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

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

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

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




Method Infoalllakes. 1llakes. 1ssand box 1lsand box 1sstora. room 1lstora. room 1sstora. room 2lstora. room 2sstora. room 2 1lstora. room 2 1sstora. room 2 2lstora. room 2 2sstora. room 3lstora. room 3stunnel 1ltunnel 1stunnel 2ltunnel 2stunnel 3ltunnel 3s
sorted bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort bysort by
CREStereotwo views1.09
1
0.51
11
2.16
1
0.11
1
0.24
21
1.67
8
0.03
7
4.24
6
6.91
42
0.80
5
1.73
4
0.44
7
0.64
7
1.58
7
0.61
5
0.00
1
0.00
1
0.00
1
0.00
1
0.01
3
0.02
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
TANstereotwo views1.32
2
0.79
20
2.93
5
0.37
8
0.11
7
1.81
16
0.01
1
3.56
3
4.36
14
4.53
23
4.96
17
0.41
6
0.68
8
1.22
3
0.59
3
0.01
33
0.00
1
0.00
1
0.00
1
0.00
1
0.03
2
sCroCo_RVCtwo views1.34
3
0.85
25
6.15
72
1.21
29
1.28
68
0.90
1
0.18
21
4.50
7
2.86
2
0.37
1
3.18
9
0.52
11
1.52
25
1.83
8
0.93
9
0.13
102
0.25
128
0.00
1
0.00
1
0.08
30
0.16
31
PMTNettwo views1.48
4
0.43
2
2.77
4
0.31
3
0.17
16
2.94
32
0.33
31
3.66
5
5.27
21
1.11
6
1.28
3
0.28
5
0.92
14
4.32
30
0.90
8
4.89
173
0.00
1
0.00
1
0.00
1
0.03
15
0.05
4
CroCo_RVCtwo views1.54
5
2.26
111
9.53
120
0.65
17
0.50
34
1.20
2
0.23
26
4.86
8
3.65
6
0.42
2
1.27
1
0.23
2
1.40
23
1.23
4
0.68
6
0.02
57
0.00
1
0.00
1
0.20
121
2.00
162
0.42
60
sAnonymous2two views1.54
5
2.26
111
9.53
120
0.65
17
0.50
34
1.20
2
0.23
26
4.86
8
3.65
6
0.42
2
1.27
1
0.23
2
1.40
23
1.23
4
0.68
6
0.02
57
0.00
1
0.00
1
0.20
121
2.00
162
0.42
60
Anonymoustwo views1.54
5
0.76
19
5.39
52
1.49
36
2.68
113
1.83
17
0.43
35
5.65
15
3.79
8
0.52
4
1.74
5
0.19
1
1.30
20
3.31
22
1.01
11
0.06
90
0.08
97
0.00
1
0.00
1
0.19
71
0.31
47
XX-TBDtwo views1.55
8
1.11
45
2.94
6
2.45
85
0.06
5
2.31
20
0.01
1
3.58
4
7.80
59
3.11
12
5.10
19
0.27
4
0.26
2
1.42
6
0.60
4
0.00
1
0.00
1
0.00
1
0.00
1
0.00
1
0.04
3
s12784htwo views1.70
9
0.50
9
3.07
9
0.45
13
0.01
1
1.69
10
0.02
5
3.09
1
8.24
65
3.82
18
7.53
41
0.49
9
0.32
3
1.05
1
0.52
1
0.00
1
0.00
1
0.00
1
3.03
175
0.01
3
0.08
7
CREStereo++_RVCtwo views1.70
9
0.50
9
3.07
9
0.45
13
0.01
1
1.69
10
0.02
5
3.09
1
8.24
65
3.82
18
7.53
41
0.49
9
0.32
3
1.05
1
0.52
1
0.00
1
0.00
1
0.00
1
3.03
175
0.01
3
0.08
7
iRaftStereo_RVCtwo views1.88
11
1.22
54
2.63
2
0.95
24
0.06
5
2.96
33
0.05
8
6.08
19
3.99
10
3.92
20
4.88
15
2.53
28
0.62
6
3.62
24
1.19
17
0.00
1
1.19
157
0.00
1
0.00
1
0.37
109
1.29
132
TestStereotwo views1.89
12
1.59
82
4.22
35
0.37
8
0.28
23
1.33
4
0.01
1
5.63
14
4.13
11
3.59
14
5.11
20
2.06
25
0.70
9
7.62
58
0.93
9
0.02
57
0.00
1
0.00
1
0.00
1
0.02
8
0.19
37
Anonymous3two views1.96
13
0.61
14
7.58
94
0.42
12
0.51
36
4.28
59
0.52
39
5.56
13
2.76
1
2.35
7
3.47
10
1.50
19
1.13
16
5.92
41
1.54
19
0.23
117
0.23
125
0.00
1
0.00
1
0.14
48
0.47
64
GMStereotwo views2.07
14
1.56
80
5.64
62
0.74
20
0.05
3
3.63
45
0.01
1
5.17
10
11.27
97
4.22
22
2.44
8
0.53
12
0.49
5
4.06
28
1.01
11
0.22
116
0.00
1
0.00
1
0.14
115
0.18
69
0.05
4
rafts_anoytwo views2.09
15
1.42
68
3.34
19
2.87
112
2.72
115
2.46
23
0.17
20
6.20
24
6.48
37
3.40
13
2.25
7
1.91
23
1.32
21
3.59
23
1.75
26
0.01
33
1.52
164
0.00
1
0.01
72
0.06
24
0.28
45
DIP-Stereotwo views2.12
16
0.98
38
3.94
28
0.20
2
0.11
7
4.73
67
0.13
15
9.27
47
4.40
15
6.46
48
5.03
18
0.45
8
0.72
10
3.79
25
1.94
28
0.00
1
0.01
51
0.00
1
0.00
1
0.05
23
0.24
42
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 views2.16
17
0.42
1
3.13
11
1.28
30
0.31
24
3.52
40
0.39
33
6.56
26
6.31
34
5.38
30
5.11
20
5.47
44
1.70
28
2.00
9
1.08
14
0.01
33
0.19
119
0.00
1
0.25
129
0.02
8
0.16
31
raftrobusttwo views2.16
17
1.31
59
2.76
3
2.38
78
3.33
128
2.72
26
0.51
38
5.34
12
4.64
16
6.13
44
3.98
11
1.63
20
2.51
36
3.02
21
1.64
21
0.00
1
0.76
148
0.00
1
0.02
80
0.14
48
0.27
43
raft+_RVCtwo views2.18
19
1.49
74
3.14
12
2.77
105
2.01
95
1.40
5
0.10
11
7.02
32
5.35
22
2.99
11
5.54
28
1.19
13
1.34
22
4.98
33
1.49
18
0.00
1
1.45
160
0.00
1
0.00
1
0.14
48
1.11
123
MSMDNettwo views2.20
20
0.46
5
3.49
21
1.48
35
0.75
44
3.69
47
0.52
39
6.91
31
3.97
9
5.62
35
5.26
22
5.77
51
1.73
30
2.61
18
1.09
15
0.01
33
0.20
122
0.00
1
0.23
126
0.02
8
0.15
25
CFNet-RSSMtwo views2.26
21
0.43
2
3.24
15
1.28
30
0.31
24
3.19
35
0.40
34
7.62
35
7.29
50
5.88
38
4.68
14
5.32
42
1.68
27
2.26
13
1.04
13
0.01
33
0.09
101
0.00
1
0.31
135
0.03
15
0.16
31
FENettwo views2.35
22
0.59
13
3.28
17
2.32
74
0.27
22
2.78
27
0.13
15
7.68
36
6.09
28
5.38
30
2.12
6
4.84
37
3.33
47
4.98
33
2.66
39
0.02
57
0.03
74
0.00
1
0.00
1
0.04
18
0.40
53
GwcNet-DCAtwo views2.35
22
0.96
34
5.51
55
0.39
10
0.19
19
1.72
15
0.25
28
6.06
18
9.33
81
6.31
46
5.42
26
1.44
18
4.14
62
2.22
12
2.28
34
0.01
33
0.00
1
0.00
1
0.00
1
0.04
18
0.66
94
RALCasStereoNettwo views2.39
24
1.73
88
3.59
22
2.38
78
2.69
114
2.84
30
0.18
21
5.29
11
4.94
18
2.45
9
5.62
29
2.22
26
0.79
13
7.48
57
1.63
20
0.00
1
1.75
165
0.00
1
0.00
1
0.14
48
2.03
150
222two views2.45
25
0.93
30
6.66
79
0.40
11
0.20
20
1.70
12
0.11
12
6.19
21
9.43
84
6.03
40
5.29
23
1.43
15
5.28
83
2.37
15
2.21
31
0.01
33
0.00
1
0.01
88
0.00
1
0.04
18
0.61
80
RAFT-Stereo + iAFFtwo views2.48
26
1.23
55
3.48
20
1.61
39
1.09
59
1.63
6
0.15
17
5.80
16
6.43
36
7.32
68
9.91
66
3.05
31
1.71
29
2.33
14
1.75
26
0.01
33
0.71
147
0.00
1
0.00
1
0.28
92
1.19
128
csctwo views2.54
27
0.93
30
5.51
55
0.32
4
0.14
11
1.70
12
0.11
12
6.19
21
12.69
108
6.03
40
5.29
23
1.43
15
5.28
83
2.37
15
2.21
31
0.01
33
0.00
1
0.01
88
0.00
1
0.02
8
0.61
80
cscssctwo views2.54
27
0.93
30
5.51
55
0.32
4
0.14
11
1.70
12
0.11
12
6.19
21
12.69
108
6.03
40
5.29
23
1.43
15
5.28
83
2.37
15
2.21
31
0.01
33
0.00
1
0.01
88
0.00
1
0.02
8
0.61
80
R-Stereo Traintwo views2.60
29
0.46
5
2.94
6
1.10
26
0.18
17
3.61
42
0.60
47
6.88
29
3.50
4
9.13
90
17.23
123
1.77
21
0.77
11
2.03
10
1.65
22
0.00
1
0.01
51
0.00
1
0.00
1
0.01
3
0.21
38
RAFT-Stereopermissivetwo views2.60
29
0.46
5
2.94
6
1.10
26
0.18
17
3.61
42
0.60
47
6.88
29
3.50
4
9.13
90
17.23
123
1.77
21
0.77
11
2.03
10
1.65
22
0.00
1
0.01
51
0.00
1
0.00
1
0.01
3
0.21
38
Lahav Lipson, Zachary Teed, and Jia Deng: RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matching. 3DV
RALAANettwo views2.68
31
1.71
86
4.73
42
2.79
107
0.80
48
3.30
38
0.15
17
7.92
37
5.92
27
5.79
37
6.06
31
3.51
33
1.13
16
6.76
46
2.28
34
0.00
1
0.28
130
0.00
1
0.10
108
0.27
90
0.13
18
GEStwo views2.69
32
0.72
16
5.34
49
1.82
54
0.05
3
4.51
65
1.04
67
10.39
59
4.18
12
5.53
34
6.72
34
1.92
24
3.30
44
5.16
36
2.67
40
0.06
90
0.01
51
0.06
120
0.01
72
0.20
74
0.14
21
HCRNettwo views2.74
33
0.83
22
4.19
32
0.45
13
3.31
127
2.18
18
0.22
24
7.59
34
5.76
26
6.08
43
4.89
16
5.21
41
2.74
39
8.04
63
3.10
44
0.00
1
0.02
63
0.00
1
0.00
1
0.10
39
0.09
9
EAI-Stereotwo views2.75
34
0.54
12
3.17
13
0.91
23
1.28
68
4.02
53
5.30
124
12.39
91
4.76
17
2.51
10
7.10
39
1.22
14
0.94
15
6.03
42
1.96
29
0.01
33
0.56
141
0.00
1
1.38
167
0.04
18
0.95
113
ACVNettwo views2.86
35
0.96
34
3.87
24
0.86
21
0.61
39
2.63
24
0.54
41
8.93
42
6.12
29
5.09
24
9.32
61
5.70
49
3.84
56
2.68
19
5.56
78
0.00
1
0.00
1
0.00
1
0.00
1
0.26
87
0.23
40
AFF-stereotwo views2.87
36
1.31
59
3.26
16
1.87
57
0.38
28
1.68
9
0.09
10
6.38
25
7.07
45
6.52
49
11.90
82
6.08
55
2.32
35
5.05
35
1.65
22
0.02
57
0.66
143
0.00
1
0.00
1
0.29
95
0.94
112
XX-Stereotwo views2.95
37
0.48
8
7.08
84
2.84
111
2.51
110
1.63
6
0.20
23
6.15
20
3.38
3
2.38
8
16.00
119
6.90
57
0.25
1
6.40
44
1.14
16
0.00
1
0.66
143
0.00
1
0.31
135
0.16
61
0.57
73
DN-CSS_ROBtwo views3.00
38
2.14
106
7.33
86
2.67
101
0.78
46
3.13
34
0.07
9
6.68
27
4.22
13
5.51
33
12.12
84
2.33
27
1.21
18
8.69
74
2.51
37
0.03
69
0.00
1
0.00
1
0.00
1
0.42
112
0.15
25
111two views3.05
39
1.41
67
5.51
55
0.32
4
0.14
11
2.83
29
0.68
55
6.76
28
14.21
118
3.59
14
9.11
59
5.50
45
3.30
44
4.05
26
2.43
36
0.01
33
0.00
1
0.00
1
0.00
1
0.13
45
1.02
118
DMCAtwo views3.08
40
1.10
43
6.20
73
1.82
54
0.90
53
3.80
51
0.55
43
9.76
51
5.42
23
6.56
51
6.47
33
7.06
62
2.09
32
4.05
26
5.46
77
0.00
1
0.01
51
0.00
1
0.00
1
0.11
42
0.16
31
HITNettwo views3.11
41
1.38
66
5.35
51
2.23
66
0.12
9
5.59
76
0.58
45
9.95
53
5.72
25
6.62
53
7.91
47
3.90
34
3.24
43
4.39
31
4.58
67
0.01
33
0.00
1
0.00
1
0.00
1
0.08
30
0.57
73
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
pcwnet_v2two views3.29
42
0.62
15
4.21
34
2.70
102
1.12
62
2.78
27
0.86
65
11.96
85
6.91
42
6.98
60
12.22
88
2.74
30
3.62
50
4.68
32
3.54
49
0.06
90
0.06
89
0.00
1
0.00
1
0.06
24
0.62
83
ARAFTtwo views3.32
43
1.74
89
7.92
98
0.64
16
0.15
14
4.45
64
2.89
90
8.20
39
6.30
33
9.23
95
7.12
40
5.58
46
1.24
19
8.60
70
1.69
25
0.01
33
0.03
74
0.00
1
0.00
1
0.35
107
0.35
50
AdaStereotwo views3.34
44
0.74
18
4.00
30
3.10
122
0.51
36
4.22
57
1.25
71
12.84
98
8.39
69
6.33
47
9.55
64
5.77
51
1.54
26
5.53
37
2.74
41
0.02
57
0.00
1
0.00
1
0.00
1
0.03
15
0.15
25
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.
UCFNet_RVCtwo views3.37
45
1.25
56
3.30
18
2.55
96
1.80
86
3.92
52
1.28
73
12.49
94
7.16
47
5.12
27
7.88
46
8.14
75
4.59
69
5.67
39
2.11
30
0.00
1
0.00
1
0.00
1
0.00
1
0.14
48
0.07
6
CFNet-pseudotwo views3.45
46
0.73
17
4.72
41
2.44
83
1.45
75
3.78
48
0.68
55
13.28
106
10.00
86
5.73
36
12.49
90
2.60
29
1.81
31
5.65
38
3.39
48
0.01
33
0.00
1
0.00
1
0.00
1
0.06
24
0.13
18
test_xeample3two views3.59
47
1.29
57
5.53
60
0.36
7
0.12
9
3.48
39
3.85
105
7.44
33
12.69
108
6.81
56
13.53
99
3.50
32
4.14
62
4.11
29
2.59
38
0.01
33
0.02
63
0.00
1
0.00
1
0.02
8
2.34
157
DMCA-RVCcopylefttwo views3.59
47
1.98
102
8.68
107
2.30
72
1.48
76
3.54
41
0.55
43
8.40
40
5.43
24
9.37
96
5.90
30
8.34
78
3.21
41
7.19
50
4.46
63
0.06
90
0.11
104
0.08
127
0.02
80
0.21
78
0.40
53
GANet-RSSMtwo views3.67
49
1.16
48
4.56
39
3.24
128
2.03
96
2.63
24
0.50
37
11.53
82
6.83
39
6.25
45
10.58
72
7.35
64
6.65
103
6.77
47
2.94
43
0.00
1
0.00
1
0.08
127
0.00
1
0.26
87
0.12
14
NOSS_ROBtwo views3.67
49
0.86
27
3.81
23
2.53
93
1.11
60
6.10
82
0.72
60
11.08
72
12.36
102
5.45
32
8.91
57
5.65
47
2.09
32
8.30
67
3.16
46
0.20
113
0.07
96
0.00
1
0.00
1
0.11
42
0.80
106
PSMNet-RSSMtwo views3.69
51
1.36
64
4.54
38
2.44
83
1.90
90
2.35
21
0.69
57
11.77
84
6.38
35
6.60
52
11.85
81
7.70
71
5.61
92
6.66
45
3.67
50
0.00
1
0.00
1
0.02
101
0.01
72
0.17
66
0.10
10
BEATNet_4xtwo views3.69
51
2.12
105
8.11
100
1.82
54
0.16
15
5.30
72
1.04
67
10.87
67
6.14
31
7.09
63
8.42
54
4.99
39
4.42
66
5.89
40
6.23
84
0.04
78
0.05
84
0.00
1
0.00
1
0.32
103
0.84
107
CFNet_RVCtwo views3.70
53
1.59
82
3.87
24
1.68
44
2.42
104
3.20
36
0.66
53
8.92
41
7.76
56
5.09
24
9.95
67
10.77
99
5.43
88
7.41
53
4.71
69
0.00
1
0.00
1
0.03
110
0.01
72
0.30
97
0.11
12
CFNet-ftpermissivetwo views3.70
53
1.59
82
3.87
24
1.68
44
2.42
104
3.20
36
0.66
53
8.93
42
7.76
56
5.09
24
9.95
67
10.77
99
5.43
88
7.41
53
4.71
69
0.00
1
0.00
1
0.03
110
0.01
72
0.30
97
0.11
12
cf-rtwo views3.76
55
1.43
69
5.34
49
2.66
99
2.79
116
2.85
31
0.62
51
9.08
45
6.91
42
7.06
61
11.90
82
9.64
86
3.83
55
6.23
43
4.53
65
0.00
1
0.00
1
0.00
1
0.00
1
0.19
71
0.17
35
MLCVtwo views3.76
55
1.52
77
6.93
82
1.66
43
0.31
24
4.31
61
0.32
30
7.92
37
7.68
55
9.55
97
12.12
84
5.17
40
2.74
39
10.00
99
4.49
64
0.01
33
0.00
1
0.00
1
0.00
1
0.16
61
0.27
43
DeepPruner_ROBtwo views3.82
57
1.87
93
5.65
63
1.31
32
1.64
77
3.62
44
0.81
63
14.47
119
5.03
20
8.03
75
10.78
74
6.95
59
2.26
34
8.92
78
4.32
61
0.07
95
0.03
74
0.00
1
0.01
72
0.37
109
0.29
46
STTStereotwo views3.90
58
1.43
69
8.20
102
3.03
118
2.28
100
3.64
46
0.64
52
10.11
54
7.09
46
10.20
102
7.08
38
8.73
83
3.57
49
7.44
55
4.00
56
0.00
1
0.02
63
0.01
88
0.01
72
0.04
18
0.40
53
acv_fttwo views3.90
58
0.96
34
5.85
65
2.56
97
4.70
158
7.27
97
0.84
64
9.17
46
6.12
29
10.00
101
9.32
61
5.70
49
4.07
60
2.68
19
8.24
99
0.00
1
0.00
1
0.00
1
0.00
1
0.35
107
0.23
40
GEStereo_RVCtwo views3.95
60
1.21
53
6.13
71
2.30
72
1.22
66
4.29
60
2.96
92
12.39
91
8.24
65
7.25
65
6.79
35
7.58
70
4.22
64
9.16
83
4.56
66
0.04
78
0.01
51
0.02
101
0.02
80
0.19
71
0.42
60
hitnet-ftcopylefttwo views3.95
60
1.16
48
3.92
27
0.89
22
0.73
43
4.88
69
0.54
41
9.78
52
7.91
62
7.42
69
10.24
69
12.50
111
4.84
75
8.15
65
5.56
78
0.03
69
0.00
1
0.06
120
0.00
1
0.24
85
0.12
14
ac_64two views3.98
62
0.86
27
5.10
46
3.17
123
3.43
132
5.80
78
0.70
58
10.87
67
4.96
19
8.94
86
6.46
32
10.70
97
5.52
90
7.44
55
5.10
75
0.03
69
0.00
1
0.00
1
0.00
1
0.14
48
0.31
47
CFNettwo views4.03
63
1.97
101
6.30
76
2.70
102
1.64
77
5.41
74
0.22
24
12.14
88
10.28
87
5.24
29
11.15
77
6.43
56
3.70
52
8.97
81
4.01
57
0.00
1
0.00
1
0.00
1
0.00
1
0.30
97
0.13
18
iResNettwo views4.04
64
1.46
73
8.72
109
3.25
129
0.36
27
4.54
66
0.47
36
9.06
44
10.40
90
7.98
72
12.82
94
4.06
35
3.22
42
8.82
76
4.95
72
0.03
69
0.01
51
0.00
1
0.00
1
0.14
48
0.55
71
NLCA_NET_v2_RVCtwo views4.11
65
1.53
79
6.87
81
3.04
119
3.33
128
4.34
63
0.59
46
11.36
77
7.92
63
8.98
87
7.98
48
9.72
88
3.82
54
7.92
61
4.43
62
0.15
104
0.02
63
0.03
110
0.03
88
0.07
27
0.14
21
Zhibo Rao, Mingyi He, Yuchao Dai, Zhidong Zhu, Bo Li, and Renjie He.: NLCA-Net: A non-local context attention network for stereo matching.
FADNet-RVC-Resampletwo views4.11
65
2.25
109
13.88
139
1.64
41
0.66
40
5.91
81
2.34
86
10.87
67
9.09
78
6.53
50
4.34
13
5.69
48
4.71
73
8.01
62
4.64
68
0.18
110
0.22
123
0.17
144
0.12
112
0.46
118
0.59
75
GwcNet-RSSMtwo views4.13
67
1.71
86
6.25
74
2.50
89
1.93
93
2.39
22
1.49
76
9.59
50
8.71
73
6.66
54
15.04
109
8.47
81
3.84
56
6.87
48
6.74
87
0.00
1
0.00
1
0.04
115
0.00
1
0.29
95
0.15
25
UPFNettwo views4.17
68
0.83
22
5.51
55
2.49
88
3.36
130
7.13
95
2.60
87
10.52
62
7.16
47
8.76
82
7.83
44
7.85
73
3.41
48
8.60
70
7.04
91
0.01
33
0.02
63
0.00
1
0.00
1
0.12
44
0.17
35
UNettwo views4.19
69
0.87
29
7.18
85
2.00
60
3.22
125
9.55
111
0.31
29
10.11
54
6.22
32
7.98
72
6.82
36
10.42
93
5.77
93
7.28
51
5.83
80
0.00
1
0.00
1
0.00
1
0.00
1
0.10
39
0.12
14
PS-NSSStwo views4.22
70
2.53
125
5.89
66
1.01
25
2.93
117
4.05
55
1.93
83
11.44
78
8.94
76
7.14
64
14.70
108
4.96
38
2.60
37
9.09
82
3.92
54
0.40
134
0.04
79
0.83
164
0.00
1
1.09
145
1.02
118
FADNet_RVCtwo views4.24
71
2.32
118
14.23
142
1.18
28
0.79
47
5.64
77
0.60
47
11.50
80
7.19
49
3.68
16
8.08
52
4.14
36
4.06
59
10.29
104
7.47
93
0.36
132
0.44
138
0.18
145
0.47
149
0.93
143
1.23
129
HSMtwo views4.25
72
1.10
43
4.14
31
1.78
51
2.14
97
6.67
90
1.08
69
12.92
99
6.90
41
7.06
61
8.17
53
14.17
119
6.03
96
9.16
83
3.11
45
0.00
1
0.00
1
0.00
1
0.00
1
0.09
36
0.41
58
psm_uptwo views4.33
73
1.34
63
4.43
36
2.27
71
2.62
112
5.29
71
3.30
103
11.49
79
6.89
40
8.05
77
8.06
50
10.09
91
11.10
127
7.12
49
3.76
52
0.24
119
0.01
51
0.00
1
0.00
1
0.20
74
0.37
51
FADNet-RVCtwo views4.36
74
2.62
130
14.51
145
2.01
61
0.44
32
4.27
58
1.27
72
12.61
96
7.89
60
3.76
17
4.05
12
10.63
96
4.34
65
9.87
95
7.66
94
0.03
69
0.09
101
0.00
1
0.03
88
0.65
130
0.50
66
HSM-Net_RVCpermissivetwo views4.40
75
0.44
4
3.22
14
0.72
19
0.68
41
7.06
93
1.82
81
12.57
95
8.48
70
9.04
88
12.20
87
14.98
122
5.14
79
7.32
52
4.19
58
0.02
57
0.02
63
0.00
1
0.00
1
0.02
8
0.14
21
Gengshan Yang, Joshua Manela, Michael Happold, and Deva Ramanan: Hierarchical Deep Stereo Matching on High-resolution Images. CVPR 2019
MMNettwo views4.44
76
1.37
65
6.26
75
1.73
50
1.64
77
10.51
123
0.61
50
9.27
47
9.66
85
9.55
97
9.22
60
7.40
66
6.52
101
7.90
60
6.98
89
0.01
33
0.00
1
0.00
1
0.00
1
0.14
48
0.10
10
DSFCAtwo views4.48
77
1.03
41
6.84
80
1.61
39
1.68
82
7.26
96
4.46
117
11.27
76
7.51
52
11.01
112
6.89
37
8.49
82
4.54
67
10.41
106
6.18
83
0.03
69
0.06
89
0.02
101
0.03
88
0.13
45
0.14
21
FADNettwo views4.55
78
2.35
119
13.60
137
1.87
57
0.80
48
4.74
68
0.75
61
14.47
119
12.41
103
4.07
21
5.45
27
9.64
86
5.03
77
8.86
77
4.82
71
0.20
113
0.04
79
0.07
123
0.05
98
1.27
152
0.40
53
HGLStereotwo views4.58
79
1.04
42
6.02
70
3.30
132
5.14
162
6.36
85
0.71
59
11.04
71
7.60
54
8.04
76
12.85
95
6.90
57
8.51
118
7.85
59
5.93
81
0.02
57
0.02
63
0.00
1
0.00
1
0.14
48
0.15
25
TDLMtwo views4.58
79
1.79
92
5.03
45
2.45
85
1.32
72
4.04
54
7.15
145
11.60
83
11.46
98
6.91
58
12.69
93
5.94
53
5.03
77
9.74
91
3.80
53
0.67
149
0.00
1
0.01
88
0.05
98
0.23
83
1.78
147
delettwo views4.59
81
1.12
47
5.93
67
1.37
33
3.73
139
10.48
122
3.90
106
10.51
61
9.17
79
8.87
85
9.74
65
8.37
79
3.74
53
8.12
64
6.36
86
0.00
1
0.00
1
0.00
1
0.00
1
0.21
78
0.12
14
RAFT + AFFtwo views4.63
82
2.28
114
7.43
91
1.51
37
2.38
102
5.81
79
6.49
136
10.74
64
8.53
71
7.75
71
8.95
58
7.47
68
6.68
105
9.17
85
5.06
74
0.08
96
0.47
140
0.00
1
0.01
72
0.66
131
1.03
121
CBMV_ROBtwo views4.66
83
0.85
25
3.94
28
1.64
41
0.75
44
8.38
103
1.90
82
11.19
73
10.36
88
9.20
94
15.21
110
7.28
63
6.99
107
9.48
88
5.24
76
0.04
78
0.05
84
0.00
1
0.00
1
0.07
27
0.64
86
iResNet_ROBtwo views4.67
84
1.77
90
5.93
67
2.47
87
0.95
54
3.78
48
0.36
32
15.65
129
17.36
131
8.75
81
11.23
78
7.05
61
4.60
70
8.94
79
4.23
59
0.01
33
0.00
1
0.00
1
0.00
1
0.16
61
0.15
25
CVANet_RVCtwo views4.68
85
1.93
99
5.57
61
2.61
98
1.86
89
4.12
56
4.94
121
12.11
86
9.29
80
8.05
77
11.08
76
10.28
92
6.48
100
9.52
89
3.19
47
0.45
138
0.01
51
0.00
1
0.12
112
0.23
83
1.75
146
iResNetv2_ROBtwo views4.72
86
2.26
111
8.87
112
3.20
126
1.25
67
4.32
62
1.73
79
14.22
117
10.62
92
10.53
106
12.13
86
7.77
72
5.54
91
8.29
66
2.76
42
0.00
1
0.00
1
0.00
1
0.00
1
0.42
112
0.55
71
NVstereo2Dtwo views4.82
87
1.29
57
7.98
99
3.68
141
3.48
133
8.02
99
3.10
96
11.25
74
15.50
123
5.14
28
7.71
43
8.21
77
4.68
71
10.19
103
4.30
60
0.02
57
0.19
119
0.13
135
0.36
142
0.07
27
1.12
125
StereoDRNet-Refinedtwo views4.84
88
0.96
34
5.13
47
2.22
65
0.46
33
9.10
107
0.16
19
10.72
63
9.40
83
13.41
129
12.50
91
10.02
89
4.58
68
8.60
70
8.71
106
0.01
33
0.00
1
0.01
88
0.05
98
0.24
85
0.53
69
Rohan Chabra, Julian Straub, Chris Sweeney, Richard Newcombe, Henry Fuchs: StereoDRNet. CVPR
DLCB_ROBtwo views4.87
89
1.45
72
4.75
43
2.52
92
1.13
64
6.25
84
3.01
94
10.44
60
8.34
68
11.05
113
13.43
97
14.04
117
3.90
58
9.74
91
6.80
88
0.00
1
0.00
1
0.00
1
0.00
1
0.08
30
0.37
51
RASNettwo views5.00
90
1.19
51
5.69
64
3.68
141
4.80
159
5.04
70
1.29
74
10.26
56
8.99
77
5.89
39
11.65
79
10.55
95
9.76
123
9.96
98
10.32
124
0.20
113
0.00
1
0.00
1
0.00
1
0.09
36
0.73
100
SGM-Foresttwo views5.40
91
0.84
24
4.19
32
1.43
34
0.68
41
10.12
119
6.41
135
12.13
87
11.86
100
11.50
115
14.43
105
11.62
105
5.20
80
9.93
97
6.16
82
0.16
106
0.04
79
0.00
1
0.00
1
0.08
30
1.24
130
Johannes L. Schönberger, Sudipta Sinha, Marc Pollefeys: Learning to Fuse Proposals from Multiple Scanline Optimizations in Semi-Global Matching. ECCV 2018
PA-Nettwo views5.40
91
2.35
119
9.55
122
2.75
104
2.21
98
8.74
104
3.92
107
13.05
102
13.35
114
7.30
67
7.98
48
8.08
74
6.39
99
11.17
112
9.30
113
0.02
57
0.03
74
0.00
1
0.03
88
0.16
61
1.55
139
Zhibo Rao, Mingyi He, Yuchao Dai, Zhelun Shen: Patch Attention Network with Generative Adversarial Model for Semi-Supervised Binocular Disparity Prediction.
AANet_RVCtwo views5.41
93
2.61
129
8.17
101
2.25
68
1.30
70
2.27
19
1.64
77
10.82
65
18.54
138
8.01
74
18.09
128
8.39
80
3.31
46
11.37
114
8.98
110
1.15
157
0.38
134
0.02
101
0.00
1
0.17
66
0.70
97
PSMNet_ROBtwo views5.41
93
2.45
122
7.39
89
2.36
76
1.90
90
10.11
118
6.54
137
16.32
135
7.59
53
6.84
57
10.83
75
12.06
109
4.13
61
8.73
75
10.24
122
0.02
57
0.01
51
0.01
88
0.11
110
0.26
87
0.40
53
CBMVpermissivetwo views5.97
95
1.52
77
4.51
37
1.99
59
0.53
38
10.02
117
6.94
142
13.27
105
13.80
117
13.08
127
15.93
117
11.36
104
8.11
114
9.82
93
7.27
92
0.08
96
0.15
113
0.00
1
0.00
1
0.30
97
0.74
102
Konstantinos Batsos, Changjiang Cai, Philippos Mordohai: CBMV: A Coalesced Bidirectional Matching Volume for Disparity Estimation. Computer Vision and Pattern Recognition (CVPR) 2018
StereoDRNettwo views6.02
96
2.40
121
8.22
103
3.43
134
4.43
149
10.57
124
4.21
110
19.44
152
11.24
96
12.86
122
8.06
50
11.64
106
4.79
74
8.65
73
9.32
114
0.01
33
0.10
103
0.00
1
0.03
88
0.34
106
0.69
96
DRN-Testtwo views6.23
97
1.51
75
7.53
92
3.07
121
3.69
138
12.17
134
3.23
100
20.80
162
11.11
95
12.64
121
13.02
96
10.91
103
5.26
81
8.49
68
10.12
120
0.05
86
0.05
84
0.04
115
0.04
97
0.28
92
0.65
87
NCCL2two views6.27
98
2.46
123
7.00
83
2.36
76
1.11
60
9.48
110
11.14
159
13.01
101
10.40
90
9.94
100
9.49
63
22.11
155
7.42
110
9.90
96
8.56
102
0.01
33
0.00
1
0.02
101
0.03
88
0.20
74
0.73
100
ETE_ROBtwo views6.28
99
2.60
128
9.22
117
1.71
48
0.84
51
6.44
86
6.78
139
13.16
103
8.80
74
13.01
125
15.93
117
21.01
153
6.18
97
10.97
108
8.11
98
0.04
78
0.01
51
0.00
1
0.02
80
0.16
61
0.70
97
APVNettwo views6.28
99
2.46
123
10.09
125
3.51
136
3.97
143
12.15
133
1.75
80
20.08
158
6.73
38
6.92
59
13.45
98
10.75
98
6.29
98
11.86
119
14.34
146
0.05
86
0.02
63
0.11
129
0.03
88
0.63
128
0.34
49
MaskLacGwcNet_RVCtwo views6.42
101
6.12
163
6.53
78
4.29
153
3.61
136
15.41
144
4.13
108
12.36
89
7.32
51
7.60
70
8.51
55
8.97
84
2.64
38
10.17
102
3.72
51
8.16
179
2.40
171
2.00
174
0.11
110
8.55
178
5.84
173
DANettwo views6.46
102
1.58
81
9.58
124
4.10
147
3.99
144
9.13
108
1.34
75
10.37
58
7.76
56
13.76
133
15.66
113
16.14
127
7.03
108
13.07
126
13.62
143
0.19
111
0.12
107
0.02
101
0.02
80
0.15
60
1.61
140
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 views6.46
102
1.88
94
7.41
90
2.81
108
0.95
54
6.45
87
6.82
140
13.58
109
8.80
74
15.50
138
15.86
115
19.59
147
7.12
109
11.41
115
9.86
118
0.02
57
0.04
79
0.00
1
0.03
88
0.20
74
0.79
105
NaN_ROBtwo views6.51
104
2.25
109
8.77
110
1.69
46
1.79
85
9.71
114
10.94
157
16.20
133
16.13
125
13.04
126
8.86
56
11.77
107
5.94
95
12.88
125
8.01
96
0.05
86
0.13
110
0.07
123
0.20
121
0.32
103
1.50
136
DISCOtwo views6.58
105
0.98
38
6.44
77
3.75
143
1.18
65
11.19
126
3.29
102
12.92
99
16.72
127
7.27
66
11.74
80
17.34
132
6.60
102
19.59
160
11.59
133
0.00
1
0.00
1
0.00
1
0.00
1
0.45
117
0.50
66
Syn2CoExtwo views6.70
106
3.69
149
12.14
133
2.78
106
1.68
82
8.36
102
3.21
99
18.59
147
12.66
106
8.86
84
20.79
134
10.52
94
10.54
124
9.68
90
8.70
105
0.01
33
0.00
1
0.00
1
0.00
1
0.69
134
1.00
116
RYNettwo views6.82
107
1.51
75
7.35
88
1.79
52
4.52
153
16.68
148
4.41
116
14.19
116
17.43
132
11.64
116
7.86
45
15.03
123
8.81
119
10.10
100
14.16
145
0.04
78
0.13
110
0.15
138
0.05
98
0.08
30
0.53
69
S-Stereotwo views6.85
108
0.93
30
10.76
129
3.34
133
3.51
134
6.55
88
6.01
129
13.75
113
23.86
153
9.65
99
15.54
112
11.81
108
8.24
115
8.59
69
11.23
131
0.13
102
0.16
114
0.00
1
0.42
146
0.14
48
2.39
158
GANettwo views6.86
109
1.78
91
5.98
69
3.05
120
1.12
62
10.12
119
9.38
154
13.41
108
8.67
72
11.46
114
29.87
155
13.94
116
8.25
116
12.06
120
6.33
85
0.40
134
0.06
89
0.16
141
0.33
139
0.18
69
0.59
75
GwcNetcopylefttwo views6.95
110
2.89
136
13.32
136
2.82
109
5.53
164
11.76
129
2.14
85
14.95
123
17.69
134
10.83
110
18.14
129
10.83
101
6.66
104
9.83
94
10.19
121
0.03
69
0.06
89
0.03
110
0.08
104
0.64
129
0.66
94
FINETtwo views6.96
111
2.98
137
13.77
138
3.20
126
4.81
160
9.65
112
9.90
156
13.68
112
23.91
154
8.28
80
13.70
100
5.36
43
5.83
94
10.79
107
8.05
97
0.90
153
1.43
159
0.11
129
0.38
143
0.48
123
1.98
149
GANetREF_RVCpermissivetwo views6.97
112
3.59
146
9.14
116
4.00
144
0.40
29
13.17
137
11.27
161
15.95
131
17.51
133
10.62
107
10.66
73
12.15
110
7.97
113
13.18
127
5.04
73
0.45
138
0.00
1
0.01
88
0.15
116
3.47
170
0.60
77
Zhang, Feihu and Prisacariu, Victor and Yang, Ruigang and Torr, Philip HS: GA-Net: Guided Aggregation Net for End- to-end Stereo Matching. CVPR 2019
FAT-Stereotwo views6.98
113
0.79
20
7.57
93
2.54
95
1.80
86
6.16
83
4.33
115
19.16
151
18.82
139
10.75
109
17.08
122
18.23
138
13.56
140
9.25
86
8.44
100
0.03
69
0.16
114
0.01
88
0.10
108
0.14
48
0.65
87
LALA_ROBtwo views7.02
114
2.54
126
8.51
104
1.53
38
0.96
56
9.96
116
8.71
152
16.70
136
12.68
107
12.99
124
13.90
103
24.47
157
5.40
87
11.61
117
9.50
115
0.04
78
0.05
84
0.00
1
0.02
80
0.17
66
0.65
87
DGSMNettwo views7.10
115
3.49
143
15.11
147
4.36
155
4.41
148
8.94
106
6.25
133
15.10
124
14.28
119
8.15
79
14.51
107
10.08
90
5.38
86
11.04
109
7.95
95
1.49
162
2.00
169
1.79
173
1.68
173
2.43
164
3.48
168
edge stereotwo views7.12
116
1.19
51
7.83
97
2.32
74
2.46
106
6.90
91
3.26
101
15.58
128
12.52
104
15.68
139
18.34
130
21.83
154
13.05
136
11.53
116
8.68
104
0.04
78
0.06
89
0.11
129
0.03
88
0.30
97
0.63
84
DeepPrunerFtwo views7.18
117
3.58
145
25.29
165
4.27
152
7.20
171
3.78
48
4.64
118
14.03
114
20.86
147
8.85
83
10.57
71
7.35
64
7.88
112
12.16
123
10.66
127
0.25
121
0.16
114
0.29
154
0.43
147
0.77
140
0.65
87
Anonymous Stereotwo views7.27
118
4.41
156
26.56
167
3.67
140
3.09
122
5.56
75
15.31
167
10.85
66
11.63
99
10.31
103
10.32
70
7.46
67
3.67
51
14.42
136
15.13
149
0.64
147
0.88
150
0.00
1
0.07
103
0.33
105
1.16
126
RGCtwo views7.40
119
2.77
132
7.71
95
4.43
157
4.87
161
8.88
105
2.96
92
16.05
132
12.76
111
11.96
119
19.57
132
18.32
139
11.57
132
13.89
134
9.58
117
0.83
152
0.02
63
0.24
152
0.41
145
0.40
111
0.77
104
RPtwo views7.41
120
1.90
97
7.33
86
4.26
151
5.25
163
6.64
89
3.46
104
12.44
93
16.23
126
10.36
104
23.00
140
18.72
140
16.59
149
11.10
110
8.47
101
1.10
156
0.02
63
0.12
133
0.16
117
0.43
114
0.65
87
Abc-Nettwo views7.43
121
2.28
114
8.58
105
3.28
130
4.48
150
8.34
100
4.27
112
20.45
160
13.66
115
9.17
92
17.52
125
16.49
128
14.31
145
13.71
130
8.95
108
1.73
164
0.12
107
0.01
88
0.50
151
0.21
78
0.60
77
Xing Li, Yangyu Fan, Guoyun Lv, and Haoyue Ma: Area-based Correlation and Non-local Attention Network for Stereo Matching. The Visual Computer
NCC-stereotwo views7.43
121
2.28
114
8.58
105
3.28
130
4.48
150
8.34
100
4.27
112
20.45
160
13.66
115
9.17
92
17.52
125
16.49
128
14.31
145
13.71
130
8.95
108
1.73
164
0.12
107
0.01
88
0.50
151
0.21
78
0.60
77
ccnettwo views7.45
123
1.89
96
7.71
95
3.18
124
2.35
101
12.19
135
6.14
132
18.95
149
9.36
82
13.33
128
20.38
133
16.00
126
9.10
122
10.10
100
15.40
150
0.44
136
0.34
133
0.25
153
0.25
129
0.48
123
1.18
127
Nwc_Nettwo views7.54
124
1.91
98
8.71
108
4.16
148
3.42
131
10.77
125
2.00
84
20.33
159
12.19
101
9.10
89
24.04
144
17.67
134
13.58
141
12.14
122
9.12
111
0.94
155
0.00
1
0.00
1
0.02
80
0.22
82
0.41
58
psmorigintwo views7.74
125
2.19
108
18.99
155
2.51
90
0.89
52
7.08
94
2.74
89
11.25
74
7.90
61
20.49
151
22.36
138
20.16
148
11.95
133
13.84
132
10.61
126
0.04
78
0.18
117
0.06
120
0.24
127
0.27
90
1.07
122
ADCReftwo views7.76
126
2.31
117
19.09
156
2.82
109
3.29
126
11.88
130
3.06
95
11.52
81
10.89
94
13.69
132
25.85
149
8.20
76
8.89
121
8.95
80
22.66
159
0.16
106
0.08
97
0.16
141
0.34
140
0.47
120
0.91
110
CC-Net-ROBtwo views8.09
127
10.11
173
9.03
114
4.07
145
3.74
140
15.24
143
4.84
120
18.44
146
10.87
93
6.73
55
14.03
104
7.54
69
4.68
71
11.61
117
3.98
55
6.59
176
2.54
174
16.51
179
1.19
163
7.88
177
2.20
154
pmcnntwo views8.17
128
1.70
85
11.49
131
3.18
124
3.07
121
9.66
113
6.06
131
13.60
110
17.71
135
15.23
137
24.02
142
27.61
163
8.81
119
10.36
105
9.51
116
0.08
96
0.08
97
0.00
1
0.00
1
0.31
102
0.87
108
stereogantwo views8.18
129
1.32
62
9.05
115
3.59
137
4.00
145
20.16
158
3.13
97
17.25
141
20.16
144
10.68
108
19.19
131
20.56
152
10.70
126
14.74
140
7.00
90
0.34
130
0.04
79
0.21
149
0.08
104
0.70
136
0.65
87
SuperBtwo views8.29
130
3.96
152
24.92
163
2.96
115
1.33
73
9.83
115
4.15
109
10.92
70
29.79
163
11.86
117
12.51
92
6.07
54
6.97
106
11.13
111
14.73
147
0.15
104
0.11
104
0.35
156
0.26
132
12.64
179
1.11
123
AF-Nettwo views8.37
131
2.16
107
9.01
113
3.65
139
4.58
156
9.40
109
2.63
88
17.71
143
21.32
149
10.38
105
24.51
145
20.43
149
16.67
150
12.54
124
10.76
129
0.70
150
0.00
1
0.12
133
0.00
1
0.47
120
0.42
60
CSANtwo views8.41
132
2.86
135
9.35
119
2.43
82
0.82
50
13.59
139
11.68
162
15.80
130
21.98
151
14.71
135
16.46
120
18.00
136
14.42
147
13.68
129
10.10
119
0.10
99
0.13
110
0.15
138
0.19
119
0.43
114
1.28
131
PWCDC_ROBbinarytwo views8.50
133
4.40
155
9.57
123
6.16
169
4.54
155
10.46
121
1.03
66
15.30
125
28.16
160
10.88
111
30.88
159
6.96
60
11.31
129
14.57
138
10.40
125
3.10
172
0.02
63
0.00
1
0.00
1
1.55
156
0.63
84
G-Nettwo views8.60
134
1.93
99
12.59
134
6.04
167
3.66
137
25.93
167
4.26
111
10.35
57
8.05
64
19.71
148
24.73
146
14.15
118
15.49
148
11.30
113
10.69
128
0.50
145
0.46
139
0.19
147
0.25
129
0.80
141
0.91
110
ADCP+two views8.60
134
2.78
133
17.44
153
1.69
46
4.27
147
16.95
151
5.29
123
13.38
107
12.52
104
13.57
131
17.57
127
10.86
102
11.28
128
18.43
156
23.91
165
0.03
69
0.05
84
0.01
88
0.18
118
0.47
120
1.31
133
PWC_ROBbinarytwo views8.89
136
4.49
157
15.62
148
2.92
113
4.49
152
7.89
98
1.23
70
17.29
142
20.00
142
16.54
141
28.06
152
14.37
120
11.98
135
14.93
141
15.43
151
0.47
142
0.00
1
0.05
118
0.00
1
0.67
132
1.31
133
aanetorigintwo views9.02
137
3.75
150
28.68
169
2.16
63
3.86
141
5.84
80
5.40
126
5.93
17
10.39
89
28.12
165
31.02
160
15.98
125
13.74
142
9.38
87
12.55
139
0.25
121
0.22
123
0.22
150
0.24
127
1.24
150
1.47
135
PASMtwo views9.11
138
5.68
162
25.06
164
4.33
154
4.18
146
7.01
92
8.48
150
12.37
90
20.95
148
13.49
130
15.37
111
18.80
141
11.52
131
14.70
139
12.70
140
0.93
154
1.19
157
0.32
155
1.21
164
1.66
157
2.22
156
MDST_ROBtwo views9.13
139
1.31
59
10.16
127
4.36
155
2.48
108
28.66
168
7.64
147
19.96
153
14.97
120
27.53
164
24.02
142
9.47
85
5.27
82
16.31
147
8.61
103
0.63
146
0.19
119
0.00
1
0.00
1
0.08
30
0.90
109
FBW_ROBtwo views9.17
140
2.08
104
10.31
128
2.51
90
1.66
81
13.22
138
6.04
130
23.09
169
20.11
143
19.55
146
15.80
114
18.97
144
11.96
134
19.34
157
11.38
132
0.66
148
0.30
132
2.15
175
0.90
158
1.15
149
2.16
153
XQCtwo views9.46
141
4.84
160
18.88
154
4.07
145
3.09
122
15.97
146
6.65
138
16.26
134
26.63
158
12.63
120
12.35
89
15.75
124
10.62
125
16.48
148
21.19
156
0.44
136
0.96
151
0.05
118
0.32
137
1.10
146
0.99
114
MSMD_ROBtwo views9.73
142
1.43
69
5.42
53
1.81
53
0.42
31
16.83
149
4.30
114
14.29
118
15.34
121
23.06
153
40.17
173
25.63
160
22.62
163
13.37
128
8.85
107
0.48
144
0.03
74
0.00
1
0.00
1
0.13
45
0.47
64
RTSCtwo views9.87
143
4.34
154
16.44
150
4.60
159
2.38
102
12.54
136
0.75
61
18.19
145
37.31
173
16.16
140
15.86
115
13.32
113
7.56
111
20.93
162
23.50
163
0.34
130
1.07
153
0.02
101
0.35
141
0.71
137
0.99
114
ADCPNettwo views10.18
144
3.40
142
33.24
171
2.38
78
1.74
84
17.75
154
7.65
148
13.20
104
13.03
113
14.09
134
21.72
136
19.36
145
11.40
130
15.43
144
23.00
161
0.28
127
1.15
156
0.39
159
1.48
169
0.68
133
2.21
155
WCMA_ROBtwo views10.20
145
1.99
103
9.33
118
3.01
117
2.46
106
15.78
145
7.67
149
12.75
97
15.41
122
25.08
162
33.42
165
27.06
162
19.86
155
13.87
133
12.47
138
1.28
159
0.38
134
0.18
145
0.20
121
0.28
92
1.54
137
PDISCO_ROBtwo views10.27
146
2.79
134
13.30
135
10.58
176
9.96
177
22.78
161
5.99
127
20.01
154
28.57
161
11.88
118
14.49
106
20.54
150
4.92
76
17.01
152
10.29
123
5.79
175
0.42
136
0.14
137
0.13
114
2.53
166
3.25
165
MFN_U_SF_DS_RVCtwo views10.36
147
5.37
161
16.61
152
2.95
114
3.02
120
24.75
164
15.44
169
14.68
121
19.73
140
19.77
149
23.97
141
16.98
130
8.40
117
17.58
154
9.23
112
1.59
163
1.75
165
0.16
141
2.63
174
0.75
138
1.85
148
SHDtwo views10.45
148
3.67
148
14.03
141
4.81
160
4.52
153
11.93
131
1.66
78
21.36
166
38.40
175
18.84
145
21.72
136
17.63
133
13.07
137
15.65
146
17.71
153
0.36
132
1.09
154
0.01
88
0.32
137
0.50
125
1.67
141
FCDSN-DCtwo views10.72
149
1.16
48
5.01
44
2.16
63
1.31
71
17.46
153
4.80
119
17.79
144
19.87
141
25.02
161
32.85
161
33.57
169
23.52
167
16.80
150
12.13
134
0.11
101
0.00
1
0.00
1
0.00
1
0.09
36
0.76
103
Dominik Hirner, Friedrich Fraundorfer: FCDSN-DC: An accurate but lightweight end-to-end trainable neural network for stereo estimation with depth completion.
SGM_RVCbinarytwo views10.77
150
1.11
45
4.66
40
2.66
99
0.41
30
21.03
159
6.87
141
20.02
155
15.54
124
25.86
163
24.97
147
33.59
170
20.51
158
20.89
161
14.78
148
0.31
129
0.26
129
0.22
150
0.26
132
0.43
114
1.01
117
Heiko Hirschmueller: Stereo processing by semiglobal matching and mutual information. TPAMI 2008, Volume 30(2), pp. 328-341
ADCLtwo views10.77
150
3.23
138
21.50
159
2.26
70
1.96
94
23.65
162
9.07
153
14.69
122
24.88
156
15.21
136
27.20
150
12.68
112
14.02
144
16.94
151
24.43
167
0.26
124
0.18
117
0.74
162
0.67
157
0.80
141
1.02
118
DPSNettwo views10.88
152
2.57
127
20.40
158
2.24
67
1.90
90
24.92
165
17.00
175
21.35
165
28.61
162
12.87
123
13.74
101
17.01
131
20.14
157
15.36
142
14.01
144
0.81
151
0.98
152
0.20
148
0.65
156
1.30
153
1.54
137
ADCMidtwo views10.97
153
4.49
157
23.66
161
2.53
93
2.52
111
11.46
128
6.31
134
15.31
126
12.85
112
23.57
157
22.99
139
17.92
135
21.04
159
19.57
159
28.05
172
0.47
142
0.43
137
1.10
167
1.32
166
1.73
159
2.10
151
FC-DCNNcopylefttwo views11.29
154
1.01
40
5.50
54
2.25
68
1.65
80
18.23
155
5.36
125
18.99
150
20.65
146
28.95
166
34.83
168
33.83
171
23.18
166
17.04
153
13.37
141
0.17
109
0.01
51
0.02
101
0.02
80
0.10
39
0.70
97
SANettwo views11.48
155
3.34
140
13.93
140
2.40
81
1.01
58
16.49
147
12.16
164
20.05
157
37.28
172
19.62
147
28.04
151
25.61
159
20.10
156
14.43
137
12.39
137
0.10
99
0.06
89
0.03
110
0.06
102
0.76
139
1.73
145
AnyNet_C32two views11.82
156
7.08
166
25.83
166
4.55
158
5.88
167
16.92
150
16.97
174
14.14
115
18.11
137
16.87
142
21.66
135
14.78
121
13.87
143
23.54
167
31.83
177
0.24
119
0.28
130
0.35
156
0.53
153
1.35
154
1.71
144
MeshStereopermissivetwo views11.94
157
1.88
94
5.24
48
2.11
62
1.44
74
20.09
157
4.95
122
21.74
168
16.75
128
33.10
171
35.07
169
39.28
177
22.72
164
19.45
158
13.59
142
0.19
111
0.06
89
0.02
101
0.00
1
0.46
118
0.65
87
C. Zhang, Z. Li, Y. Cheng, R. Cai, H. Chao, Y. Rui: MeshStereo: A Global Stereo Model with Mesh Alignment Regularization for View Interpolation. ICCV 2015
MFN_U_SF_RVCtwo views13.54
158
4.69
159
28.43
168
4.24
150
2.50
109
24.26
163
5.99
127
28.23
174
20.18
145
24.06
160
29.38
154
30.35
166
16.92
151
20.98
163
21.39
157
1.35
160
1.13
155
1.02
165
1.41
168
1.88
161
2.50
160
ADCStwo views13.76
159
6.46
164
30.74
170
3.62
138
2.93
117
14.28
141
11.22
160
20.02
155
34.42
168
23.20
154
25.42
148
19.51
146
19.62
154
24.54
168
33.88
179
0.27
126
0.23
125
0.37
158
0.19
119
1.47
155
2.84
162
MFMNet_retwo views14.09
160
9.67
172