13
57
G-Nettwo views8.41
102
1.54
88
10.97
102
5.73
133
3.60
106
26.19
131
4.41
86
10.10
36
7.42
36
19.71
113
24.99
112
14.38
85
15.83
115
10.99
82
9.53
92
0.50
118
0.46
128
0.19
116
0.25
107
0.80
113
0.66
110
STTStereo_v2two views8.41
102
1.54
88
10.97
102
5.73
133
3.60
106
26.19
131
4.41
86
10.10
36
7.42
36
19.71
113
24.99
112
14.38
85
15.83
115
10.99
82
9.53
92
0.50
118
0.46
128
0.19
116
0.25
107
0.80
113
0.66
110
XQCtwo views8.43
104
3.58
126
16.40
120
2.92
95
2.17
75
13.22
107
3.60
74
14.64
95
25.86
122
11.87
84
12.04
57
15.06
88
10.67
97
15.24
113
19.41
121
0.39
109
0.08
85
0.05
93
0.07
88
0.84
115
0.45
94
FBW_ROBtwo views8.50
105
1.03
57
7.98
90
1.93
48
1.28
50
13.10
106
6.23
102
22.50
134
18.98
105
18.82
110
14.91
79
19.06
109
10.04
94
18.41
122
9.83
95
0.62
122
0.22
111
1.82
141
0.82
131
0.99
119
1.36
126
aanetorigintwo views8.72
106
3.29
125
27.55
134
1.95
50
3.84
112
4.93
41
5.39
97
5.49
3
10.05
65
27.85
130
31.47
125
15.80
90
12.78
106
9.04
61
11.98
106
0.25
97
0.22
111
0.22
119
0.25
107
1.28
125
0.84
117
RTSCtwo views9.15
107
3.00
119
13.57
115
3.72
117
1.76
66
11.82
100
0.46
15
16.95
109
36.83
137
15.80
104
15.53
82
12.91
77
7.46
78
20.01
128
21.76
128
0.31
104
0.13
99
0.01
63
0.08
90
0.57
102
0.41
92
WCMA_ROBtwo views9.21
108
0.87
43
7.37
85
2.54
82
2.13
73
13.59
108
5.80
98
11.64
61
14.01
86
24.43
126
32.99
130
27.09
126
18.02
119
12.51
95
9.85
97
0.81
126
0.07
83
0.01
63
0.01
59
0.16
56
0.23
71
MSMD_ROBtwo views9.28
109
1.09
63
4.65
44
1.58
29
0.39
14
16.52
114
4.41
86
13.60
84
14.87
89
22.34
118
39.89
138
25.67
124
20.71
128
12.42
94
6.98
64
0.34
106
0.03
60
0.00
1
0.00
1
0.05
24
0.09
41
ADCPNettwo views9.54
110
2.39
112
31.46
136
2.09
57
1.60
57
16.71
117
6.39
107
12.11
67
11.45
73
13.53
98
21.45
101
19.41
110
10.94
99
14.38
108
21.54
127
0.27
102
1.16
135
0.39
130
1.49
139
0.58
103
1.45
127
SHDtwo views9.61
111
2.60
113
12.46
109
3.69
116
3.54
105
9.47
79
1.25
45
20.16
131
37.84
140
18.19
109
21.24
100
16.96
97
12.83
108
14.47
110
16.05
118
0.32
105
0.13
99
0.01
63
0.08
90
0.38
91
0.48
97
PDISCO_ROBtwo views9.62
112
1.99
107
11.51
104
9.88
141
9.61
142
21.48
125
3.83
78
19.33
122
28.49
126
11.27
82
14.17
74
19.92
113
5.02
55
16.35
118
9.18
89
5.28
142
0.41
123
0.14
110
0.09
93
2.05
132
2.36
137
MFN_U_SF_DS_RVCtwo views9.78
113
4.27
129
14.47
116
2.29
69
2.85
92
23.40
129
13.62
133
13.60
84
19.54
109
19.42
112
24.27
109
16.74
96
8.59
85
17.05
120
7.98
74
1.25
133
1.68
139
0.17
113
2.63
141
0.72
111
1.04
120
SGM_RVCbinarytwo views10.08
114
0.60
25
3.42
24
2.30
70
0.32
12
19.41
122
6.33
105
18.95
119
14.64
87
25.14
128
24.32
110
33.34
135
18.79
123
19.86
126
12.55
110
0.25
97
0.26
117
0.22
119
0.24
106
0.34
87
0.40
90
Heiko Hirschmueller: Stereo processing by semiglobal matching and mutual information. TPAMI 2008, Volume 30(2), pp. 328-341
DPSNettwo views10.14
115
1.88
105
16.82
121
1.85
41
1.73
65
24.84
130
17.20
142
19.92
129
27.41
124
12.23
88
13.62
72
16.52
95
18.35
120
14.42
109
12.50
109
0.78
124
0.54
132
0.08
99
0.25
107
1.18
122
0.59
109
ADCLtwo views10.16
116
2.11
108
19.36
125
1.92
46
1.88
70
22.23
126
8.91
117
14.04
89
23.56
119
14.62
101
26.19
115
12.75
76
13.59
111
16.06
117
22.95
131
0.26
99
0.18
108
0.75
133
0.65
127
0.69
110
0.58
106
ADCMidtwo views10.24
117
3.13
120
20.70
126
2.21
65
2.39
82
11.23
96
6.19
100
14.17
91
11.19
70
23.20
124
22.25
102
17.89
102
19.54
125
18.51
123
26.21
137
0.45
116
0.42
125
1.10
137
1.29
136
1.56
130
1.18
123
FCDSN-DCtwo views10.24
117
0.56
22
3.49
25
1.96
51
1.29
51
16.90
118
4.59
91
17.16
111
19.10
107
24.64
127
32.46
129
33.82
136
22.14
133
15.93
116
10.45
98
0.04
60
0.00
1
0.00
1
0.00
1
0.05
24
0.21
68
Dominik Hirner, Friedrich Fraundorfer: FCDSN-DC: An accurate but lightweight end-to-end trainable neural network for stereo estimation with depth completion.
SANettwo views10.64
119
1.86
104
10.91
100
1.76
38
0.71
27
14.62
111
9.23
120
19.18
121
37.14
138
19.22
111
27.96
116
25.86
125
19.11
124
13.02
99
10.63
99
0.08
74
0.06
76
0.03
83
0.02
63
0.62
106
0.81
115
FC-DCNNcopylefttwo views10.72
120
0.52
17
4.27
36
1.88
43
1.63
58
17.18
119
5.29
96
18.20
115
19.69
111
28.50
131
34.51
133
34.03
137
21.48
131
15.89
115
11.15
105
0.03
52
0.01
40
0.02
75
0.01
59
0.07
31
0.09
41
AnyNet_C32two views10.98
121
5.58
131
22.79
129
4.16
123
5.83
133
15.64
112
14.30
134
13.18
80
17.15
101
16.44
107
20.52
98
14.68
87
13.44
110
22.46
130
30.08
142
0.17
89
0.26
117
0.36
128
0.36
117
1.23
123
0.91
118
MeshStereopermissivetwo views11.52
122
1.52
87
4.55
42
1.89
44
1.46
53
19.87
124
5.11
94
20.66
132
15.91
94
32.67
136
34.51
133
39.34
142
21.15
129
18.74
124
12.10
107
0.11
78
0.06
76
0.01
63
0.00
1
0.45
99
0.22
70
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 views12.94
123
3.66
127
25.81
133
3.61
114
2.26
79
22.77
127
4.55
90
27.10
139
20.06
112
23.90
125
28.99
119
30.53
132
16.98
117
19.92
127
20.26
123
1.24
132
1.07
134
0.98
136
1.33
137
1.80
131
2.04
133
ADCStwo views13.02
124
4.93
130
28.38
135
3.17
105
2.67
89
13.61
109
10.83
129
18.70
116
33.46
132
22.59
119
24.78
111
19.59
111
18.51
122
23.40
133
32.16
144
0.10
77
0.19
109
0.37
129
0.18
101
1.26
124
1.46
128
MFMNet_retwo views13.29
125
8.60
138
18.29
122
9.75
140
7.25
139
19.65
123
14.84
138
20.71
133
30.72
129
23.03
122
28.77
118
18.85
108
26.09
138
13.55
103
9.82
94
2.44
137
1.35
138
0.34
126
0.23
105
4.78
140
6.69
141
EDNetEfficienttwo views13.82
126
6.19
133
54.79
142
2.83
90
3.60
106
4.86
39
8.57
116
8.61
19
32.90
131
40.77
142
30.59
123
18.33
105
21.87
132
11.38
88
25.36
135
0.15
85
0.23
113
0.06
95
1.03
134
2.54
134
1.79
129
LSMtwo views14.01
127
5.95
132
33.49
137
6.78
138
43.61
146
10.22
87
9.98
125
15.16
100
22.93
118
23.07
123
32.34
128
18.52
106
12.67
105
15.45
114
11.10
104
0.16
88
0.51
131
0.09
102
0.32
113
1.08
120
16.85
146
SAMSARAtwo views14.63
128
2.74
116
12.38
108
12.65
143
6.74
136
36.50
140
72.93
148
19.36
123
23.77
120
16.20
106
13.04
69
29.21
129
12.78
106
16.98
119
15.21
117
0.11
78
0.26
117
0.03
83
0.14
98
0.76
112
0.77
113
SPS-STEREOcopylefttwo views15.04
129
6.23
134
13.21
114
11.34
142
11.65
144
23.30
128
7.15
113
24.16
136
15.65
92
31.78
135
29.19
121
31.62
133
21.32
130
24.62
134
19.50
122
7.59
144
4.19
145
3.22
142
1.48
138
6.99
143
6.54
140
K. Yamaguchi, D. McAllester, R. Urtasun: Efficient Joint Segmentation, Occlusion Labeling, Stereo and Flow Estimation. ECCV 2014
PVDtwo views15.44
130
2.93
118
14.67
118
4.21
126
3.39
102
17.43
120
4.16
80
27.84
142
48.84
145
31.02
134
43.54
142
29.76
131
30.81
142
25.97
136
21.40
126
0.23
95
0.41
123
0.04
90
0.33
115
0.41
97
1.33
125
SGM+DAISYtwo views15.62
131
7.26
137
19.28
124
8.94
139
10.11
143
26.25
133
10.49
127
19.36
123
14.65
88
30.64
133
33.59
131
33.00
134
22.32
134
24.96
135
16.42
119
7.90
145
6.25
146
4.51
143
3.37
142
5.86
141
7.20
142
NVStereoNet_ROBtwo views16.04
132
6.75
135
12.90
112
6.37
137
7.42
140
12.89
104
9.74
123
22.78
135
25.12
121
30.32
132
46.19
144
34.37
138
25.38
136
21.48
129
21.38
125
5.94
143
3.10
143
6.07
144
10.09
146
4.01
138
8.54
144
Nikolai Smolyanskiy, Alexey Kamenev, Stan Birchfield: On the Importance of Stereo for Accurate Depth Estimation: An Efficient Semi-Supervised Deep Neural Network Approach. Arxiv
AnyNet_C01two views16.12
133
10.81
142
59.36
143
4.42
127
2.49
86
30.06
136
15.15
140
17.51
112
16.51
97
17.88
108
37.69
137
24.04
121
17.54
118
29.60
139
33.29
145
0.28
103
0.38
121
0.43
131
0.42
121
2.57
135
1.98
132
MSC_U_SF_DS_RVCtwo views16.41
134
6.93
136
21.83
127
5.94
136
2.81
91
38.71
141
14.59
135
24.55
137
34.87
135
33.66
137
34.35
132
29.24
130
24.20
135
22.59
131
17.95
120
2.52
138
2.81
141
1.17
139
1.51
140
5.89
142
2.16
134
ELAS_RVCcopylefttwo views16.54
135
2.26
111
10.09
99
5.50
132
4.46
122
28.28
135
16.72
141
25.55
138
33.54
133
40.19
140
40.30
140
36.68
140
30.03
140
29.40
138
20.61
124
0.98
131
1.21
136
0.86
135
0.70
129
1.39
127
2.16
134
A. Geiger, M. Roser, R. Urtasun: Efficient large-scale stereo matching. ACCV 2010
ELAScopylefttwo views16.72
136
2.14
109
9.23
94
4.92
129
4.53
125
32.66
139
15.11
139
27.40
140
28.68
127
40.27
141
44.90
143
38.33
141
30.50
141
26.44
137
21.94
130
0.88
127
1.23
137
0.67
132
0.89
133
1.49
128
2.18
136
A. Geiger, M. Roser, R. Urtasun: Efficient large-scale stereo matching. ACCV 2010
LE_ROBtwo views16.73
137
1.28
78
11.61
105
3.72
117
1.65
59
16.67
116
9.17
119
14.39
92
55.91
146
63.81
146
40.86
141
35.94
139
37.73
145
14.24
107
26.87
138
0.05
68
0.10
91
0.13
109
0.22
104
0.12
45
0.15
61
SGM-ForestMtwo views16.99
138
1.08
62
5.74
60
2.12
59
0.75
29
31.63
138
12.21
131
27.80
141
32.25
130
37.88
138
39.99
139
52.96
145
35.20
144
33.60
141
24.47
133
0.26
99
0.39
122
0.31
125
0.39
119
0.26
79
0.53
104
DispFullNettwo views17.47
139
26.01
144
33.98
138
22.58
145
20.86
145
13.84
110
1.28
46
16.50
106
26.27
123
19.97
116
17.17
89
20.52
117
18.49
121
22.86
132
10.76
102
5.13
141
2.83
142
30.72
146
7.72
144
20.86
145
11.01
145
EDNetEfficientorigintwo views17.87
140
14.00
143
99.53
150
1.64
34
0.92
40
10.55
90
6.26
104
13.21
81
34.84
134
38.20
139
36.91
136
27.47
128
26.32
139
14.71
111
25.69
136
0.04
60
0.06
76
0.14
110
0.52
126
3.16
137
3.27
138
RTStwo views18.87
141
9.32
140
86.48
145
4.95
130
6.10
134
42.08
143
14.70
136
15.49
102
41.06
142
22.65
120
32.32
126
13.77
78
19.54
125
37.98
142
28.96
139
0.41
111
0.23
113
0.00
1
0.02
63
0.91
116
0.50
100
RTSAtwo views18.87
141
9.32
140
86.48
145
4.95
130
6.10
134
42.08
143
14.70
136
15.49
102
41.06
142
22.65
120
32.32
126
13.77
78
19.54
125
37.98
142
28.96
139
0.41
111
0.23
113
0.00
1
0.02
63
0.91
116
0.50
100
MANEtwo views19.47
143
1.27
75
5.07
49
4.69
128
5.55
131
30.49
137
9.94
124
34.01
144
37.27
139
44.13
143
51.57
147
52.51
144
40.41
146
33.58
140
24.81
134
0.89
128
0.86
133
1.11
138
9.72
145
0.38
91
1.06
121
BEATNet-Init1two views23.31
144
9.03
139
41.67
139
4.17
124
2.53
87
45.68
145
19.47
143
33.43
143
38.45
141
47.59
145
49.10
145
59.31
146
41.80
147
38.35
144
29.21
141
0.47
117
0.50
130
0.81
134
0.66
128
2.10
133
1.86
130
MADNet+two views27.07
145
33.84
145
90.97
147
20.14
144
7.47
141
48.43
146
47.10
145
35.43
145
36.46
136
20.11
117
30.05
122
25.29
123
35.08
143
45.50
146
50.28
146
2.13
136
2.00
140
1.19
140
0.76
130
4.71
139
4.43
139
PWCKtwo views30.53
146
44.32
146
47.25
141
29.76
146
7.23
138
40.78
142
27.10
144
44.73
146
44.32
144
47.31
144
36.37
135
47.16
143
26.05
137
41.26
145
31.87
143
21.83
146
4.03
144
29.50
145
4.67
143
27.17
146
7.80
143
DPSimNet_ROBtwo views53.45
147
64.73
147
44.39
140
53.97
147
45.39
147
53.66
147
54.83
146
55.15
147
57.87
147
64.16
147
50.83
146
63.40
147
53.34
148
46.45
147
65.81
147
63.13
147
26.54
147
57.94
147
51.11
147
45.52
147
50.69
147
MADNet++two views82.84
148
82.38
149
73.57
144
87.72
148
82.97
148
93.14
148
69.15
147
86.42
148
82.50
148
93.46
148
86.70
148
86.28
148
80.92
149
88.34
148
88.84
148
86.83
148
84.17
148
72.64
148
68.92
148
80.47
148
81.42
148
MEDIAN_ROBtwo views98.41
149
99.70
150
99.30
149
97.09
149
97.02
149
96.89
149
95.77
150
97.66
149
97.28
149
98.79
151
98.94
149
99.18
149
98.14
150
96.89
149
96.88
149
99.96
151
99.16
149
100.00
149
99.99
149
99.69
149
99.88
149
AVERAGE_ROBtwo views99.62
150
99.95
151
98.81
148
100.00
154
100.00
150
98.08
150
95.47
149
100.00
152
100.00
150
100.00
152
100.00
150
100.00
152
100.00
151
100.00
152
99.99
150
100.00
153
100.00
150
100.00
149
100.00
150
100.00
152
100.00
153
DGTPSM_ROBtwo views99.90
151
100.00
152
99.99
151
99.99
152
100.00
150
100.00
151
100.00
151
99.97
150
100.00
150
98.35
149
100.00
150
99.84
150
100.00
151
99.98
150
99.99
150
99.99
152
100.00
150
100.00
149
100.00
150
100.00
152
100.00
153
DPSMNet_ROBtwo views99.91
152
100.00
152
99.99
151
99.99
152
100.00
150
100.00
151
100.00
151
99.98
151
100.00
150
98.35
149
100.00
150
99.84
150
100.00
151
99.98
150
99.99
150
100.00
153
100.00
150
100.00
149
100.00
150
100.00
152
100.00
153
DPSM_ROBtwo views99.95
153
100.00
152
100.00
153
99.76
150
100.00
150
100.00
151
100.00
151
100.00
152
100.00
150
100.00
152
100.00
150
100.00
152
100.00
151
100.00
152
100.00
153
99.21
149
100.00
150
100.00
149
100.00
150
99.99
150
99.95
150
DPSMtwo views99.95
153
100.00
152
100.00
153
99.76
150
100.00
150
100.00
151
100.00
151
100.00
152
100.00
150
100.00
152
100.00
150
100.00
152
100.00
151
100.00
152
100.00
153
99.21
149
100.00
150
100.00
149
100.00
150
99.99
150
99.95
150
LSM0two views100.00
155
100.00
152
100.00
153
100.00
154
100.00
150
100.00
151
100.00
151
100.00
152
100.00
150
100.00
152
100.00
150
100.00
152
100.00
151
100.00
152
100.00
153
100.00
153
100.00
150
100.00
149
100.00
150
100.00
152
99.99
152
MSMDNettwo views1.26
8
ASD4two views81.57
148