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	0.98 1	0.42 10	1.86 1	0.07 1	0.22 7	1.65 1	0.02 1	4.07 2	6.33 19	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

PMTNet	1.41 2	0.37 8	2.06 4	0.26 3	0.15 4	2.94 13	0.34 9	3.39 1	5.33 10	1.00 2	1.27 1	0.28 1	0.93 5	4.04 9	0.74 4	5.00 130	0.00 1	0.00 1	0.00 1	0.03 13	0.01 2

DPM-Stereo	1.97 3	0.64 26	2.95 13	0.17 2	0.10 1	4.83 36	0.13 3	8.60 17	4.06 4	6.42 26	4.92 6	0.44 3	0.72 2	3.57 8	1.80 7	0.00 1	0.01 36	0.00 1	0.00 1	0.05 19	0.04 24

Gwc-CoAtRS	2.02 4	0.33 6	2.61 5	1.04 10	0.30 10	3.56 16	0.40 12	5.89 4	5.94 16	5.08 13	5.01 7	5.55 15	1.72 9	1.60 2	0.91 5	0.01 27	0.09 81	0.00 1	0.25 99	0.01 2	0.02 6

FENet	2.19 5	0.34 7	3.04 14	2.05 49	0.26 9	2.82 10	0.13 3	7.02 8	5.55 12	4.97 12	1.94 3	4.91 11	3.39 22	4.71 12	2.50 13	0.02 34	0.03 53	0.00 1	0.00 1	0.00 1	0.21 61

R-Stereo Train	2.44 6	0.32 1	1.93 2	0.94 7	0.16 5	3.67 18	0.61 23	6.37 6	3.08 1	9.14 49	17.44 83	1.80 4	0.77 3	1.76 3	0.70 2	0.00 1	0.01 36	0.00 1	0.00 1	0.01 2	0.03 15

RAFT-Stereo	2.44 6	0.32 1	1.93 2	0.94 7	0.16 5	3.67 18	0.61 23	6.37 6	3.08 1	9.14 49	17.44 83	1.80 4	0.77 3	1.76 3	0.70 2	0.00 1	0.01 36	0.00 1	0.00 1	0.01 2	0.03 15
Lahav Lipson, Zachary Teed, and Jia Deng: RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matching. 3DV
ACVNet	2.58 8	0.53 17	2.69 7	0.84 5	0.59 20	2.61 6	0.55 18	8.35 12	5.70 13	4.45 7	9.26 29	5.77 17	3.50 24	2.24 5	4.41 38	0.00 1	0.00 1	0.00 1	0.00 1	0.17 51	0.01 2

DN-CSS_ROB	2.69 9	1.40 73	5.34 47	2.31 64	0.75 27	3.14 15	0.06 2	6.11 5	3.87 3	5.34 16	12.18 52	2.34 6	1.22 6	7.84 32	1.48 6	0.03 45	0.00 1	0.00 1	0.00 1	0.35 80	0.03 15

HITNet	2.79 10	0.77 30	4.02 30	2.03 48	0.11 3	5.58 43	0.59 22	9.24 23	5.15 9	6.42 26	7.26 18	3.66 7	2.92 20	4.07 10	3.87 35	0.00 1	0.00 1	0.00 1	0.00 1	0.06 24	0.02 6
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
DMCA	2.82 11	0.64 26	4.45 36	1.61 27	0.89 34	3.86 22	0.57 21	9.18 21	5.10 8	6.24 22	6.49 12	7.17 31	2.13 12	3.31 7	4.73 42	0.00 1	0.01 36	0.00 1	0.00 1	0.04 16	0.02 6

ccs	3.04 12	0.39 9	3.08 16	1.78 34	0.52 19	2.04 2	0.50 16	13.09 71	13.71 76	3.54 5	5.36 10	5.50 14	2.45 14	4.81 13	2.88 16	0.09 66	0.08 77	0.12 96	0.10 86	0.20 56	0.50 90

AdaStereo	3.09 13	0.58 21	3.04 14	2.84 82	0.48 18	4.08 25	1.29 44	12.16 62	7.77 38	6.03 20	9.62 31	5.79 19	1.53 8	4.56 11	1.93 9	0.00 1	0.00 1	0.00 1	0.00 1	0.02 7	0.02 6
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-RVC	3.17 14	1.26 66	6.39 65	2.10 52	1.51 53	3.02 14	0.56 20	7.84 11	5.06 7	9.23 52	5.19 8	8.46 39	2.72 18	6.31 18	3.22 23	0.06 62	0.11 84	0.09 93	0.02 56	0.16 49	0.08 31

BEATNet_4x	3.24 15	1.27 67	5.89 56	1.56 24	0.10 1	5.26 40	1.07 39	10.08 30	5.50 11	6.89 31	7.73 20	4.53 10	4.13 34	5.05 14	5.27 46	0.04 53	0.05 62	0.00 1	0.00 1	0.23 70	0.23 64

GANet-RSSM	3.27 16	0.60 22	3.38 20	2.94 89	2.00 68	2.61 6	0.48 14	10.93 50	6.11 17	5.47 17	10.20 35	6.57 23	5.44 55	6.25 17	2.23 11	0.00 1	0.00 1	0.08 90	0.00 1	0.17 51	0.03 15

NOSS_ROB	3.30 17	0.46 11	2.62 6	2.08 50	1.01 42	5.60 44	0.74 34	10.37 38	11.48 65	5.15 14	8.43 25	5.67 16	1.73 10	7.97 34	2.34 12	0.02 34	0.06 68	0.00 1	0.00 1	0.07 26	0.14 52

CFNet-ft	3.31 18	0.94 44	2.69 7	1.50 21	2.38 75	2.81 8	0.68 28	8.35 12	7.43 32	4.45 7	9.94 32	10.20 49	4.60 40	6.50 21	3.41 27	0.00 1	0.00 1	0.03 74	0.00 1	0.22 65	0.03 15

CFNet_RVC	3.31 18	0.94 44	2.69 7	1.50 21	2.38 75	2.81 8	0.68 28	8.35 12	7.43 32	4.45 7	9.94 32	10.20 49	4.60 40	6.49 20	3.41 27	0.00 1	0.00 1	0.03 74	0.00 1	0.22 65	0.03 15

PSMNet-RSSM	3.33 20	0.77 30	3.22 19	2.12 53	1.90 67	2.33 4	0.71 31	11.17 51	5.83 15	5.90 18	11.43 47	7.04 27	4.83 49	5.96 16	3.17 21	0.00 1	0.00 1	0.02 66	0.00 1	0.08 28	0.02 6

cf-r	3.42 21	0.81 33	3.97 29	2.41 68	2.78 84	2.85 11	0.64 27	8.44 15	6.32 18	6.25 23	11.57 48	8.97 43	3.54 26	5.93 15	3.83 34	0.00 1	0.00 1	0.00 1	0.00 1	0.09 32	0.07 28

MLCV	3.44 22	0.88 37	5.60 51	1.39 17	0.25 8	4.36 29	0.33 8	7.25 9	7.28 29	9.17 51	12.24 54	5.09 12	2.47 15	9.15 56	3.23 24	0.00 1	0.00 1	0.00 1	0.00 1	0.10 33	0.02 6

DeepPruner_ROB	3.52 23	1.14 60	4.06 31	1.12 11	1.65 57	3.65 17	0.83 36	13.96 80	4.47 6	7.80 37	10.84 40	7.05 29	2.16 13	8.14 41	3.08 20	0.07 64	0.03 53	0.00 1	0.01 52	0.32 76	0.06 27

STTStereo	3.60 24	0.93 43	6.34 63	2.71 79	2.23 73	3.68 20	0.63 26	9.42 24	6.73 22	9.87 61	6.97 16	8.84 42	3.65 27	6.85 22	3.04 18	0.00 1	0.02 47	0.01 54	0.00 1	0.02 7	0.02 6

ccs_rob	3.63 25	1.12 59	4.42 34	2.52 71	0.91 37	5.50 42	0.21 6	10.11 33	9.11 48	6.55 29	11.28 44	8.32 38	2.55 16	7.66 29	2.01 10	0.00 1	0.00 1	0.00 1	0.00 1	0.20 56	0.08 31

hitnet-ft	3.66 26	0.65 28	2.79 11	0.89 6	0.72 26	4.25 27	0.55 18	9.11 20	7.52 34	7.14 33	10.19 34	12.49 66	4.70 44	7.44 27	4.41 38	0.03 45	0.00 1	0.06 86	0.00 1	0.20 56	0.04 24

iResNet	3.68 27	0.91 40	7.94 81	2.97 90	0.34 12	4.44 33	0.48 14	7.70 10	9.74 52	7.72 36	12.74 58	4.03 8	2.87 19	8.05 37	3.37 26	0.02 34	0.01 36	0.00 1	0.00 1	0.10 33	0.09 35

acv_ft	3.69 28	0.53 17	4.58 38	2.31 64	4.71 116	7.44 63	0.86 37	8.54 16	5.70 13	9.29 53	9.26 29	5.77 17	4.15 35	2.24 5	8.13 71	0.00 1	0.00 1	0.00 1	0.00 1	0.29 72	0.01 2

ac_64	3.70 29	0.47 12	3.50 23	2.87 83	3.50 96	5.91 47	0.72 32	10.17 35	4.34 5	8.16 41	5.80 11	10.82 54	5.64 57	7.08 24	4.73 42	0.03 45	0.00 1	0.00 1	0.00 1	0.08 28	0.14 52

CFNet	3.72 30	1.10 56	5.03 42	2.49 70	1.59 54	4.90 38	0.22 7	11.38 52	9.88 54	4.80 10	11.25 43	6.44 22	3.68 28	8.33 43	3.00 17	0.00 1	0.00 1	0.00 1	0.00 1	0.22 65	0.07 28

GwcNet-RSSM	3.77 31	1.05 50	5.06 43	2.21 58	1.87 65	2.36 5	1.52 47	8.98 19	7.96 39	5.91 19	14.59 68	7.94 36	3.50 24	6.43 19	5.70 49	0.00 1	0.00 1	0.04 81	0.00 1	0.20 56	0.08 31

FADNet-RVC-Resample	3.79 32	1.62 85	12.06 96	1.43 19	0.66 22	5.94 48	2.41 55	10.18 36	8.58 46	6.28 24	4.22 5	5.33 13	4.80 48	7.71 30	3.19 22	0.17 80	0.21 101	0.17 103	0.12 88	0.41 88	0.29 77

CC-Net-ROB	3.84 33	1.07 52	5.23 45	2.65 76	2.96 87	4.22 26	0.69 30	10.43 40	7.72 36	8.78 46	8.29 24	9.61 46	4.02 33	7.16 26	3.65 32	0.13 73	0.03 53	0.02 66	0.03 65	0.05 19	0.03 15

NLCA_NET_v2_RVC	3.84 33	1.06 51	5.23 45	2.72 80	3.27 89	4.36 29	0.61 23	10.71 45	7.56 35	8.75 45	7.89 21	9.86 48	3.90 31	7.15 25	3.44 29	0.14 74	0.02 47	0.02 66	0.03 65	0.04 16	0.03 15
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	3.91 35	1.67 88	12.95 103	0.96 9	0.75 27	5.71 46	0.54 17	10.83 48	6.60 21	3.46 3	8.09 22	4.10 9	3.40 23	9.43 60	6.33 52	0.36 97	0.44 115	0.17 103	0.46 116	0.91 105	0.95 109

FADNet-RVC	3.98 36	1.84 94	12.48 99	1.69 32	0.44 16	4.33 28	1.31 45	11.84 56	7.15 27	3.53 4	3.50 4	10.63 53	4.43 39	9.12 55	6.25 51	0.03 45	0.10 82	0.00 1	0.03 65	0.60 95	0.25 70

HSM	4.00 37	0.79 32	3.16 18	1.59 26	2.17 71	6.77 57	1.11 40	12.28 63	6.35 20	6.75 30	8.11 23	13.90 72	5.37 54	8.85 51	2.71 15	0.00 1	0.00 1	0.00 1	0.00 1	0.02 7	0.02 6

TDLM	4.11 38	1.11 58	3.54 24	1.62 28	1.04 43	3.91 23	7.41 105	10.60 43	10.67 59	6.38 25	12.59 57	5.95 20	4.77 46	8.79 50	3.04 18	0.58 109	0.00 1	0.01 54	0.00 1	0.19 55	0.12 47

CBMV_ROB	4.14 39	0.52 14	3.14 17	1.30 14	0.77 30	6.92 58	1.97 53	10.11 33	9.58 50	8.92 48	14.20 66	7.12 30	5.90 60	8.65 47	3.50 31	0.01 27	0.05 62	0.00 1	0.00 1	0.04 16	0.09 35

CVANet_RVC	4.16 40	1.16 61	3.60 25	1.94 44	1.46 51	3.92 24	4.68 84	10.89 49	8.34 44	7.58 35	10.84 40	10.27 51	6.62 64	8.56 46	2.69 14	0.39 99	0.00 1	0.00 1	0.01 52	0.21 63	0.09 35

HSM-Net_RVC	4.20 41	0.32 1	2.76 10	0.63 4	0.69 24	6.95 59	1.69 50	11.96 57	8.36 45	8.83 47	12.17 51	15.18 80	4.21 37	6.91 23	3.30 25	0.02 34	0.02 47	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
DSFCA	4.21 42	0.50 13	5.45 49	1.34 15	1.68 59	7.04 60	4.51 81	10.73 46	7.00 24	10.78 69	6.80 13	8.48 40	4.63 42	9.91 64	5.12 45	0.02 34	0.06 68	0.02 66	0.03 65	0.08 28	0.03 15

FADNet	4.23 43	1.65 87	11.75 95	1.64 30	0.80 32	4.80 35	0.77 35	13.76 79	11.65 67	3.97 6	5.24 9	9.62 47	5.14 51	8.40 44	3.78 33	0.21 84	0.04 58	0.07 89	0.05 76	1.14 110	0.10 42

iResNet_ROB	4.23 43	1.02 48	4.90 41	2.18 55	0.93 40	2.92 12	0.37 11	15.10 92	16.91 91	7.89 39	10.51 38	7.03 26	3.07 21	8.16 42	3.46 30	0.01 27	0.00 1	0.00 1	0.00 1	0.10 33	0.02 6

HGLStereo	4.25 45	0.52 14	4.65 39	3.09 92	5.18 118	6.51 55	0.73 33	10.37 38	6.74 23	7.32 34	12.44 56	7.00 25	7.21 68	7.46 28	5.51 48	0.00 1	0.02 47	0.00 1	0.00 1	0.06 24	0.10 42

iResNetv2_ROB	4.28 46	1.43 74	7.17 76	2.91 84	1.26 47	4.36 29	1.62 48	13.64 78	10.25 58	9.83 60	11.41 45	7.68 33	4.00 32	7.75 31	1.85 8	0.00 1	0.00 1	0.00 1	0.00 1	0.37 82	0.09 35

StereoDRNet-Refined	4.46 47	0.62 25	3.80 28	1.92 41	0.40 14	9.35 71	0.15 5	10.02 28	8.83 47	12.69 81	11.62 49	9.34 44	3.87 30	8.06 38	8.02 66	0.00 1	0.00 1	0.01 54	0.05 76	0.20 56	0.26 73
Rohan Chabra, Julian Straub, Chris Sweeney, Richard Newcombe, Henry Fuchs: StereoDRNet. CVPR
NVstereo2D	4.51 48	0.82 34	6.86 74	3.28 97	3.38 93	8.16 66	3.13 60	10.51 41	15.15 81	4.90 11	6.89 15	7.87 34	4.78 47	9.88 63	3.91 36	0.01 27	0.00 1	0.00 1	0.06 78	0.02 7	0.58 96

DLCB_ROB	4.51 48	0.91 40	3.78 27	2.19 56	1.07 44	6.28 49	3.09 59	9.78 27	7.72 36	10.65 67	12.97 59	13.91 73	3.71 29	8.72 48	5.30 47	0.00 1	0.00 1	0.00 1	0.00 1	0.03 13	0.10 42

RASNet	4.52 50	0.61 24	4.42 34	3.42 101	4.68 115	4.58 34	0.99 38	9.54 26	8.01 40	5.28 15	11.42 46	10.34 52	8.88 79	9.28 58	8.68 76	0.15 76	0.00 1	0.00 1	0.00 1	0.03 13	0.04 24

SGM-Forest	4.96 51	0.32 1	2.84 12	1.21 12	0.64 21	10.23 81	6.64 100	11.55 53	10.98 60	10.94 71	13.59 62	11.65 61	4.30 38	8.94 53	4.63 41	0.11 69	0.04 58	0.00 1	0.00 1	0.05 19	0.46 87
Johannes L. Schönberger, Sudipta Sinha, Marc Pollefeys: Learning to Fuse Proposals from Multiple Scanline Optimizations in Semi-Global Matching. ECCV 2018
PA-Net	4.98 52	1.47 76	7.42 78	2.40 67	2.14 70	8.73 68	3.64 69	12.42 64	13.11 72	7.03 32	7.57 19	7.88 35	6.52 63	10.16 66	7.82 64	0.02 34	0.03 53	0.00 1	0.00 1	0.11 37	1.07 112
Zhibo Rao, Mingyi He, Yuchao Dai, Zhelun Shen: Patch Attention Network with Generative Adversarial Model for Semi-Supervised Binocular Disparity Prediction.
AANet_RVC	5.01 53	1.74 89	6.38 64	1.96 46	1.29 49	2.26 3	1.69 50	10.07 29	18.53 94	7.88 38	18.15 85	8.49 41	2.70 17	10.59 70	7.04 57	0.96 119	0.15 93	0.02 66	0.00 1	0.13 42	0.12 47

PSMNet_ROB	5.02 54	1.63 86	6.03 58	1.90 40	1.83 64	9.57 75	6.35 97	15.58 97	7.23 28	6.15 21	10.48 37	12.22 64	4.16 36	8.02 35	8.71 77	0.02 34	0.01 36	0.01 54	0.10 86	0.20 56	0.12 47

CBMV	5.35 55	0.91 40	3.67 26	1.62 28	0.44 16	10.09 79	7.19 104	12.49 65	12.33 71	12.22 77	14.69 69	10.93 56	6.48 62	8.51 45	4.96 44	0.02 34	0.15 93	0.00 1	0.00 1	0.17 51	0.17 57
Konstantinos Batsos, Changjiang Cai, Philippos Mordohai: CBMV: A Coalesced Bidirectional Matching Volume for Disparity Estimation. Computer Vision and Pattern Recognition (CVPR) 2018
StereoDRNet	5.59 56	1.75 90	6.80 72	3.12 93	4.45 110	10.61 84	4.35 77	18.80 109	9.73 51	12.22 77	6.87 14	11.44 59	4.65 43	8.09 40	8.26 72	0.02 34	0.11 84	0.00 1	0.03 65	0.20 56	0.28 76

ETE_ROB	5.80 57	1.77 91	6.33 62	1.44 20	0.78 31	6.43 53	6.90 101	12.53 66	8.08 41	12.93 85	14.89 70	21.13 108	5.87 59	9.83 62	6.57 55	0.04 53	0.01 36	0.00 1	0.02 56	0.08 28	0.33 78

DRN-Test	5.87 58	0.98 46	5.89 56	2.69 78	3.65 101	12.37 92	3.35 63	20.07 120	10.20 57	11.93 76	12.31 55	11.06 58	5.31 53	7.89 33	9.05 80	0.04 53	0.05 62	0.04 81	0.04 74	0.18 54	0.25 70

NCCL2	5.88 59	1.59 83	5.44 48	1.87 37	0.92 38	9.55 74	11.55 121	12.11 59	9.94 55	9.67 59	8.85 27	22.28 110	7.41 69	8.78 49	7.17 58	0.01 27	0.00 1	0.03 74	0.00 1	0.13 42	0.23 64

NaN_ROB	6.00 60	1.24 64	6.29 61	1.34 15	1.68 59	9.60 76	10.31 117	15.09 90	15.79 84	12.62 80	8.95 28	11.67 62	5.83 58	11.78 79	6.41 54	0.05 60	0.13 90	0.08 90	0.20 95	0.22 65	0.79 104

DANet	6.02 61	1.23 63	8.45 83	3.86 110	3.94 104	7.64 65	1.34 46	9.51 25	7.00 24	13.39 87	15.53 74	15.99 84	7.02 67	12.14 80	12.37 100	0.19 82	0.12 89	0.02 66	0.03 65	0.13 42	0.56 95

XPNet_ROB	6.03 62	1.22 62	5.61 52	2.56 74	0.90 36	6.32 50	7.07 102	12.92 69	8.30 43	14.76 92	15.13 72	19.84 102	6.66 66	10.36 67	8.58 75	0.02 34	0.04 58	0.00 1	0.03 65	0.11 37	0.24 67

Anonymous Stereo	6.16 63	3.15 112	23.75 121	2.97 90	2.48 80	4.39 32	13.30 123	9.21 22	9.86 53	9.56 58	8.76 26	6.79 24	1.99 11	13.50 92	13.04 103	0.01 27	0.05 62	0.00 1	0.06 78	0.22 65	0.19 59

GANet	6.22 64	1.07 52	4.07 32	2.27 60	0.89 34	9.19 70	9.52 112	12.02 58	8.13 42	10.72 68	29.09 111	13.86 71	7.52 71	11.00 74	4.39 37	0.36 97	0.00 1	0.02 66	0.02 56	0.12 39	0.08 31

DISCO	6.28 65	0.57 20	5.78 54	3.43 102	1.17 45	11.22 86	3.39 64	12.14 61	16.16 86	6.52 28	11.22 42	16.96 87	6.32 61	19.51 115	10.74 93	0.00 1	0.00 1	0.00 1	0.00 1	0.35 80	0.11 45

RYNet	6.34 66	0.89 39	5.88 55	1.41 18	4.48 112	15.97 103	4.18 73	13.41 74	16.49 87	10.81 70	7.00 17	14.33 75	8.72 77	9.43 60	13.71 104	0.00 1	0.01 36	0.00 1	0.00 1	0.02 7	0.07 28

GwcNet	6.42 67	1.97 97	10.92 90	2.59 75	5.58 121	11.55 88	2.21 54	14.10 82	16.52 90	10.04 63	17.19 82	10.86 55	5.61 56	9.23 57	8.84 79	0.01 27	0.06 68	0.03 74	0.08 82	0.56 91	0.40 82

GANetREF_RVC	6.56 68	2.89 107	7.58 80	3.41 100	0.40 14	12.96 95	9.58 113	15.09 90	17.25 93	10.33 64	10.62 39	12.27 65	8.16 73	12.21 81	4.53 40	0.41 101	0.00 1	0.00 1	0.02 56	3.12 126	0.39 81
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_ROB	6.58 69	1.80 93	6.25 60	1.26 13	0.94 41	10.08 78	9.02 108	16.00 98	11.51 66	12.74 82	13.02 60	24.77 112	5.25 52	10.56 69	8.02 66	0.04 53	0.05 62	0.00 1	0.02 56	0.10 33	0.25 70

S-Stereo	6.63 70	0.84 35	9.67 87	3.15 94	3.48 95	6.49 54	6.22 92	12.99 70	22.84 107	9.48 54	15.51 73	12.00 63	8.43 74	8.04 36	10.70 92	0.12 72	0.17 97	0.00 1	0.38 109	0.13 42	1.92 121

DeepPrunerF	6.75 71	2.69 105	23.31 120	3.68 105	7.16 127	3.78 21	4.29 75	13.42 75	20.13 104	8.13 40	10.46 36	7.18 32	8.06 72	11.10 75	9.44 82	0.24 86	0.15 93	0.29 112	0.42 112	0.66 98	0.45 85

edge stereo	6.76 72	1.01 47	6.76 71	2.20 57	2.45 79	6.41 52	2.45 56	14.84 88	11.98 70	15.29 93	18.31 86	22.02 109	12.56 94	10.82 71	7.49 60	0.03 45	0.06 68	0.11 95	0.03 65	0.30 73	0.14 52

NCC-stereo	6.77 73	1.49 77	6.48 66	2.92 86	4.40 106	7.43 61	3.61 67	19.52 116	13.29 73	8.39 43	16.91 77	15.96 82	12.13 92	12.85 86	7.70 62	1.47 123	0.11 84	0.01 54	0.42 112	0.14 47	0.24 67

Abc-Net	6.77 73	1.49 77	6.48 66	2.92 86	4.40 106	7.43 61	3.61 67	19.52 116	13.29 73	8.39 43	16.91 77	15.96 82	12.13 92	12.85 86	7.70 62	1.47 123	0.11 84	0.01 54	0.42 112	0.14 47	0.24 67
Xing Li, Yangyu Fan, Guoyun Lv, and Haoyue Ma: Area-based Correlation and Non-local Attention Network for Stereo Matching. The Visual Computer
FAT-Stereo	6.78 75	0.68 29	6.80 72	2.30 62	1.77 63	5.63 45	4.20 74	18.79 108	18.62 95	10.53 66	17.15 80	18.52 96	13.74 102	8.86 52	7.38 59	0.03 45	0.15 93	0.01 54	0.07 80	0.12 39	0.26 73

RP	6.84 76	1.29 71	5.53 50	3.92 111	5.18 118	6.32 50	3.53 65	11.73 55	15.31 82	9.54 57	22.38 94	18.25 94	14.47 103	10.11 65	7.49 60	0.91 118	0.01 36	0.12 96	0.15 91	0.33 77	0.19 59

RGC	6.88 77	2.23 101	6.13 59	4.05 112	4.73 117	8.94 69	2.78 58	15.19 94	11.74 68	11.13 72	19.34 89	17.86 91	10.42 86	13.02 89	8.03 68	0.73 112	0.01 36	0.24 111	0.41 111	0.31 75	0.38 80

Nwc_Net	6.97 78	1.25 65	6.63 69	3.82 109	3.37 92	10.83 85	1.67 49	19.56 118	11.35 63	8.36 42	23.62 97	17.19 89	11.44 91	11.21 76	8.08 70	0.80 114	0.00 1	0.00 1	0.02 56	0.13 42	0.09 35

STTRV1_RVC	7.02 79	1.10 56	12.88 101	3.32 98	6.92 126	11.90 91	4.00 71	15.07 89	11.94 69	9.51 55	14.57 67	11.63 60	8.73 78	12.65 85	8.06 69	3.32 129	2.75 130	0.41 120	0.12 88	1.38 116	0.11 45

ADCRef	7.27 80	1.38 72	16.37 109	2.52 71	3.30 91	11.63 89	3.16 61	10.80 47	9.35 49	13.03 86	25.27 105	8.17 37	8.92 80	8.06 38	21.81 119	0.15 76	0.08 77	0.16 102	0.34 107	0.38 83	0.58 96

psmorigin	7.34 81	1.58 82	18.31 113	2.35 66	0.87 33	6.72 56	1.70 52	10.63 44	7.14 26	19.77 105	22.71 95	20.13 104	11.34 90	12.96 88	9.49 83	0.04 53	0.17 97	0.06 86	0.17 92	0.21 63	0.50 90

CSAN	7.62 82	1.60 84	6.56 68	1.83 35	0.66 22	12.40 93	10.52 119	14.45 85	21.32 106	14.19 89	15.98 76	17.84 90	13.02 99	12.32 82	8.38 73	0.09 66	0.07 75	0.03 74	0.04 74	0.33 77	0.67 102

stereogan	7.69 83	0.88 37	7.08 75	3.49 103	3.93 103	18.98 111	3.23 62	16.52 100	19.58 101	9.93 62	18.92 87	20.50 106	9.04 81	14.07 96	6.14 50	0.26 89	0.04 58	0.21 108	0.03 65	0.63 97	0.33 78

pmcnn	7.72 84	1.27 67	9.42 86	2.91 84	3.14 88	9.44 72	6.23 93	12.56 67	16.51 88	14.53 90	24.08 99	27.44 118	8.49 75	9.32 59	8.44 74	0.06 62	0.08 77	0.00 1	0.00 1	0.30 73	0.15 55

AF-Net	7.78 85	1.44 75	6.68 70	3.37 99	4.50 113	8.61 67	2.69 57	17.07 103	20.17 105	9.52 56	24.02 98	20.31 105	14.59 104	11.58 78	9.84 88	0.61 110	0.00 1	0.12 96	0.00 1	0.38 83	0.12 47

PASM	7.90 86	4.22 118	21.97 118	3.25 96	3.29 90	5.39 41	6.57 99	10.57 42	19.09 97	12.77 83	13.92 64	18.11 93	9.51 83	13.79 95	10.77 95	0.19 82	0.45 116	0.29 112	1.08 125	1.49 118	1.19 114

PWCDC_ROB	7.92 87	3.17 114	7.48 79	5.73 123	4.40 106	10.45 82	0.35 10	14.52 86	28.19 115	10.36 65	31.27 115	7.04 27	9.14 82	13.22 91	8.78 78	2.74 128	0.02 47	0.00 1	0.00 1	1.31 115	0.17 57

ADCP+	8.09 88	1.79 92	14.50 107	1.54 23	4.28 105	16.57 105	5.20 86	12.80 68	11.20 62	12.83 84	17.07 79	11.02 57	10.80 88	17.59 111	23.18 122	0.03 45	0.05 62	0.01 54	0.18 93	0.39 87	0.81 105

SuperB	8.10 89	3.15 112	24.67 122	2.65 76	1.23 46	9.88 77	4.29 75	10.18 36	30.07 118	11.53 74	12.18 52	6.12 21	6.65 65	10.50 68	14.47 106	0.14 74	0.11 84	0.35 116	0.25 99	13.06 135	0.48 88

PWC_ROB	8.24 90	3.13 110	12.74 100	2.43 69	4.43 109	7.51 64	1.22 41	16.63 101	19.24 99	16.08 95	28.29 108	13.99 74	10.16 85	13.63 94	14.06 105	0.42 104	0.00 1	0.05 84	0.00 1	0.59 94	0.27 75

MDST_ROB	8.37 91	0.32 1	9.03 84	4.18 115	2.42 78	26.86 125	6.14 90	19.36 113	13.52 75	27.09 120	22.75 96	9.47 45	4.74 45	15.06 102	6.34 53	0.02 34	0.02 47	0.00 1	0.00 1	0.02 7	0.13 51

G-Net	8.41 92	1.54 80	10.97 91	5.73 123	3.60 98	26.19 122	4.41 78	10.10 31	7.42 30	19.71 103	24.99 103	14.38 76	15.83 105	10.99 72	9.53 84	0.50 107	0.46 117	0.19 106	0.25 99	0.80 102	0.66 100

STTStereo_v2	8.41 92	1.54 80	10.97 91	5.73 123	3.60 98	26.19 122	4.41 78	10.10 31	7.42 30	19.71 103	24.99 103	14.38 76	15.83 105	10.99 72	9.53 84	0.50 107	0.46 117	0.19 106	0.25 99	0.80 102	0.66 100

XQC	8.43 94	3.58 116	16.40 110	2.92 86	2.17 71	13.22 97	3.60 66	14.64 87	25.86 112	11.87 75	12.04 50	15.06 79	10.67 87	15.24 103	19.41 111	0.39 99	0.08 77	0.05 84	0.07 80	0.84 104	0.45 85

FBW_ROB	8.50 95	1.03 49	7.98 82	1.93 43	1.28 48	13.10 96	6.23 93	22.50 124	18.98 96	18.82 100	14.91 71	19.06 99	10.04 84	18.41 112	9.83 87	0.62 111	0.22 102	1.82 131	0.82 122	0.99 108	1.36 116

aanetorigin	8.72 96	3.29 115	27.55 124	1.95 45	3.84 102	4.93 39	5.39 88	5.49 3	10.05 56	27.85 121	31.47 116	15.80 81	12.78 96	9.04 54	11.98 98	0.25 87	0.22 102	0.22 109	0.25 99	1.28 114	0.84 107

RTSC	9.15 97	3.00 109	13.57 105	3.72 107	1.76 62	11.82 90	0.46 13	16.95 102	36.83 127	15.80 94	15.53 74	12.91 68	7.46 70	20.01 118	21.76 118	0.31 94	0.13 90	0.01 54	0.08 82	0.57 92	0.41 84

WCMA_ROB	9.21 98	0.87 36	7.37 77	2.54 73	2.13 69	13.59 98	5.80 89	11.64 54	14.01 77	24.43 117	32.99 121	27.09 117	18.02 109	12.51 84	9.85 89	0.81 115	0.07 75	0.01 54	0.01 52	0.16 49	0.23 64

MSMD_ROB	9.28 99	1.09 55	4.65 39	1.58 25	0.39 13	16.52 104	4.41 78	13.60 76	14.87 80	22.34 109	39.89 129	25.67 114	20.71 118	12.42 83	6.98 56	0.34 96	0.03 53	0.00 1	0.00 1	0.05 19	0.09 35

ADCPNet	9.54 100	2.39 103	31.46 126	2.09 51	1.60 55	16.71 107	6.39 98	12.11 59	11.45 64	13.53 88	21.45 92	19.41 100	10.94 89	14.38 98	21.54 117	0.27 92	1.16 124	0.39 119	1.49 129	0.58 93	1.45 117

SHD	9.61 101	2.60 104	12.46 98	3.69 106	3.54 97	9.47 73	1.25 42	20.16 121	37.84 130	18.19 99	21.24 91	16.96 87	12.83 98	14.47 100	16.05 108	0.32 95	0.13 90	0.01 54	0.08 82	0.38 83	0.48 88

PDISCO_ROB	9.62 102	1.99 98	11.51 93	9.88 131	9.61 132	21.48 115	3.83 70	19.33 112	28.49 116	11.27 73	14.17 65	19.92 103	5.02 50	16.35 108	9.18 81	5.28 132	0.41 112	0.14 100	0.09 85	2.05 122	2.36 127

MFN_U_SF_DS_RVC	9.78 103	4.27 119	14.47 106	2.29 61	2.85 86	23.40 120	13.62 124	13.60 76	19.54 100	19.42 102	24.27 100	16.74 86	8.59 76	17.05 110	7.98 65	1.25 122	1.68 128	0.17 103	2.63 131	0.72 100	1.04 110

SGM_RVC	10.08 104	0.60 22	3.42 21	2.30 62	0.32 11	19.41 112	6.33 96	18.95 110	14.64 78	25.14 119	24.32 101	33.34 126	18.79 113	19.86 116	12.55 102	0.25 87	0.26 107	0.22 109	0.24 98	0.34 79	0.40 82
Heiko Hirschmueller: Stereo processing by semiglobal matching and mutual information. TPAMI 2008, Volume 30(2), pp. 328-341
DPSNet	10.14 105	1.88 96	16.82 111	1.85 36	1.73 61	24.84 121	17.20 133	19.92 119	27.41 114	12.23 79	13.62 63	16.52 85	18.35 110	14.42 99	12.50 101	0.78 113	0.54 121	0.08 90	0.25 99	1.18 111	0.59 99

ADCL	10.16 106	2.11 99	19.36 115	1.92 41	1.88 66	22.23 116	8.91 107	14.04 81	23.56 109	14.62 91	26.19 106	12.75 67	13.59 101	16.06 107	22.95 121	0.26 89	0.18 99	0.75 123	0.65 118	0.69 99	0.58 96

FCDSN-DC	10.24 107	0.56 19	3.49 22	1.96 46	1.29 49	16.90 108	4.59 83	17.16 104	19.10 98	24.64 118	32.46 120	33.82 127	22.14 123	15.93 106	10.45 90	0.04 53	0.00 1	0.00 1	0.00 1	0.05 19	0.21 61
Dominik Hirner, Friedrich Fraundorfer: FCDSN-DC: An accurate but lightweight end-to-end trainable neural network for stereo estimation with depth completion.
ADCMid	10.24 107	3.13 110	20.70 116	2.21 58	2.39 77	11.23 87	6.19 91	14.17 83	11.19 61	23.20 115	22.25 93	17.89 92	19.54 115	18.51 113	26.21 127	0.45 105	0.42 114	1.10 127	1.29 126	1.56 120	1.18 113

SANet	10.64 109	1.86 95	10.91 89	1.76 33	0.71 25	14.62 101	9.23 111	19.18 111	37.14 128	19.22 101	27.96 107	25.86 115	19.11 114	13.02 89	10.63 91	0.08 65	0.06 68	0.03 74	0.02 56	0.62 96	0.81 105

FC-DCNN	10.72 110	0.52 14	4.27 33	1.88 38	1.63 56	17.18 109	5.29 87	18.20 106	19.69 102	28.50 122	34.51 124	34.03 128	21.48 121	15.89 105	11.15 97	0.03 45	0.01 36	0.02 66	0.01 52	0.07 26	0.09 35

AnyNet_C32	10.98 111	5.58 121	22.79 119	4.16 113	5.83 122	15.64 102	14.30 125	13.18 72	17.15 92	16.44 97	20.52 90	14.68 78	13.44 100	22.46 120	30.08 132	0.17 80	0.26 107	0.36 117	0.36 108	1.23 112	0.91 108

MeshStereo	11.52 112	1.52 79	4.55 37	1.89 39	1.46 51	19.87 114	5.11 85	20.66 122	15.91 85	32.67 127	34.51 124	39.34 133	21.15 119	18.74 114	12.10 99	0.11 69	0.06 68	0.01 54	0.00 1	0.45 90	0.22 63
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_RVC	12.94 113	3.66 117	25.81 123	3.61 104	2.26 74	22.77 118	4.55 82	27.10 129	20.06 103	23.90 116	28.99 110	30.53 123	16.98 107	19.92 117	20.26 113	1.24 121	1.07 123	0.98 126	1.33 127	1.80 121	2.04 123

ADCS	13.02 114	4.93 120	28.38 125	3.17 95	2.67 83	13.61 99	10.83 120	18.70 107	33.46 122	22.59 110	24.78 102	19.59 101	18.51 112	23.40 123	32.16 134	0.10 68	0.19 100	0.37 118	0.18 93	1.26 113	1.46 118

MFMNet_re	13.29 115	8.60 128	18.29 112	9.75 130	7.25 129	19.65 113	14.84 129	20.71 123	30.72 119	23.03 113	28.77 109	18.85 98	26.09 128	13.55 93	9.82 86	2.44 126	1.35 127	0.34 115	0.23 97	4.78 130	6.69 132

EDNetEfficient	13.82 116	6.19 123	54.79 133	2.83 81	3.60 98	4.86 37	8.57 106	8.61 18	32.90 121	40.77 133	30.59 114	18.33 95	21.87 122	11.38 77	25.36 125	0.15 76	0.23 104	0.06 86	1.03 124	2.54 124	1.79 119

LSM	14.01 117	5.95 122	33.49 127	6.78 128	43.61 137	10.22 80	9.98 116	15.16 93	22.93 108	23.07 114	32.34 119	18.52 96	12.67 95	15.45 104	11.10 96	0.16 79	0.51 120	0.09 93	0.32 105	1.08 109	16.85 137

SAMSARA	14.63 118	2.74 106	12.38 97	12.65 134	6.74 125	36.50 131	72.93 139	19.36 113	23.77 110	16.20 96	13.04 61	29.21 120	12.78 96	16.98 109	15.21 107	0.11 69	0.26 107	0.03 74	0.14 90	0.76 101	0.77 103

SPS-STEREO	15.04 119	6.23 124	13.21 104	11.34 133	11.65 135	23.30 119	7.15 103	24.16 126	15.65 83	31.78 126	29.19 112	31.62 124	21.32 120	24.62 124	19.50 112	7.59 134	4.19 135	3.22 132	1.48 128	6.99 134	6.54 131
K. Yamaguchi, D. McAllester, R. Urtasun: Efficient Joint Segmentation, Occlusion Labeling, Stereo and Flow Estimation. ECCV 2014
PVD	15.44 120	2.93 108	14.67 108	4.21 116	3.39 94	17.43 110	4.16 72	27.84 132	48.84 135	31.02 125	43.54 133	29.76 122	30.81 132	25.97 126	21.40 116	0.23 85	0.41 112	0.04 81	0.33 106	0.41 88	1.33 115

SGM+DAISY	15.62 121	7.26 127	19.28 114	8.94 129	10.11 134	26.25 124	10.49 118	19.36 113	14.65 79	30.64 124	33.59 122	33.00 125	22.32 124	24.96 125	16.42 109	7.90 135	6.25 136	4.51 133	3.37 132	5.86 132	7.20 133

NVStereoNet_ROB	16.04 122	6.75 125	12.90 102	6.37 127	7.42 130	12.89 94	9.74 114	22.78 125	25.12 111	30.32 123	46.19 135	34.37 129	25.38 126	21.48 119	21.38 115	5.94 133	3.10 133	6.07 134	10.09 136	4.01 128	8.54 135
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_C01	16.12 123	10.81 132	59.36 134	4.42 117	2.49 81	30.06 127	15.15 131	17.51 105	16.51 88	17.88 98	37.69 128	24.04 111	17.54 108	29.60 129	33.29 135	0.28 93	0.38 110	0.43 121	0.42 112	2.57 125	1.98 122

MSC_U_SF_DS_RVC	16.41 124	6.93 126	21.83 117	5.94 126	2.81 85	38.71 132	14.59 126	24.55 127	34.87 125	33.66 128	34.35 123	29.24 121	24.20 125	22.59 121	17.95 110	2.52 127	2.81 131	1.17 129	1.51 130	5.89 133	2.16 124

ELAS_RVC	16.54 125	2.26 102	10.09 88	5.50 122	4.46 111	28.28 126	16.72 132	25.55 128	33.54 123	40.19 131	40.30 131	36.68 131	30.03 130	29.40 128	20.61 114	0.98 120	1.21 125	0.86 125	0.70 120	1.39 117	2.16 124
A. Geiger, M. Roser, R. Urtasun: Efficient large-scale stereo matching. ACCV 2010
ELAS	16.72 126	2.14 100	9.23 85	4.92 119	4.53 114	32.66 130	15.11 130	27.40 130	28.68 117	40.27 132	44.90 134	38.33 132	30.50 131	26.44 127	21.94 120	0.88 116	1.23 126	0.67 122	0.89 123	1.49 118	2.18 126
A. Geiger, M. Roser, R. Urtasun: Efficient large-scale stereo matching. ACCV 2010
LE_ROB	16.73 127	1.28 70	11.61 94	3.72 107	1.65 57	16.67 106	9.17 109	14.39 84	55.91 137	63.81 137	40.86 132	35.94 130	37.73 136	14.24 97	26.87 128	0.05 60	0.10 82	0.13 99	0.22 96	0.12 39	0.15 55

SGM-ForestM	16.99 128	1.08 54	5.74 53	2.12 53	0.75 27	31.63 129	12.21 122	27.80 131	32.25 120	37.88 129	39.99 130	52.96 136	35.20 135	33.60 131	24.47 123	0.26 89	0.39 111	0.31 114	0.39 110	0.26 71	0.53 94

DispFullNet	17.47 129	26.01 135	33.98 128	22.58 136	20.86 136	13.84 100	1.28 43	16.50 99	26.27 113	19.97 106	17.17 81	20.52 107	18.49 111	22.86 122	10.76 94	5.13 131	2.83 132	30.72 136	7.72 134	20.86 136	11.01 136

EDNetEfficientorigin	17.87 130	14.00 133	99.53 142	1.64 30	0.92 38	10.55 83	6.26 95	13.21 73	34.84 124	38.20 130	36.91 127	27.47 119	26.32 129	14.71 101	25.69 126	0.04 53	0.06 68	0.14 100	0.52 117	3.16 127	3.27 128

RTSA	18.87 131	9.32 130	86.48 137	4.95 120	6.10 123	42.08 134	14.70 127	15.49 95	41.06 132	22.65 111	32.32 117	13.77 69	19.54 115	37.98 132	28.96 129	0.41 101	0.23 104	0.00 1	0.02 56	0.91 105	0.50 90

RTS	18.87 131	9.32 130	86.48 137	4.95 120	6.10 123	42.08 134	14.70 127	15.49 95	41.06 132	22.65 111	32.32 117	13.77 69	19.54 115	37.98 132	28.96 129	0.41 101	0.23 104	0.00 1	0.02 56	0.91 105	0.50 90

MANE	19.47 133	1.27 67	5.07 44	4.69 118	5.55 120	30.49 128	9.94 115	34.01 134	37.27 129	44.13 134	51.57 138	52.51 135	40.41 137	33.58 130	24.81 124	0.89 117	0.86 122	1.11 128	9.72 135	0.38 83	1.06 111

BEATNet-Init1	23.31 134	9.03 129	41.67 129	4.17 114	2.53 82	45.68 136	19.47 134	33.43 133	38.45 131	47.59 136	49.10 136	59.31 137	41.80 138	38.35 134	29.21 131	0.47 106	0.50 119	0.81 124	0.66 119	2.10 123	1.86 120

FADE	25.68 135	17.27 134	50.60 132	10.46 132	9.90 133	22.50 117	9.17 109	35.80 136	53.05 136	20.32 108	19.01 88	26.54 116	34.42 133	39.35 135	33.52 136	30.62 137	14.22 137	38.39 137	37.63 137	5.22 131	5.56 130

MADNet+	27.07 136	33.84 136	90.97 139	20.14 135	7.47 131	48.43 137	47.10 136	35.43 135	36.46 126	20.11 107	30.05 113	25.29 113	35.08 134	45.50 137	50.28 137	2.13 125	2.00 129	1.19 130	0.76 121	4.71 129	4.43 129

PWCK	30.53 137	44.32 137	47.25 131	29.76 137	7.23 128	40.78 133	27.10 135	44.73 137	44.32 134	47.31 135	36.37 126	47.16 134	26.05 127	41.26 136	31.87 133	21.83 136	4.03 134	29.50 135	4.67 133	27.17 137	7.80 134

DPSimNet_ROB	53.45 138	64.73 138	44.39 130	53.97 138	45.39 138	53.66 138	54.83 137	55.15 138	57.87 138	64.16 138	50.83 137	63.40 138	53.34 139	46.45 138	65.81 138	63.13 139	26.54 138	57.94 139	51.11 138	45.52 138	50.69 138

ASD4	75.85 139	81.57 139	69.25 135	68.23 139	80.87 139	96.24 140	99.15 142	77.50 139	93.47 140	77.59 139	85.63 139	78.32 139	76.43 140	77.89 139	82.61 139	58.30 138	62.67 139	55.02 138	62.43 139	68.35 139	65.49 139

MADNet++	82.84 140	82.38 140	73.57 136	87.72 140	82.97 140	93.14 139	69.15 138	86.42 140	82.50 139	93.46 140	86.70 140	86.28 140	80.92 141	88.34 140	88.84 140	86.83 140	84.17 140	72.64 140	68.92 140	80.47 140	81.42 140

MEDIAN_ROB	98.41 141	99.70 141	99.30 141	97.09 141	97.02 141	96.89 141	95.77 141	97.66 141	97.28 141	98.79 143	98.94 141	99.18 141	98.14 142	96.89 141	96.88 141	99.96 143	99.16 141	100.00 141	99.99 141	99.69 141	99.88 141

AVERAGE_ROB	99.62 142	99.95 142	98.81 140	100.00 146	100.00 142	98.08 142	95.47 140	100.00 144	100.00 142	100.00 144	100.00 142	100.00 144	100.00 143	100.00 144	99.99 142	100.00 145	100.00 142	100.00 141	100.00 142	100.00 144	100.00 145

DGTPSM_ROB	99.90 143	100.00 143	99.99 143	99.99 144	100.00 142	100.00 143	100.00 143	99.97 142	100.00 142	98.35 141	100.00 142	99.84 142	100.00 143	99.98 142	99.99 142	99.99 144	100.00 142	100.00 141	100.00 142	100.00 144	100.00 145

DPSMNet_ROB	99.91 144	100.00 143	99.99 143	99.99 144	100.00 142	100.00 143	100.00 143	99.98 143	100.00 142	98.35 141	100.00 142	99.84 142	100.00 143	99.98 142	99.99 142	100.00 145	100.00 142	100.00 141	100.00 142	100.00 144	100.00 145

DPSM_ROB	99.95 145	100.00 143	100.00 145	99.76 142	100.00 142	100.00 143	100.00 143	100.00 144	100.00 142	100.00 144	100.00 142	100.00 144	100.00 143	100.00 144	100.00 145	99.21 141	100.00 142	100.00 141	100.00 142	99.99 142	99.95 142

DPSM	99.95 145	100.00 143	100.00 145	99.76 142	100.00 142	100.00 143	100.00 143	100.00 144	100.00 142	100.00 144	100.00 142	100.00 144	100.00 143	100.00 144	100.00 145	99.21 141	100.00 142	100.00 141	100.00 142	99.99 142	99.95 142

LSM0	100.00 147	100.00 143	100.00 145	100.00 146	100.00 142	100.00 143	100.00 143	100.00 144	100.00 142	100.00 144	100.00 142	100.00 144	100.00 143	100.00 144	100.00 145	100.00 145	100.00 142	100.00 141	100.00 142	100.00 144	99.99 144

MSMDNet													1.26 7

Low-res two-view benchmark