Low-res two-view results

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):

bad 0.5, 1.0, 2.0, 4.0 [%]: Fraction of pixels with errors larger than the given number of disparities.
Average error [px]: The per-pixel average disparity error.
Root mean square error [px]: The per-pixel root mean square disparity error.
50%, 90%, 95%, 99% error quantile [px]: The highest disparity error within the given percentage of best pixels (for 50%, this is the median error).
Time [s]: The runtime of the method.

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	all	lakes. 1l	lakes. 1s	sand box 1l	sand box 1s	stora. room 1l	stora. room 1s	stora. room 2l	stora. room 2s	stora. room 2 1l	stora. room 2 1s	stora. room 2 2l	stora. room 2 2s	stora. room 3l	stora. room 3s	tunnel 1l	tunnel 1s	tunnel 2l	tunnel 2s	tunnel 3l	tunnel 3s

CREStereo	3.58 1	1.05 5	3.35 2	2.17 2	0.59 8	5.61 15	0.40 5	10.50 6	13.34 17	6.93 5	10.44 5	5.29 10	3.47 7	5.82 5	2.62 6	0.01 8	0.01 4	0.02 29	0.01 26	0.04 4	0.01 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
XX-TBD	4.45 2	10.67 141	3.01 1	10.28 110	0.57 6	5.84 18	0.07 1	7.57 1	12.24 13	11.43 17	14.74 13	4.07 4	1.66 1	4.51 1	2.05 1	0.01 8	0.01 4	0.00 1	0.00 1	0.26 23	0.01 1

s12784h	4.61 3	0.99 2	3.70 4	4.08 11	2.01 27	3.94 4	0.10 2	9.10 3	16.94 26	8.65 7	16.77 21	4.46 5	2.18 2	5.72 3	3.12 15	0.00 1	0.00 1	0.00 1	10.32 168	0.02 1	0.01 1

CREStereo++_RVC	4.61 3	0.99 2	3.70 4	4.08 11	2.01 27	3.94 4	0.10 2	9.10 3	16.94 26	8.65 7	16.77 21	4.46 5	2.18 2	5.72 3	3.12 15	0.00 1	0.00 1	0.00 1	10.32 168	0.02 1	0.01 1

iRaftStereo_RVC	4.68 5	1.51 20	6.89 22	4.91 19	0.72 13	5.26 13	0.46 6	17.60 30	10.47 9	9.53 10	15.13 15	5.75 11	3.24 5	8.93 17	2.22 4	0.00 1	0.09 54	0.00 1	0.00 1	0.46 39	0.41 43

TANstereo	4.70 6	1.08 6	4.01 6	3.53 6	0.47 5	14.26 59	0.15 4	8.67 2	12.54 14	12.50 22	18.43 25	5.20 9	4.54 12	5.68 2	2.07 2	0.00 1	0.13 64	0.00 1	0.01 26	0.02 1	0.63 57

sCroCo_RVC	5.06 7	3.29 78	16.45 100	10.72 113	6.64 105	5.83 17	2.74 28	9.67 5	8.06 6	3.66 1	9.20 4	4.81 7	4.95 13	8.12 11	4.64 25	0.64 92	0.50 106	0.06 42	0.03 42	0.70 62	0.42 44

EAI-Stereo	5.21 8	1.74 37	8.46 45	5.69 30	1.37 21	5.67 16	3.52 29	18.77 31	5.65 2	9.75 11	16.08 16	6.36 13	3.57 8	11.73 34	3.64 18	0.01 8	0.33 87	0.03 31	1.06 119	0.21 19	0.65 63

MSMDNet	5.23 9	1.32 11	8.33 41	4.10 14	1.51 23	6.39 23	1.42 14	15.36 16	8.28 7	12.19 20	14.83 14	12.21 32	5.84 22	8.97 18	3.05 13	0.01 8	0.24 76	0.00 1	0.27 80	0.27 27	0.07 8

PMTNet	5.36 10	0.97 1	3.42 3	1.97 1	1.30 16	8.96 31	0.69 8	12.33 7	12.16 12	8.50 6	11.35 8	6.09 12	4.50 11	9.34 21	2.70 7	22.71 176	0.01 4	0.04 34	0.02 34	0.06 5	0.01 1

RAFT-Stereo + iAFF	5.48 11	1.66 31	7.56 35	3.49 4	0.40 4	2.75 1	0.67 7	16.07 24	14.76 22	12.27 21	18.67 26	11.95 29	8.08 29	7.84 10	2.89 8	0.01 8	0.06 37	0.00 1	0.00 1	0.37 34	0.04 7

Gwc-CoAtRS	5.54 12	1.22 8	8.40 43	4.28 15	1.36 20	6.93 26	1.31 12	15.48 17	15.03 23	12.62 23	13.32 10	12.74 34	5.71 21	8.61 12	2.98 12	0.01 8	0.22 73	0.00 1	0.29 86	0.28 28	0.07 8

CFNet-RSSM	5.55 13	1.08 6	8.60 47	4.30 17	1.21 15	6.42 24	1.25 11	16.12 25	17.84 28	13.12 26	11.14 7	11.84 28	5.52 15	9.14 20	2.35 5	0.01 8	0.20 71	0.00 1	0.36 93	0.34 31	0.07 8

raftrobust	5.56 14	1.42 15	6.00 13	7.04 59	5.11 78	4.90 12	1.88 22	17.13 28	13.72 18	11.87 18	11.58 9	7.82 18	9.21 33	10.28 22	2.92 11	0.02 20	0.05 29	0.00 1	0.00 1	0.22 20	0.11 14

GMStereo	5.94 15	3.84 90	10.86 65	5.13 20	0.71 12	7.27 27	2.66 27	12.38 8	25.14 55	12.01 19	10.66 6	5.12 8	3.46 6	11.39 30	6.16 37	0.45 81	0.02 14	0.05 36	0.31 88	0.86 71	0.25 29

RALCasStereoNet	5.95 16	2.02 50	7.09 26	7.06 60	5.10 77	6.36 22	5.44 50	13.97 11	13.25 16	9.93 12	17.45 23	10.37 25	3.89 9	11.64 33	4.24 23	0.02 20	0.46 103	0.10 53	0.00 1	0.20 18	0.31 34

TestStereo	6.02 17	2.55 63	7.33 31	5.57 27	1.68 25	6.05 19	1.78 21	13.73 9	16.40 24	10.87 16	23.79 37	9.00 22	5.63 18	11.26 29	3.96 22	0.09 38	0.01 4	0.00 1	0.00 1	0.69 60	0.01 1

rafts_anoy	6.02 17	1.67 32	6.27 16	6.53 48	4.98 73	5.30 14	2.53 25	13.77 10	21.01 41	12.70 24	16.23 18	8.56 20	5.65 19	9.09 19	5.22 28	0.04 27	0.31 84	0.09 52	0.03 42	0.25 22	0.17 25

XX-Stereo	6.14 19	1.47 18	9.11 50	11.87 125	6.76 109	3.27 2	0.93 10	15.13 13	5.43 1	8.79 9	28.46 58	13.71 37	2.22 4	11.51 31	2.07 2	0.20 54	0.24 76	0.08 48	0.29 86	1.04 76	0.13 18

AFF-stereo	6.17 20	1.61 28	7.02 24	4.08 11	0.64 9	3.53 3	0.76 9	16.00 23	13.14 15	12.86 25	25.42 41	15.67 42	7.86 25	11.12 28	3.21 17	0.02 20	0.06 37	0.00 1	0.00 1	0.40 36	0.08 11

GwcNet-DCA	6.46 21	1.59 25	7.28 27	3.51 5	0.68 10	4.77 11	1.75 20	15.33 15	23.62 50	15.55 41	20.73 32	7.31 17	11.56 46	8.83 16	5.26 29	0.00 1	0.05 29	0.00 1	0.00 1	0.26 23	1.11 83

RALAANet	6.49 22	2.49 61	9.69 55	7.09 61	4.98 73	6.23 20	1.68 16	19.90 32	14.26 19	15.04 34	14.56 11	12.69 33	4.38 10	12.91 41	3.07 14	0.02 20	0.07 43	0.00 1	0.10 61	0.51 46	0.15 21

222	6.59 23	1.55 22	9.64 54	3.42 3	0.68 10	4.73 8	1.70 17	15.75 20	24.33 54	15.11 35	20.25 28	6.90 14	12.80 51	8.63 13	5.33 30	0.01 8	0.04 21	0.00 1	0.00 1	0.26 23	0.63 57

sAnonymous2	6.59 23	6.21 115	19.05 114	8.91 85	5.27 83	4.58 6	6.69 69	15.56 18	11.12 10	4.86 3	8.71 2	3.81 2	7.93 26	7.38 6	2.89 8	0.09 38	0.07 43	0.05 36	0.28 83	11.25 164	7.10 154

CroCo_RVC	6.59 23	6.21 115	19.05 114	8.91 85	5.27 83	4.58 6	6.69 69	15.56 18	11.12 10	4.86 3	8.71 2	3.81 2	7.93 26	7.38 6	2.89 8	0.09 38	0.07 43	0.05 36	0.28 83	11.25 164	7.10 154

raft+_RVC	6.69 26	1.61 28	5.57 12	6.37 43	4.62 64	6.51 25	6.18 63	20.39 33	23.38 49	10.62 15	17.82 24	9.89 24	5.65 19	10.70 24	3.67 19	0.00 1	0.14 65	0.00 1	0.00 1	0.34 31	0.27 32

Anonymous	6.73 27	2.83 67	18.61 111	11.88 126	7.83 126	14.52 61	9.37 83	14.49 12	10.15 8	4.28 2	6.89 1	3.13 1	7.67 24	11.82 35	5.66 33	0.83 98	0.35 91	0.08 48	0.20 71	2.86 115	1.09 82

DIP-Stereo	6.74 28	1.93 45	10.02 59	4.28 15	1.64 24	13.54 52	1.52 15	15.16 14	6.63 3	14.30 27	26.81 50	10.90 26	5.19 14	14.78 52	7.56 40	0.00 1	0.06 37	0.05 36	0.00 1	0.17 15	0.17 25
Zihua Zheng, Ni Nie, Zhi Ling, Pengfei Xiong, Jiangyu Liu, Hao Wang, Jiankun Li: DIP: Deep Inverse Patchmatch for High-Resolution Optical Flow. cvpr2022
csc	6.83 29	1.55 22	7.28 27	3.56 7	0.32 1	4.73 8	1.70 17	15.75 20	31.87 84	15.11 35	20.25 28	6.90 14	12.80 51	8.63 13	5.33 30	0.01 8	0.04 21	0.00 1	0.00 1	0.09 6	0.63 57

cscssc	6.83 29	1.55 22	7.28 27	3.56 7	0.32 1	4.73 8	1.70 17	15.75 20	31.87 84	15.11 35	20.25 28	6.90 14	12.80 51	8.63 13	5.33 30	0.01 8	0.04 21	0.00 1	0.00 1	0.09 6	0.63 57

RAFT-Stereo	7.04 31	1.34 13	7.89 37	6.23 40	2.30 33	7.51 29	5.54 52	23.96 47	6.94 4	16.26 46	33.57 93	12.04 30	5.61 16	7.67 8	3.74 20	0.01 8	0.05 29	0.01 25	0.00 1	0.10 9	0.12 15
Lahav Lipson, Zachary Teed, and Jia Deng: RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matching. 3DV
R-Stereo Train	7.04 31	1.34 13	7.89 37	6.23 40	2.30 33	7.51 29	5.54 52	23.96 47	6.94 4	16.26 46	33.57 93	12.04 30	5.61 16	7.67 8	3.74 20	0.01 8	0.05 29	0.01 25	0.00 1	0.10 9	0.12 15

111	7.26 33	1.65 30	7.28 27	3.56 7	0.32 1	10.76 35	3.53 30	20.55 34	27.51 67	10.07 13	16.62 20	14.30 38	9.53 36	12.02 36	5.87 35	0.14 47	0.01 4	0.00 1	0.00 1	0.31 29	1.14 87

HITNet	7.83 34	2.93 70	8.39 42	4.76 18	1.48 22	12.75 44	4.64 41	20.97 35	14.37 21	15.14 39	22.15 33	14.57 40	10.62 38	14.38 50	8.14 45	0.04 27	0.01 4	0.48 85	0.02 34	0.57 53	0.10 13
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
FENet	7.94 35	1.01 4	6.72 19	5.87 33	2.07 30	11.11 37	2.03 23	21.39 37	20.81 39	17.14 54	22.65 34	14.43 39	11.54 45	12.97 43	7.89 42	0.15 49	0.24 76	0.03 31	0.02 34	0.14 12	0.62 56

DN-CSS_ROB	8.17 36	2.59 64	14.14 83	6.11 38	3.22 43	6.31 21	3.53 30	22.92 45	20.59 38	16.66 51	30.05 67	9.31 23	7.95 28	13.49 45	4.78 26	0.72 96	0.00 1	0.45 83	0.00 1	0.54 49	0.09 12

test_xeample3	8.30 37	1.88 41	7.48 33	3.74 10	0.57 6	7.49 28	9.01 80	16.85 26	31.87 84	14.47 28	28.31 57	12.88 35	11.56 46	10.87 27	4.39 24	0.04 27	0.10 57	0.00 1	0.02 34	0.09 6	4.31 135

ARAFT	8.50 38	3.19 77	16.58 101	5.52 26	0.91 14	12.13 41	11.20 101	17.19 29	18.28 30	18.40 59	16.33 19	17.10 44	8.58 31	18.56 73	5.00 27	0.11 43	0.04 21	0.00 1	0.00 1	0.55 51	0.25 29

DMCA	8.51 39	1.90 43	9.96 58	5.16 21	4.14 55	10.29 34	4.46 38	25.52 55	19.55 35	16.33 48	20.69 31	17.59 47	11.26 41	10.42 23	11.36 66	0.50 84	0.08 50	0.35 75	0.05 50	0.36 33	0.31 34

GES	8.51 39	1.98 48	11.82 69	6.39 45	1.77 26	12.36 43	6.22 64	21.55 39	21.34 42	19.02 62	26.02 44	8.79 21	9.34 34	13.94 48	7.24 39	0.38 71	0.10 57	0.30 73	0.14 66	0.69 60	0.73 68

HCRNet	8.80 41	1.43 16	7.74 36	14.41 140	5.25 79	10.85 36	2.56 26	26.07 61	18.04 29	15.48 40	25.03 40	18.63 50	9.93 37	13.01 44	6.76 38	0.04 27	0.04 21	0.10 53	0.01 26	0.18 17	0.38 38

CFNet-pseudo	9.29 42	1.31 10	7.48 33	6.39 45	3.68 47	13.69 53	4.61 39	32.46 99	28.01 68	17.36 55	29.86 64	11.43 27	8.43 30	10.79 25	9.29 53	0.04 27	0.09 54	0.08 48	0.04 47	0.17 15	0.60 55

MLCV	9.32 43	2.22 58	12.55 74	5.22 23	1.32 17	13.79 54	1.35 13	21.14 36	23.65 51	21.91 80	31.37 77	17.14 45	7.57 23	16.00 60	10.18 56	0.22 56	0.01 4	0.03 31	0.02 34	0.40 36	0.29 33

BEATNet_4x	9.37 44	4.85 101	12.05 71	5.48 25	1.35 19	13.15 49	6.15 62	25.06 54	16.74 25	16.18 45	24.65 39	17.29 46	12.51 49	16.91 66	10.92 61	0.38 71	0.18 69	0.81 102	0.05 50	1.95 101	0.65 63

pcwnet_v2	9.48 45	1.46 17	7.39 32	5.95 35	3.44 45	15.21 65	4.63 40	28.87 75	18.88 32	16.52 49	38.32 117	15.41 41	11.27 42	10.79 25	8.99 50	0.26 61	0.30 82	0.07 45	0.11 63	0.48 40	1.19 89

UCFNet_RVC	9.48 45	2.12 54	4.89 7	7.03 58	3.78 50	16.05 72	6.37 65	28.25 72	20.00 37	15.93 43	23.07 35	25.21 79	12.79 50	12.57 39	8.01 44	0.07 35	0.04 21	1.17 116	0.23 76	1.31 88	0.69 65

GANet-RSSM	9.48 45	1.32 11	6.45 17	6.72 53	5.51 88	12.78 45	6.03 59	26.55 63	18.96 33	14.92 30	27.95 56	25.41 81	14.69 71	13.52 46	7.60 41	0.10 42	0.10 57	0.16 63	0.01 26	0.51 46	0.39 39

Anonymous3	9.60 48	3.45 82	29.83 137	9.58 95	4.75 66	24.51 113	10.20 91	24.04 49	18.77 31	10.42 14	14.64 12	7.97 19	8.75 32	14.05 49	5.80 34	2.44 137	1.39 134	0.08 48	0.06 55	0.73 66	0.48 49

PSMNet-RSSM	9.71 49	1.73 36	6.09 14	7.47 68	4.27 58	10.06 32	6.00 57	29.14 77	21.74 44	15.76 42	32.88 90	22.63 63	13.18 54	12.94 42	9.12 52	0.09 38	0.02 14	0.07 45	0.10 61	0.70 62	0.24 28

CFNet-ft	9.87 50	2.18 56	5.49 9	7.45 66	5.25 79	15.76 69	5.98 55	21.89 40	21.87 45	14.96 31	30.02 65	25.68 82	13.68 58	12.46 37	10.94 62	0.08 36	0.06 37	1.83 126	0.22 73	1.09 77	0.43 45

CFNet_RVC	9.87 50	2.18 56	5.49 9	7.45 66	5.25 79	15.76 69	5.98 55	21.89 40	21.87 45	14.96 31	30.02 65	25.68 82	13.68 58	12.46 37	10.94 62	0.08 36	0.06 37	1.83 126	0.22 73	1.09 77	0.43 45

cf-r	10.06 52	1.60 26	7.99 39	5.82 31	5.70 94	12.92 47	4.78 44	24.18 51	27.47 66	16.91 53	31.02 74	25.12 78	14.05 63	11.63 32	11.01 64	0.03 25	0.03 18	0.05 36	0.03 42	0.43 38	0.39 39

AdaStereo	10.22 53	3.63 83	9.14 53	9.15 88	3.24 44	15.79 71	5.39 49	31.94 93	25.42 56	18.39 58	26.47 46	19.44 53	9.50 35	16.60 65	8.59 47	0.44 80	0.05 29	0.40 81	0.00 1	0.57 53	0.16 24
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.
iResNet	10.26 54	3.10 76	15.72 93	7.35 63	2.13 31	13.86 55	7.07 75	22.80 44	28.01 68	20.10 71	30.40 70	16.79 43	11.06 40	14.46 51	10.29 59	0.39 75	0.05 29	0.00 1	0.07 59	0.73 66	0.73 68

RAFT + AFF	10.27 55	3.77 86	17.13 103	10.24 108	4.97 71	13.08 48	11.52 106	25.69 56	25.80 58	14.72 29	16.12 17	19.16 52	11.35 44	16.24 63	9.04 51	0.30 65	1.46 136	0.66 95	0.20 71	2.03 105	1.94 107

ACVNet	10.36 56	1.68 33	5.49 9	6.01 36	2.75 38	12.78 45	3.56 32	21.89 40	34.81 99	14.96 31	30.78 72	24.72 76	13.88 61	15.81 56	17.20 89	0.13 45	0.01 4	0.06 42	0.06 55	0.51 46	0.15 21

GwcNet-RSSM	10.65 57	1.94 46	8.57 46	6.89 56	5.26 82	12.26 42	5.21 47	25.80 59	31.50 81	16.67 52	33.52 92	23.70 71	13.88 61	12.84 40	13.67 71	0.05 32	0.10 57	0.14 61	0.02 34	0.65 58	0.39 39

CFNet	10.67 58	2.33 59	8.90 49	6.66 50	4.09 54	16.05 72	4.26 36	30.44 84	31.01 80	18.53 60	26.48 47	23.69 70	13.68 58	15.27 54	10.59 60	0.05 32	0.02 14	0.44 82	0.05 50	0.73 66	0.23 27

DeepPruner_ROB	10.77 59	4.56 96	13.13 77	6.37 43	4.28 59	10.14 33	6.86 71	36.42 124	19.59 36	17.87 56	27.37 54	23.19 66	12.26 48	19.09 80	10.23 58	1.05 112	0.48 105	0.23 67	0.15 68	0.89 73	1.17 88

HSM-Net_RVC	10.88 60	1.23 9	5.46 8	5.57 27	2.63 36	20.12 92	6.06 60	27.67 67	28.35 71	20.63 75	25.51 42	34.01 114	14.38 70	16.18 62	9.29 53	0.11 43	0.07 43	0.01 25	0.00 1	0.13 11	0.14 20
Gengshan Yang, Joshua Manela, Michael Happold, and Deva Ramanan: Hierarchical Deep Stereo Matching on High-resolution Images. CVPR 2019
NOSS_ROB	10.99 61	3.81 88	6.88 21	6.69 51	2.30 33	16.06 74	10.94 98	30.06 82	32.41 90	16.10 44	18.72 27	22.78 64	14.21 67	17.52 69	8.62 48	2.35 132	1.84 140	3.32 147	1.57 130	2.06 106	1.52 98

ac_64	11.10 62	1.49 19	9.88 57	8.24 82	5.45 87	19.12 86	6.46 67	24.10 50	14.34 20	22.80 83	23.87 38	33.28 112	15.54 75	18.87 77	17.26 91	0.28 63	0.12 63	0.02 29	0.02 34	0.26 23	0.63 57

HSM	11.33 63	1.80 38	7.04 25	6.13 39	3.99 53	16.16 77	6.92 73	29.69 80	19.39 34	22.95 84	26.05 45	41.08 142	15.79 78	20.25 92	8.89 49	0.02 20	0.03 18	0.01 25	0.00 1	0.15 13	0.34 37

DMCA-RVC	11.41 64	3.32 79	11.68 68	10.12 106	4.81 67	11.92 40	4.11 34	24.67 52	23.66 52	20.06 70	28.51 59	26.71 86	24.96 119	15.95 59	14.82 75	0.41 76	0.34 90	0.56 90	0.16 69	0.72 65	0.69 65

RASNet	11.59 65	1.71 35	10.36 61	6.30 42	5.66 92	15.48 68	6.12 61	34.50 113	20.86 40	16.61 50	32.51 87	22.04 62	23.00 108	19.44 83	16.15 84	0.71 94	0.03 18	0.05 36	0.01 26	0.15 13	0.13 18

iResNet_ROB	11.71 66	1.96 47	10.77 62	6.01 36	3.68 47	15.13 63	2.34 24	31.72 92	37.58 110	26.16 104	32.77 88	25.88 84	15.34 74	16.35 64	7.94 43	0.19 52	0.01 4	0.00 1	0.00 1	0.31 29	0.12 15

UPFNet	11.72 67	1.52 21	10.33 60	7.38 64	7.14 116	20.05 91	9.62 86	28.90 76	25.94 60	20.02 68	29.73 63	20.91 57	16.34 81	18.61 75	16.17 85	0.26 61	0.17 68	0.25 69	0.03 42	0.55 51	0.56 53

FADNet-RVC-Resample	11.72 67	4.35 94	30.33 140	9.46 93	3.90 51	14.11 56	7.17 76	25.78 57	23.31 48	23.27 87	29.52 62	19.50 55	15.00 73	15.38 55	8.33 46	0.53 85	0.50 106	0.49 86	0.50 102	1.38 90	1.67 102

CBMV_ROB	11.77 69	2.89 68	6.92 23	5.20 22	2.88 40	14.12 57	4.79 45	26.69 64	29.84 73	26.17 105	31.92 81	23.61 69	20.74 104	20.20 91	10.18 56	1.78 125	1.80 139	2.17 132	1.26 126	1.61 92	0.55 51

DSFCA	11.85 70	2.01 49	12.93 76	5.40 24	6.87 111	15.36 66	10.45 95	30.68 86	31.84 83	25.99 102	23.52 36	22.03 61	14.05 63	20.28 94	12.46 67	0.57 87	0.41 95	0.28 72	0.38 96	1.02 74	0.47 48

FADNet_RVC	11.94 71	5.17 105	39.74 153	6.44 47	3.46 46	11.42 38	4.17 35	29.16 78	26.89 63	17.97 57	25.96 43	13.41 36	15.67 76	15.84 58	13.54 70	0.90 104	0.75 123	0.55 89	1.20 124	3.65 128	2.85 119

iResNetv2_ROB	12.00 72	2.91 69	13.46 79	5.82 31	3.73 49	13.15 49	6.45 66	34.15 108	36.02 104	25.25 98	33.81 96	29.79 96	14.24 68	13.61 47	6.10 36	0.21 55	0.02 14	0.12 58	0.00 1	0.82 70	0.25 29

hitnet-ft	12.04 73	2.11 53	6.14 15	7.27 62	3.90 51	25.98 121	3.56 32	22.56 43	23.74 53	20.55 74	33.81 96	31.59 103	18.67 96	19.47 84	17.20 89	0.42 79	1.36 133	0.35 75	0.07 59	1.11 80	0.86 72

NLCA_NET_v2_RVC	12.05 74	3.43 81	16.25 97	7.62 70	5.58 89	14.57 62	6.00 57	32.98 103	27.41 65	21.69 79	31.40 78	25.21 79	13.47 57	14.96 53	15.60 80	0.84 100	0.33 87	0.35 75	0.35 92	1.29 86	1.78 104
Zhibo Rao, Mingyi He, Yuchao Dai, Zhidong Zhu, Bo Li, and Renjie He.: NLCA-Net: A non-local context attention network for stereo matching.
acv_ft	12.16 75	1.68 33	11.97 70	7.70 72	6.20 97	19.61 89	5.52 51	25.78 57	34.81 99	25.62 101	30.78 72	24.72 76	13.20 56	15.81 56	18.80 96	0.13 45	0.01 4	0.06 42	0.06 55	0.63 56	0.15 21

UNet	12.64 76	2.02 50	13.36 78	6.85 55	7.76 125	25.31 119	4.74 43	34.10 107	28.19 70	20.03 69	31.07 75	27.02 89	17.36 88	20.18 90	13.87 72	0.06 34	0.05 29	0.27 70	0.01 26	0.23 21	0.39 39

delet	12.65 77	1.85 39	11.36 67	7.71 74	6.22 99	22.52 104	10.22 93	28.14 71	37.67 111	19.66 63	29.24 61	21.00 58	15.99 80	21.97 100	17.77 94	0.03 25	0.07 43	0.19 65	0.04 47	0.60 55	0.75 70

HGLStereo	12.72 78	2.66 65	12.34 72	9.93 102	7.55 121	16.16 77	7.45 77	31.48 91	27.01 64	20.01 67	35.31 102	26.98 88	18.90 97	17.87 72	18.90 97	0.33 67	0.16 67	0.07 45	0.05 50	0.48 40	0.70 67

GEStereo_RVC	12.81 79	2.49 61	13.64 80	9.68 98	4.97 71	18.16 80	11.30 103	31.26 87	47.47 143	19.84 64	26.97 51	19.03 51	13.19 55	20.14 89	14.86 77	0.83 98	0.51 108	0.10 53	0.13 65	0.54 49	0.99 79

SGM-Forest	12.92 80	1.87 40	6.61 18	5.68 29	2.05 29	23.19 109	11.30 103	34.24 109	30.77 78	26.54 107	31.99 82	29.98 97	15.74 77	21.17 99	13.36 69	1.05 112	0.33 87	0.84 104	0.01 26	0.71 64	0.97 77
Johannes L. Schönberger, Sudipta Sinha, Marc Pollefeys: Learning to Fuse Proposals from Multiple Scanline Optimizations in Semi-Global Matching. ECCV 2018
STTStereo	12.97 81	4.88 102	26.57 132	7.54 69	5.41 86	13.43 51	6.49 68	30.53 85	26.35 62	25.14 97	31.43 79	32.00 104	14.12 65	16.01 61	14.83 76	0.29 64	0.31 84	0.67 96	1.08 120	1.21 83	1.11 83

MMNet	12.97 81	2.08 52	12.74 75	6.98 57	6.27 101	26.39 123	4.73 42	23.10 46	41.79 125	25.48 99	32.29 86	21.09 59	17.79 91	19.52 85	17.90 95	0.19 52	0.07 43	0.10 53	0.00 1	0.37 34	0.59 54

TDLM	12.98 83	3.78 87	9.77 56	10.75 114	4.94 69	16.10 75	14.08 118	35.05 116	31.87 84	22.36 81	26.51 49	26.09 85	14.97 72	24.73 113	13.11 68	1.05 112	0.05 29	1.21 120	0.27 80	1.83 97	1.00 80

AANet_RVC	13.16 84	5.34 107	10.83 64	8.20 81	4.44 60	14.13 58	9.46 84	28.78 74	37.67 111	23.44 89	37.47 110	23.48 67	15.83 79	22.39 102	16.63 87	1.67 123	0.86 127	0.24 68	0.02 34	0.63 56	1.60 100

psm_up	13.18 85	1.91 44	8.81 48	9.59 96	5.84 95	15.17 64	11.37 105	38.74 139	30.81 79	20.66 76	30.32 69	30.00 98	23.12 109	19.19 81	16.02 81	0.58 88	0.10 57	0.17 64	0.01 26	0.65 58	0.55 51

CVANet_RVC	13.24 86	3.39 80	8.43 44	8.42 83	5.04 75	18.52 84	11.72 108	32.03 95	33.85 94	23.17 86	32.77 88	29.78 95	16.51 83	23.20 104	11.28 65	0.94 105	0.04 21	1.19 119	0.47 100	3.13 121	1.01 81

FADNet-RVC	13.27 87	11.57 147	39.71 152	7.94 78	4.50 61	15.41 67	6.95 74	27.96 68	22.98 47	20.37 73	30.73 71	26.86 87	11.31 43	20.50 97	14.28 74	0.14 47	0.14 65	0.13 60	0.18 70	2.83 114	0.86 72

DLCB_ROB	13.35 88	3.00 73	9.12 51	9.43 91	5.68 93	21.80 100	10.12 89	29.19 79	29.92 74	27.71 111	31.45 80	32.37 105	19.50 98	19.02 78	16.73 88	0.22 56	0.04 21	0.38 80	0.06 55	0.49 44	0.85 71

StereoDRNet-Refined	13.36 89	2.80 66	10.82 63	7.94 78	3.10 41	18.95 85	4.45 37	28.47 73	29.65 72	29.11 116	41.47 132	24.44 74	17.64 90	23.34 105	21.66 103	0.16 50	0.07 43	0.53 88	0.32 90	0.77 69	1.64 101
Rohan Chabra, Julian Straub, Chris Sweeney, Richard Newcombe, Henry Fuchs: StereoDRNet. CVPR
CBMV	13.69 90	3.63 83	8.02 40	5.93 34	2.69 37	22.57 105	12.44 110	29.81 81	31.57 82	31.20 126	33.79 95	31.04 99	17.41 89	25.28 116	14.11 73	0.71 94	0.60 115	0.60 92	0.11 63	1.11 80	1.13 86
Konstantinos Batsos, Changjiang Cai, Philippos Mordohai: CBMV: A Coalesced Bidirectional Matching Volume for Disparity Estimation. Computer Vision and Pattern Recognition (CVPR) 2018
PS-NSSS	13.70 91	7.41 126	14.68 88	6.70 52	5.63 90	18.43 83	12.59 111	38.89 141	25.56 57	20.27 72	35.71 103	21.11 60	16.35 82	17.16 67	17.56 92	2.28 130	0.53 110	5.37 152	0.36 93	4.28 134	3.09 123

FADNet	14.18 92	10.92 144	37.56 149	7.79 76	6.64 105	17.48 79	6.91 72	34.73 114	34.03 95	18.85 61	27.02 52	28.26 90	14.24 68	17.77 71	10.09 55	0.41 76	0.67 118	0.37 78	0.37 95	8.13 151	1.34 91

pmcnn	14.59 93	3.68 85	19.78 117	6.83 54	4.51 62	21.27 97	14.67 121	27.37 65	30.54 76	27.09 110	40.19 126	38.71 132	18.53 95	17.70 70	19.27 98	0.84 100	0.09 54	0.00 1	0.00 1	0.48 40	0.31 34

NVstereo2D	15.15 94	2.97 71	16.28 98	9.23 89	7.56 122	30.03 134	9.53 85	42.80 155	42.72 128	15.11 35	27.03 53	23.16 65	16.93 86	24.28 110	15.45 79	3.07 142	0.43 100	1.44 122	0.61 106	6.74 144	7.60 158

StereoDRNet	15.30 95	4.47 95	14.35 87	10.71 112	8.69 136	25.02 117	11.03 99	39.66 142	35.62 102	29.17 117	28.81 60	34.05 115	17.80 92	18.80 76	23.77 113	0.41 76	0.30 82	0.15 62	0.14 66	1.64 93	1.42 94

FINET	15.49 96	6.08 111	24.99 128	8.12 80	7.69 123	21.80 100	22.35 152	33.76 106	48.27 145	25.99 102	32.27 85	19.48 54	16.83 85	19.07 79	16.38 86	1.37 119	0.70 119	0.95 109	0.27 80	2.61 112	0.90 74

DRN-Test	15.88 97	3.83 89	14.70 89	9.96 104	8.02 131	28.93 132	14.65 120	42.34 151	38.25 115	28.18 114	34.31 101	29.22 93	18.35 94	20.01 87	22.68 108	0.37 70	0.41 95	0.46 84	0.28 83	1.28 85	1.34 91

PA-Net	16.29 98	6.88 123	27.34 133	9.88 101	11.12 149	21.77 99	28.15 166	32.06 96	39.49 116	19.99 66	27.91 55	23.50 68	20.59 103	20.25 92	24.04 116	0.24 58	2.26 148	0.22 66	4.59 155	1.13 82	4.48 139
Zhibo Rao, Mingyi He, Yuchao Dai, Zhelun Shen: Patch Attention Network with Generative Adversarial Model for Semi-Supervised Binocular Disparity Prediction.
DISCO	16.36 99	1.88 41	16.15 95	7.78 75	4.19 56	26.92 125	10.21 92	30.31 83	44.26 135	21.35 78	33.91 98	35.80 125	22.40 107	35.65 151	33.87 149	0.18 51	0.08 50	0.04 34	0.04 47	1.71 95	0.54 50

GwcNet	16.48 100	5.23 106	20.03 118	10.24 108	8.48 134	28.79 130	10.09 87	38.60 138	44.51 136	23.29 88	36.95 108	31.29 101	19.79 101	20.42 96	25.70 129	0.54 86	0.72 120	0.57 91	0.39 97	1.37 89	2.64 117

NaN_ROB	16.51 101	6.51 119	16.22 96	10.53 111	4.64 65	31.76 141	17.78 139	37.00 128	43.37 131	29.68 121	31.29 76	32.67 110	20.32 102	28.00 126	16.06 82	0.38 71	0.41 95	0.32 74	0.41 98	0.87 72	1.94 107

FAT-Stereo	16.81 102	2.97 71	17.56 104	9.80 100	6.21 98	16.12 76	13.23 115	38.18 137	39.67 118	27.00 109	39.73 124	37.55 129	31.30 143	20.39 95	21.52 102	3.44 144	2.05 142	0.81 102	1.10 121	1.89 100	5.72 144

DANet	16.85 103	8.00 127	26.04 131	14.56 143	7.25 118	20.59 93	5.37 48	26.22 62	26.02 61	32.17 131	36.00 104	37.20 128	29.77 137	30.75 131	24.62 121	1.14 116	1.34 131	2.21 133	0.51 103	3.18 122	4.07 133
Z Ling, K Yang, J Li, Y Zhang, X Gao, L Luo, L Xie: Domain-adaptive modules for stereo matching network. Neurocomputing 2021
S-Stereo	16.85 103	3.95 91	20.67 121	12.08 127	8.91 139	21.68 98	12.61 112	32.20 98	44.61 137	22.37 82	38.42 118	29.15 92	28.02 131	18.58 74	22.36 107	2.88 141	3.11 158	0.87 107	1.82 134	3.58 127	9.08 164

MaskLacGwcNet_RVC	16.87 105	16.27 158	17.93 109	11.24 122	6.69 108	28.95 133	10.44 94	41.23 148	21.49 43	29.68 121	34.27 100	32.48 106	11.00 39	20.05 88	15.09 78	8.22 164	2.59 151	12.23 170	0.75 112	10.40 159	6.34 147

PSMNet_ROB	16.90 106	5.45 108	14.16 84	13.77 134	7.25 118	27.65 127	32.74 170	43.21 158	37.88 113	24.00 90	32.97 91	34.55 119	16.69 84	17.41 68	23.96 115	0.38 71	0.22 73	0.93 108	2.05 138	1.83 97	0.96 76

GANet	16.92 107	4.32 93	12.46 73	10.84 116	4.24 57	23.04 108	15.36 128	39.91 143	34.47 98	32.42 132	45.10 146	40.04 140	27.54 127	23.61 108	20.20 99	1.02 111	0.08 50	0.67 96	0.22 73	1.87 99	0.97 77

Syn2CoEx	17.03 108	10.40 139	25.65 130	20.35 157	8.02 131	26.28 122	11.27 102	37.16 129	32.69 92	28.89 115	40.17 125	29.27 94	26.63 122	19.19 81	17.63 93	0.98 107	0.54 111	1.23 121	0.23 76	1.82 96	2.24 114

NCCL2	17.23 109	6.37 118	14.30 86	23.45 166	7.18 117	24.16 112	17.05 137	34.39 111	25.88 59	31.55 127	38.01 113	42.09 147	26.81 123	20.94 98	22.69 109	0.45 81	0.23 75	2.63 138	1.80 133	2.02 104	2.57 115

ADCRef	17.30 110	6.73 121	42.01 156	10.11 105	8.78 138	26.61 124	10.45 95	31.29 89	32.75 93	31.19 125	41.62 134	18.47 49	19.78 100	24.29 111	34.06 151	0.76 97	0.42 99	2.16 131	1.79 132	1.23 84	1.56 99

XPNet_ROB	17.33 111	4.84 100	15.44 92	11.15 119	6.64 105	20.88 94	16.10 131	36.26 123	39.54 117	31.80 130	41.32 131	41.30 144	23.91 111	24.67 112	27.40 135	1.07 115	0.74 121	0.79 100	0.25 78	1.10 79	1.31 90

RP	17.58 112	4.92 103	15.27 91	15.44 147	11.95 155	21.05 95	11.75 109	27.64 66	45.09 140	22.99 85	42.05 136	39.39 136	27.74 129	25.54 117	22.33 106	5.45 157	0.59 114	4.33 149	1.62 131	3.53 126	2.86 120

SuperB	17.60 113	6.22 117	56.40 169	7.91 77	5.07 76	19.70 90	9.03 81	24.87 53	50.15 155	26.25 106	39.13 119	17.61 48	20.84 105	22.28 101	25.31 125	0.70 93	0.31 84	1.10 114	0.71 109	16.58 173	1.86 105

ETE_ROB	17.67 114	9.48 133	17.66 105	13.60 132	4.57 63	23.03 107	21.09 149	32.49 100	35.24 101	30.49 123	38.06 114	46.18 158	24.49 114	26.40 120	24.21 118	0.46 83	0.24 76	1.14 115	0.84 114	1.52 91	2.13 111

APVNet	17.75 115	5.46 109	20.35 120	12.31 128	8.40 133	30.16 135	24.10 160	41.14 147	30.12 75	25.08 96	41.09 129	31.24 100	23.12 109	27.07 122	27.86 140	0.34 68	0.56 113	3.02 143	0.52 104	1.68 94	1.45 95

PWCDC_ROB	17.80 116	9.62 134	18.38 110	17.95 153	7.74 124	24.61 115	5.56 54	34.41 112	52.79 159	29.66 120	50.20 156	20.63 56	19.69 99	29.85 130	20.87 100	6.18 160	0.26 81	0.10 53	0.05 50	5.36 142	2.00 109

MDST_ROB	17.92 117	1.60 26	13.68 81	13.88 136	6.44 103	43.05 164	14.86 123	42.74 154	41.66 123	43.25 155	42.85 139	28.43 91	17.28 87	29.35 128	16.06 82	0.88 102	0.11 62	0.63 94	0.47 100	0.50 45	0.64 62

Nwc_Net	17.98 118	4.67 98	17.79 108	14.26 138	11.65 154	25.14 118	14.88 124	40.85 146	39.83 119	21.07 77	43.24 141	34.50 118	27.64 128	26.44 121	25.03 122	3.22 143	0.18 69	2.52 137	3.18 145	1.98 102	1.51 97

ADCP+	18.16 119	4.61 97	32.94 143	9.31 90	10.23 145	28.87 131	11.69 107	33.43 104	36.49 106	29.28 119	40.49 127	24.18 72	24.61 116	33.89 141	35.33 152	0.25 59	0.40 94	2.00 128	1.17 123	2.16 107	1.87 106

PWC_ROB	18.37 120	10.55 140	25.19 129	10.21 107	6.29 102	23.75 110	4.79 45	35.93 122	48.75 148	32.56 133	44.46 145	34.68 120	24.78 117	30.99 132	25.72 130	1.38 120	0.08 50	1.57 123	0.26 79	2.18 108	3.17 125

AF-Net	18.38 121	5.60 110	14.21 85	15.95 148	10.78 147	22.26 102	10.45 95	35.39 118	50.22 156	24.71 95	37.77 111	41.55 145	29.24 135	29.11 127	26.67 131	4.35 153	0.06 37	4.56 150	0.90 116	2.42 109	1.38 93

Anonymous Stereo	18.60 122	11.75 148	49.81 162	14.43 141	12.02 157	14.33 60	23.23 159	32.16 97	43.08 130	24.32 93	34.16 99	24.24 73	14.12 65	31.67 135	30.84 146	0.89 103	0.91 128	1.74 125	1.92 137	3.23 123	3.23 126

stereogan	18.62 123	3.07 74	16.31 99	13.04 131	9.99 143	35.74 152	9.09 82	38.17 136	45.27 141	24.39 94	41.24 130	39.93 138	25.46 120	31.29 134	24.33 120	1.80 126	0.91 128	2.71 140	0.75 112	5.15 141	3.70 127

edge stereo	18.75 124	5.12 104	17.66 105	10.77 115	7.84 128	22.41 103	11.03 99	35.23 117	42.07 127	34.62 138	42.74 138	41.28 143	35.50 150	25.14 115	24.17 117	2.42 136	2.42 150	6.35 156	1.45 128	2.89 118	3.80 128

RYNet	18.92 125	4.69 99	16.88 102	10.99 117	15.84 164	46.72 166	16.57 134	37.84 133	48.03 144	26.92 108	26.48 47	37.84 131	24.92 118	19.80 86	31.04 147	0.25 59	0.25 80	0.62 93	0.03 42	6.34 143	6.39 148

LALA_ROB	19.19 126	7.15 125	15.93 94	12.96 130	5.30 85	28.30 129	23.21 158	42.51 152	36.69 107	33.45 136	40.81 128	50.96 164	24.53 115	27.92 124	25.66 128	0.60 89	0.44 101	2.04 129	1.22 125	2.54 111	1.47 96

aanetorigin	19.21 127	12.59 150	53.50 165	9.67 97	7.83 126	11.85 39	13.87 116	17.05 27	34.15 97	49.18 164	48.89 154	34.05 115	30.22 139	22.86 103	28.00 144	1.01 110	0.75 123	0.67 96	0.89 115	3.07 119	4.14 134

SGM_RVC	19.52 128	2.12 54	6.79 20	6.59 49	1.33 18	38.20 158	15.74 129	38.13 135	34.07 96	45.71 160	41.91 135	51.41 165	38.10 155	38.77 161	27.80 139	0.61 90	0.36 92	0.51 87	0.33 91	1.03 75	0.90 74
Heiko Hirschmueller: Stereo processing by semiglobal matching and mutual information. TPAMI 2008, Volume 30(2), pp. 328-341
RTSC	19.64 129	9.36 130	31.06 141	10.99 117	6.24 100	28.10 128	10.14 90	37.50 132	58.34 172	31.74 129	36.70 107	32.50 107	21.03 106	36.00 154	37.24 157	1.23 117	0.44 101	0.12 58	0.31 88	1.98 102	1.69 103

NCC-stereo	19.75 130	6.08 111	23.96 125	15.05 144	16.17 165	18.42 81	21.45 150	43.31 159	41.00 121	24.30 91	38.12 115	34.98 121	26.92 124	35.46 148	25.65 126	4.29 151	2.07 143	3.24 145	8.90 166	2.86 115	2.86 120

Abc-Net	19.75 130	6.08 111	23.96 125	15.05 144	16.17 165	18.42 81	21.45 150	43.31 159	41.00 121	24.30 91	38.12 115	34.98 121	26.92 124	35.46 148	25.65 126	4.29 151	2.07 143	3.24 145	8.90 166	2.86 115	2.86 120
Xing Li, Yangyu Fan, Guoyun Lv, and Haoyue Ma: Area-based Correlation and Non-local Attention Network for Stereo Matching. The Visual Computer
DGSMNet	19.79 132	12.06 149	33.43 145	13.79 135	16.23 167	22.97 106	15.88 130	35.41 119	41.96 126	19.90 65	32.16 84	35.60 124	18.14 93	29.48 129	23.10 110	3.85 148	3.89 166	6.16 155	4.75 156	13.56 169	13.44 172

RGC	19.90 133	13.26 151	20.24 119	18.19 154	14.06 162	21.21 96	14.33 119	34.97 115	43.60 133	27.73 112	41.49 133	39.76 137	27.98 130	34.52 144	21.37 101	3.66 147	0.54 111	8.67 164	4.16 153	4.19 132	4.02 131

DeepPrunerF	19.92 134	11.02 146	44.29 157	20.74 159	17.75 169	19.19 87	22.57 153	36.86 127	49.93 151	25.48 99	36.62 106	24.69 75	23.97 112	23.44 106	21.79 104	2.68 140	1.63 138	5.70 153	3.95 152	3.38 125	2.67 118

WCMA_ROB	20.02 135	4.30 92	19.72 116	9.47 94	7.00 114	32.71 146	13.99 117	31.97 94	32.48 91	41.51 151	52.00 159	44.09 151	36.14 152	32.43 137	24.29 119	6.19 161	2.79 155	1.09 113	1.10 121	3.95 129	3.13 124

psmorigin	20.95 136	10.24 138	35.84 146	9.75 99	7.05 115	24.59 114	10.11 88	32.85 102	32.17 89	43.09 154	54.57 164	42.79 148	36.62 153	35.99 153	27.41 136	1.58 122	2.72 154	0.79 100	1.05 118	3.26 124	6.50 149

SANet	20.96 137	6.60 120	29.81 136	8.83 84	3.19 42	31.27 140	20.56 148	41.86 150	56.09 168	39.30 147	43.62 142	44.95 153	31.93 145	27.83 123	23.67 112	0.94 105	0.52 109	0.99 110	0.42 99	4.72 135	2.03 110

FBW_ROB	21.00 138	10.23 137	22.72 123	13.71 133	6.77 110	30.49 139	16.46 133	44.58 161	43.57 132	44.25 158	39.59 122	43.27 150	26.12 121	33.96 142	21.80 105	2.58 139	1.84 140	7.14 163	2.78 144	4.00 130	4.06 132

G-Net	21.25 139	8.24 128	37.97 151	15.14 146	9.26 141	50.87 169	16.80 136	27.96 68	30.54 76	38.92 146	42.93 140	33.27 111	32.17 146	27.99 125	25.25 124	4.40 154	3.83 164	3.60 148	0.97 117	7.91 150	6.95 152

SHD	21.32 140	10.03 136	30.16 139	14.30 139	8.68 135	24.15 111	8.93 78	40.82 145	61.17 174	35.79 141	44.15 144	38.73 133	30.20 138	33.88 140	31.09 148	2.00 127	0.81 125	1.17 116	1.55 129	4.08 131	4.78 141

CSAN	21.34 141	9.45 132	23.34 124	20.99 161	4.95 70	32.57 144	34.26 171	38.83 140	49.95 153	36.97 143	39.72 123	44.97 154	31.73 144	26.05 118	23.94 114	1.52 121	0.47 104	0.85 105	1.43 127	2.73 113	2.14 113

ADCL	21.64 142	6.20 114	47.32 160	9.93 102	6.91 113	38.69 160	19.97 146	31.26 87	54.04 160	27.89 113	47.97 152	31.35 102	30.66 141	33.18 139	36.62 154	0.99 108	0.92 130	2.63 138	1.82 134	2.45 110	2.13 111

ADCPNet	21.93 143	9.38 131	57.92 170	11.76 123	6.88 112	36.03 153	18.44 141	32.80 101	35.93 103	32.91 135	43.80 143	39.21 135	26.99 126	31.14 133	36.94 156	2.05 128	2.67 153	2.30 135	3.45 148	4.22 133	3.82 129

CC-Net-ROB	22.04 144	22.91 168	28.86 134	14.05 137	8.73 137	33.79 148	15.26 126	46.84 165	36.06 105	33.72 137	32.10 83	35.47 123	24.02 113	23.57 107	27.88 141	8.95 167	2.95 157	24.94 176	3.68 149	12.41 167	4.62 140

MeshStereo	22.27 145	6.87 122	11.15 66	7.69 71	4.87 68	37.70 156	13.06 113	41.64 149	36.95 108	50.92 166	53.41 161	58.12 169	41.93 163	37.73 158	27.62 138	2.52 138	2.37 149	2.99 142	2.07 139	3.09 120	2.60 116
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
ADCMid	22.76 146	10.75 142	41.73 155	11.23 121	7.97 130	27.26 126	19.18 145	35.44 120	38.02 114	40.23 148	46.38 149	36.56 127	42.70 164	38.20 159	40.75 164	1.77 124	1.43 135	3.05 144	3.81 151	4.96 137	3.82 129

MSMD_ROB	22.90 147	9.64 135	14.76 90	17.46 152	9.88 142	34.94 150	18.83 142	33.67 105	37.47 109	40.97 150	59.80 169	45.93 156	38.34 156	31.74 136	23.24 111	6.22 162	3.58 162	8.78 165	11.03 170	6.76 145	5.06 143

LE_ROB	22.93 148	3.07 74	14.02 82	7.70 72	2.86 39	31.99 143	17.35 138	28.10 70	67.19 181	70.51 180	55.61 167	49.07 161	47.62 168	24.76 114	36.49 153	0.35 69	0.20 71	0.27 70	0.55 105	0.48 40	0.44 47

AnyNet_C32	23.37 149	13.28 152	40.29 154	12.36 129	11.44 151	30.39 136	29.15 167	31.36 90	44.90 139	35.36 139	48.86 153	33.61 113	34.20 148	43.15 165	41.62 168	1.27 118	1.35 132	1.18 118	2.58 142	4.77 136	6.27 146

SGM-ForestM	23.72 150	2.44 60	9.13 52	7.43 65	2.25 32	44.80 165	19.11 144	44.90 162	49.97 154	50.74 165	51.18 157	62.06 176	45.83 167	44.96 166	33.88 150	1.00 109	0.84 126	0.78 99	0.62 107	1.29 86	1.11 83

MFN_U_SF_DS_RVC	23.92 151	13.60 153	30.09 138	25.46 168	10.02 144	38.51 159	20.40 147	35.77 121	43.76 134	42.15 152	47.10 151	39.99 139	28.20 133	35.96 152	27.06 133	3.64 146	3.85 165	6.66 157	16.30 174	4.97 138	4.98 142

ccnet	24.02 152	16.16 157	21.70 122	11.83 124	17.75 169	37.87 157	16.18 132	45.46 164	31.93 88	36.99 144	46.46 150	38.76 134	29.52 136	37.55 157	40.93 167	9.20 168	6.29 171	9.63 167	7.71 163	11.19 163	7.21 156

XQC	24.10 153	16.76 159	50.68 163	21.37 164	11.01 148	35.24 151	18.84 143	37.28 131	55.11 164	31.64 128	30.06 68	37.71 130	30.31 140	37.17 155	39.66 161	4.20 150	0.41 95	2.76 141	1.89 136	11.46 166	8.42 162

FC-DCNN	24.37 154	10.99 145	19.02 113	18.44 155	9.16 140	36.98 154	23.07 156	40.44 144	43.05 129	46.25 162	53.58 163	50.44 162	37.86 154	35.45 147	27.57 137	6.87 163	3.37 159	6.01 154	4.88 157	7.86 149	6.05 145

DPSNet	24.40 155	6.97 124	33.14 144	11.16 120	6.54 104	53.33 171	43.32 177	51.28 171	59.37 173	30.89 124	39.36 120	40.35 141	34.30 149	32.57 138	25.09 123	4.65 156	1.62 137	0.37 78	0.66 108	8.51 153	4.44 138

PDISCO_ROB	24.45 156	9.24 129	29.28 135	28.68 172	19.96 171	37.14 155	15.33 127	45.04 163	54.69 162	29.24 118	42.67 137	46.07 157	28.17 132	35.35 146	30.30 145	9.92 170	2.13 145	6.90 158	3.20 146	8.74 154	6.87 151

EDNetEfficient	25.16 157	22.23 167	77.22 178	9.09 87	7.38 120	19.38 88	16.78 135	21.41 38	55.74 167	61.52 176	53.51 162	36.22 126	40.65 162	24.22 109	38.27 158	0.62 91	0.74 121	0.86 106	2.60 143	7.70 148	7.09 153

GANetREF_RVC	25.41 158	29.28 172	24.63 127	25.18 167	6.04 96	30.44 137	35.39 173	47.84 166	50.60 157	36.88 142	36.04 105	34.11 117	31.23 142	37.46 156	27.96 143	8.58 165	2.66 152	16.29 172	4.95 158	14.37 172	8.32 161
Zhang, Feihu and Prisacariu, Victor and Yang, Ruigang and Torr, Philip HS: GA-Net: Guided Aggregation Net for End- to-end Stereo Matching. CVPR 2019
FCDSN-DC	25.59 159	14.91 156	18.83 112	28.55 171	15.68 163	39.29 161	18.16 140	42.83 157	41.69 124	44.43 159	51.94 158	50.82 163	39.22 159	34.59 145	27.29 134	6.05 159	3.63 163	7.01 161	6.41 159	10.35 158	10.17 168
Dominik Hirner, Friedrich Fraundorfer: FCDSN-DC: An accurate but lightweight end-to-end trainable neural network for stereo estimation with depth completion.
ADCS	26.20 160	13.71 154

Low-res two-view benchmark