The KITTI Vision Benchmark Suite

The stereo / flow benchmark consists of 194 training image pairs and 195 test image pairs, saved in loss less png format. Our evaluation server computes the average number of bad pixels for all non-occluded or occluded (=all groundtruth) pixels. We require that all methods use the same parameter set for all test pairs. Our development kit provides details about the data format as well as MATLAB / C++ utility functions for reading and writing disparity maps and flow fields.

• Download stereo/optical flow data set (2 GB)
• Download stereo/optical flow calibration files (1 MB)
• Download multi-view extension (20 frames per scene, all cameras) (17 GB)
• Download stereo/optical flow development kit (3 MB)
• Semantic and instance labels for 60 images and car labels for all training images (1 MB)

Our evaluation table ranks all methods according to the number of non-occluded erroneous pixels at the specified disparity / end-point error threshold. All methods providing less than 100 % density have been interpolated using simple background interpolation as explained in the corresponding header file in the development kit. For each method we show:

• Out-Noc: Percentage of erroneous pixels in non-occluded areas
• Out-All: Percentage of erroneous pixels in total
• Avg-Noc: Average disparity / end-point error in non-occluded areas
• Avg-All: Average disparity / end-point error in total
• Density: Percentage of pixels for which ground truth has been provided by the method

Note: On 04.11.2013 we have improved the ground truth disparity maps and flow fields leading to slightly improvements for all methods. Please download the stereo/flow dataset with the improved ground truth for training again, if you have downloaded the dataset prior to 04.11.2013. Please consider reporting these new number for all future submissions. Links to last leaderboards before the updates: stereo and flow!

	Method	Setting	Code	Out-Noc	Out-All	Avg-Noc	Avg-All	Runtime	Environment
1	HART			3.14 %	3.92 %	0.8 px	0.8 px	0.34 s	GPU @ 2.5 Ghz (Python)
2	StereoBase		code	3.29 %	4.07 %	0.8 px	0.9 px	0.29 s	GPU @ 1.5 Ghz (Python)
X. Guo, J. Lu, C. Zhang, Y. Wang, Y. Duan, T. Yang, Z. Zhu and L. Chen: OpenStereo: A Comprehensive Benchmark for Stereo Matching and Strong Baseline. arXiv preprint arXiv:2312.00343 2023.
3	LoS			3.47 %	4.45 %	0.8 px	0.9 px	0.19 s	1 core @ 2.5 Ghz (Python)
4	EGLCR-Stereo			3.62 %	4.59 %	0.9 px	1.0 px	0.45 s	1 core @ 2.5 Ghz (C/C++)
5	MoCha-V2			3.65 %	4.50 %	0.9 px	0.9 px	0.31 s	NVIDIA Tesla A6000 (Python)
6	Selective-IGEV			3.79 %	4.38 %	0.9 px	1.0 px	0.24 s	1 core @ 2.5 Ghz (Python)
7	MoCha-Stereo			3.83 %	4.50 %	0.8 px	0.9 px	0.33 s	NVIDIA Tesla A6000 (PyTorch)
8	MPFV-Stereo			3.88 %	4.84 %	0.8 px	0.9 px	0.31 s	1 core @ 2.5 Ghz (Python)
9	DR Stereo			3.96 %	4.58 %	0.9 px	1.0 px	0.18 s	1 core @ 2.5 Ghz (C/C++)
10	4D-IteraStereo			4.01 %	4.61 %	0.9 px	1.0 px	0.4 s	GPU @ 2.5 Ghz (Python)
11	GeoNet			4.03 %	4.99 %	0.9 px	1.0 px	0.22 s	1 core @ 2.5 Ghz (C/C++)
12	MC-Stereo			4.10 %	5.30 %	1.0 px	1.1 px	0.40 s	1 core @ 2.5 Ghz (Python)
13	IGEV-Stereo(32)		code	4.11 %	4.76 %	0.9 px	1.0 px	0.32 s	NVIDIA RTX 3090 (PyTorch)
G. Xu, X. Wang, X. Ding and X. Yang: Iterative Geometry Encoding Volume for Stereo Matching. CVPR 2023.
14	RiskMin			4.11 %	5.51 %	0.9 px	1.1 px	0.20 s	GPU @ 2.5 Ghz (Python)
15	IGEV-ICGNet			4.12 %	5.09 %	0.9 px	1.0 px	0.18 s	GPU @ 2.5 Ghz (C/C++)
16	Any-IGEV			4.16 %	4.74 %	1.0 px	1.1 px	0.32 s	1 core @ 2.5 Ghz (C/C++)
17	IGE_Corr			4.19 %	5.16 %	0.9 px	1.0 px	0.2 s	1 core @ 2.5 Ghz (C/C++)
18	GSSNet			4.28 %	5.48 %	1.1 px	1.2 px	0.78 s	1 core @ 2.5 Ghz (Python)
19	ICGNet-abl			4.29 %	5.19 %	1.0 px	1.0 px	0.18s	1 core @ 2.5 Ghz (C/C++)
20	Selective-RAFT			4.35 %	4.68 %	1.1 px	1.2 px	0.45 s	1 core @ 2.5 Ghz (Python)
21	IGEV-Stereo			4.35 %	5.00 %	1.0 px	1.1 px	0.18 s	NVIDIA RTX 3090 (PyTorch)
G. Xu, X. Wang, X. Ding and X. Yang: Iterative Geometry Encoding Volume for Stereo Matching. CVPR 2023.
22	PCMAnet			4.57 %	6.29 %	1.0 px	1.1 px	0.27 s	1 core @ 2.5 Ghz (C/C++)
23	DualNet-kd*			4.60 %	5.73 %	1.0 px	1.1 px	0.3 s	1 core @ 2.5 Ghz (C/C++)
24	RCGSNP			4.68 %	5.84 %	1.2 px	1.3 px	0.12 s	GPU @ 2.5 Ghz (Python)
25	EGA-Stereo		code	4.80 %	6.22 %	1.0 px	1.1 px	0.41 s	1 core @ 2.5 Ghz (Python)
26	GANet+ADL			4.84 %	6.10 %	1.2 px	1.4 px	0.67 s	1 core @ 2.5 Ghz (Python)
27	Any-RAFT			4.85 %	5.71 %	1.2 px	1.2 px	0.32 s	1 core @ 2.5 Ghz (C/C++)
28	PCWNet-SCE			4.97 %	6.28 %	1.1 px	1.3 px	0.44 s	1 core @ 2.5 Ghz (C/C++)
29	OnestageStereo			4.99 %	6.50 %	1.1 px	1.2 px	1 s	GPU @ 2.5 Ghz (C/C++)
30	HD^3-Stereo		code	4.99 %	6.77 %	1.0 px	1.1 px	0.14 s	NVIDIA Pascal Titan XP
Z. Yin, T. Darrell and F. Yu: Hierarchical Discrete Distribution Decomposition for Match Density Estimation. CVPR 2019.
31	PCWNet		code	4.99 %	6.20 %	1.0 px	1.2 px	0.44 s	1 core @ 2.5 Ghz (C/C++)
Z. Shen, Y. Dai, X. Song, Z. Rao, D. Zhou and L. Zhang: PCW-Net: Pyramid Combination and Warping Cost Volume for Stereo Matching. European Conference on Computer Vision(ECCV) 2022.
32	SSMF			5.00 %	6.19 %	1.2 px	1.4 px	0.20 s	1 core @ 2.5 Ghz (Python)
33	CGF-ACV		code	5.02 %	6.32 %	1.0 px	1.1 px	0.24 s	NVIDIA RTX 3090 (PyTorch)
34	MVACVNet			5.02 %	6.32 %	1.0 px	1.1 px	0.01 s	GPU @ 2.5 Ghz (Python)
35	DiffuVolume			5.03 %	6.23 %	1.0 px	1.2 px	0.36 s	GPU @ 2.5 Ghz (Python)
36	DN+ACVNet			5.05 %	6.53 %	1.2 px	1.4 px	0.24 s	1 core @ 2.5 Ghz (C/C++)
37	HSJ_STEREO			5.07 %	6.33 %	1.0 px	1.1 px	0.1 s	1 core @ 2.5 Ghz (C/C++)
38	WiCRI_STEREO			5.07 %	6.33 %	1.0 px	1.1 px	1 s	1 core @ 2.5 Ghz (C/C++)
39	PCMAnet		code	5.10 %	6.98 %	1.1 px	1.2 px	0.27 s	1 core @ 2.5 Ghz (C/C++)
40	BSDual-CNN			5.14 %	6.19 %	1.0 px	1.1 px	0.45 s	GPU @ 2.5 Ghz (Python)
41	UGNet			5.16 %	6.78 %	1.1 px	1.3 px	0.3 s	GPU @ 2.5 Ghz (Python)
42	NeXt-Stereo			5.22 %	6.16 %	1.1 px	1.2 px	0.06 s	GPU @ 2.0 Ghz (Python)
43	FGDS-Net			5.29 %	6.81 %	1.1 px	1.2 px	0.3 s	1 core @ 2.5 Ghz (Python)
44	GEMAStereo			5.34 %	6.83 %	1.1 px	1.2 px	0.03 s	GPU @ 2.5 Ghz (Python)
45	LEAStereo		code	5.35 %	6.50 %	1.1 px	1.2 px	0.3 s	GPU @ 2.5 Ghz (Python)
X. Cheng, Y. Zhong, M. Harandi, Y. Dai, X. Chang, H. Li, T. Drummond and Z. Ge: Hierarchical Neural Architecture Search for Deep Stereo Matching. Advances in Neural Information Processing Systems 2020.
46	RAFT-Stereo		code	5.40 %	6.48 %	1.3 px	1.3 px	0.38 s	1 core @ 2.5 Ghz (Python)
47	ICVP		code	5.43 %	6.54 %	1.2 px	1.2 px	0.17 s	GPU @ 1.5 Ghz (Python)
O. Kwon and E. Zell: Image-Coupled Volume Propagation for Stereo Matching. 2023 IEEE International Conference on Image Processing (ICIP) 2023.
48	ADStereo			5.54 %	7.03 %	1.1 px	1.2 px	0.05 s	GPU @ 2.5 Ghz (Python)
49	MAF-Stereo		code	5.60 %	7.17 %	1.2 px	1.4 px	0.07 s	GPU @ 2.5 Ghz (Python)
50	MDA			5.64 %	7.22 %	1.2 px	1.4 px	0.32s	1 core @ 2.5 Ghz (Python)
51	DVANet			5.68 %	7.48 %	1.2 px	1.3 px	0.1 s	NVIDIA 3090 (PyTorch)
52	AutoDispNet-CSS		code	5.69 %	6.68 %	1.0 px	1.1 px	0.9 s	1 core @ 2.5 Ghz (C/C++)
T. Saikia, Y. Marrakchi, A. Zela, F. Hutter and T. Brox: AutoDispNet: Improving Disparity Estimation with AutoML. The IEEE International Conference on Computer Vision (ICCV) 2019.
53	PSMNet+CBAM			5.81 %	7.56 %	1.2 px	1.3 px	0.36 s	NVIDIA RTX 3090 (Python)
54	URDAD			5.81 %	7.27 %	1.5 px	1.6 px	0.35 s	1 core @ 2.5 Ghz (C/C++)
55	UCFNet		code	5.83 %	7.12 %	1.1 px	1.2 px	0.21 s	1 core @ 2.5 Ghz (C/C++)
Z. Shen, X. Song, Y. Dai, D. Zhou, Z. Rao and L. Zhang: Digging Into Uncertainty-Based Pseudo- Label for Robust Stereo Matching. IEEE Transactions on Pattern Analysis and Machine Intelligence 2023.
56	SCVFormer			5.83 %	7.42 %	1.4 px	1.6 px	0.09 s	NVIDIA RTX 3090 (PyTorch)
57	PFSMNet		code	5.83 %	7.39 %	1.2 px	1.3 px	0.31 s	1 core @ 2.5 Ghz (C/C++)
K. Zeng, Y. Wang, Q. Zhu, J. Mao and H. Zhang: Deep Progressive Fusion Stereo Network. IEEE Transactions on Intelligent Transportation Systems 2021.
58	EdgeStereo-V2			5.84 %	7.51 %	1.0 px	1.2 px	0.32 s	Nvidia GTX Titan Xp
X. Song, X. Zhao, L. Fang, H. Hu and Y. Yu: Edgestereo: An effective multi-task learning network for stereo matching and edge detection. International Journal of Computer Vision (IJCV) 2019.
59	HITNet		code	5.91 %	7.54 %	1.0 px	1.2 px	0.02 s	GPU @ 2.5 Ghz (Python + C/C++)
V. Tankovich, C. Häne, Y. Zhang, A. Kowdle, S. Fanello and S. Bouaziz: HITNet: Hierarchical Iterative Tile Refinement Network for Real-time Stereo Matching. CVPR 2021.
60	ADPNet			5.93 %	7.74 %	1.3 px	1.5 px	0.06 s	1 core @ 2.5 Ghz (C/C++)
61	CFNet		code	5.96 %	7.29 %	1.2 px	1.3 px	0.18 s	1 core @ 2.5 Ghz (Python)
Z. Shen, Y. Dai and Z. Rao: CFNet: Cascade and Fused Cost Volume for Robust Stereo Matching. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2021. Z. Shen, X. Song, Y. Dai, D. Zhou, Z. Rao and L. Zhang: Digging Into Uncertainty-Based Pseudo- Label for Robust Stereo Matching. IEEE Transactions on Pattern Analysis and Machine Intelligence 2023.
62	HCRNet			6.01 %	7.68 %	1.2 px	1.3 px	0.19 s	GPU @ 2.5 Ghz (Python)
63	LaC+GANet		code	6.02 %	7.34 %	1.3 px	1.5 px	1.8 s	1 core @ 2.5 Ghz (C/C++)
B. Liu, H. Yu and Y. Long: Local Similarity Pattern and Cost Self- Reassembling for Deep Stereo Matching Networks. Proceedings of the AAAI Conference on Artificial Intelligence 2022.
64	taugr1215			6.05 %	7.76 %	1.2 px	1.3 px	0.35 s	1 core @ 2.5 Ghz (C/C++)
65	DPCTF-S			6.12 %	7.81 %	1.2 px	1.4 px	0.11 s	GPU @ 2.5 Ghz (Python)
Y. Deng, J. Xiao, S. Zhou and J. Feng: Detail Preserving Coarse-to-Fine Matching for Stereo Matching and Optical Flow. IEEE Transactions on Image Processing 2021.
66	ProNet			6.15 %	7.72 %	1.2 px	1.4 px	0.33 s	GPU @ 2.5 Ghz (Python)
67	NLCA-Net v2		code	6.17 %	7.65 %	1.2 px	1.3 px	0.67 s	GPU @ >3.5 Ghz (Python)
Z. Rao, D. Yuchao, S. Zhelun and H. Renjie: Rethinking Training Strategy in Stereo Matching. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS .
68	TBFE-Net			6.19 %	7.80 %	1.3 px	1.4 px	0.3 s	1 core @ 2.5 Ghz (Python)
69	GANet-deep		code	6.22 %	7.92 %	1.2 px	1.3 px	1.8 s	GPU @ 2.5 Ghz (Python)
F. Zhang, V. Prisacariu, R. Yang and P. Torr: GA-Net: Guided Aggregation Net for End-to-end Stereo Matching. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2019.
70	taugr12			6.25 %	8.02 %	1.2 px	1.3 px	0.35 s	1 core @ 2.5 Ghz (C/C++)
71	LaC+GwcNet		code	6.26 %	8.02 %	1.5 px	1.7 px	0.65 s	GPU @ 2.5 Ghz (Python)
B. Liu, H. Yu and Y. Long: Local Similarity Pattern and Cost Self- Reassembling for Deep Stereo Matching Networks. Proceedings of the AAAI Conference on Artificial Intelligence 2022.
72	GINet			6.26 %	7.91 %	1.3 px	1.4 px	0.25 s	2 cores @ 2.5 Ghz (Python)
73	CREStereo		code	6.27 %	7.27 %	1.4 px	1.4 px	0.40 s	GPU @ >3.5 Ghz (C/C++)
J. Li, P. Wang, P. Xiong, T. Cai, Z. Yan, L. Yang, J. Liu, H. Fan and S. Liu: Practical Stereo Matching via Cascaded Recurrent Network with Adaptive Correlation. 2022.
74	VCNet			6.31 %	7.67 %	1.4 px	1.5 px	0.6 s	1 core @ 2.5 Ghz (Python)
ERROR: Wrong syntax in BIBTEX file.
75	pcanet		code	6.33 %	8.50 %	1.4 px	1.6 px	0.27 s	1 core @ 2.5 Ghz (C/C++)
76	NMRF-Stereo			6.35 %	8.11 %	1.4 px	1.6 px	0.09 s	NVIDIA RTX 3090 (PyTorch)
77	SegStereo		code	6.35 %	8.06 %	1.1 px	1.3 px	0.6 s	Nvidia GTX Titan Xp
G. Yang, H. Zhao, J. Shi, Z. Deng and J. Jia: SegStereo: Exploiting Semantic Information for Disparity Estimation. ECCV 2018.
78	FusionStereo			6.39 %	8.10 %	1.2 px	1.3 px	16 s	1 core @ 2.5 Ghz (Python)
79	CAL-Net			6.40 %	8.17 %	1.5 px	1.7 px	0.44 s	4 cores @ 2.5 Ghz (Python)
S. Chen, B. Li, W. Wang, H. Zhang, H. Li and Z. Wang: Cost Affinity Learning Network for Stereo Matching. IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2021, Toronto, ON, Canada, June 6-11, 2021 2021.
80	BGNet+			6.44 %	8.41 %	1.2 px	1.4 px	0.02 s	GPU @ 2.5 Ghz (Python)
B. Xu, Y. Xu, X. Yang, W. Jia and Y. Guo: Bilateral Grid Learning for Stereo Matching Network. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2021.
81	OptStereo			6.46 %	8.12 %	1.3 px	1.4 px	0.10 s	GPU @ 2.5 Ghz (Python)
H. Wang, R. Fan, P. Cai and M. Liu: PVStereo: Pyramid voting module for end-to-end self-supervised stereo matching. IEEE Robotics and Automation Letters 2021.
82	PDSNet			6.50 %	8.70 %	1.4 px	1.6 px	0.5 s	1 core @ 2.5 Ghz (Python + C/C++)
S. Tulyakov, A. Ivanov and F. Fleuret: Practical Deep Stereo (PDS): Toward applications-friendly deep stereo matching. Proceedings of the international conference on Neural Information Processing Systems (NIPS) 2018.
83	AMNet			6.55 %	8.16 %	1.2 px	1.3 px	0.9 s	GPU @ 2.5 Ghz (Python)
X. Du, M. El-Khamy and J. Lee: AMNet: Deep Atrous Multiscale Stereo Disparity Estimation Networks. 2019.
84	DANet-Stereo			6.59 %	8.18 %	1.3 px	1.5 px	2.7 s	GPU @ 2.5 Ghz (Python)
85	SCV-Stereo		code	6.59 %	8.34 %	1.2 px	1.4 px	0.08 s	GPU @ 2.5 Ghz (Python)
H. Wang, R. Fan and M. Liu: SCV-Stereo: Learning stereo matching from a sparse cost volume. 2021 IEEE International Conference on Image Processing (ICIP) 2021.
86	DMCNet			6.62 %	8.45 %	1.3 px	1.4 px	0.27 s	GPU @ 2.5 Ghz (Python)
87	ERSCNet			6.63 %	9.04 %	1.2 px	1.4 px	0.28 s	GPU @ 2.5 Ghz (Python)
Anonymous: ERSCNet. Proceedings of the European Conference on Computer Vision (ECCV) 2020.
88	PGNet			6.67 %	8.54 %	1.3 px	1.5 px	0.7 s	1 core @ 2.5 Ghz (python)
S. Chen, Z. Xiang, C. Qiao, Y. Chen and T. Bai: PGNet: Panoptic parsing guided deep stereo matching. Neurocomputing 2021.
89	Abc-Net			6.80 %	8.40 %	1.4 px	1.5 px	0.72 s	4 cores @ 2.5 Ghz (Python)
X. Li, Y. Fan, G. Lv and H. Ma: Area-based correlation and non-local attention network for stereo matching. The Visual Computer 2021.
90	CoEx		code	6.83 %	8.63 %	1.3 px	1.4 px	0.027 s	RTX 2080Ti (Python)
A. Bangunharcana, J. Cho, S. Lee, I. Kweon, K. Kim and S. Kim: Correlate-and-Excite: Real-Time Stereo Matching via Guided Cost Volume Excitation. 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2021.
91	NLCA-Net-3		code	6.85 %	8.19 %	1.2 px	1.4 px	0.44 s	GPU @ 2.5 Ghz (Python)
Z. Rao, M. He, Y. Dai, Z. Zhu, B. Li and R. He: NLCA-Net: a non-local context attention network for stereo matching. APSIPA Transactions on Signal and Information Processing 2020.
92	UAIStereo			6.90 %	8.63 %	1.3 px	1.4 px	0.06 s	GPU @ 3.5 Ghz (Python)
93	GCGANet-V1			6.92 %	8.31 %	1.3 px	1.5 px	0.15 s	1 core @ 2.5 Ghz (C/C++)
94	AcfNet		code	6.93 %	8.52 %	1.8 px	1.9 px	0.48 s	1 core @ 2.5 Ghz (Python)
Y. Zhang, Y. Chen, X. Bai, S. Yu, K. Yu, Z. Li and K. Yang: Adaptive Unimodal Cost Volume Filtering for Deep Stereo Matching. AAAI 2020.
95	SASNet			6.98 %	8.83 %	1.3 px	1.5 px	0.21 s	GPU @ >3.5 Ghz (Python)
96	IEG-Net			7.01 %	8.85 %	1.2 px	1.4 px	0.40 s	1 core @ 2.5 Ghz (Python)
97	AAG			7.01 %	8.85 %	1.2 px	1.4 px	1.2 s	1 core @ 2.5 Ghz (C/C++)
98	SGNet			7.02 %	8.89 %	1.4 px	1.5 px	0.6 s	1 core @ 2.5 Ghz (Python + C/C++)
S. Chen, Z. Xiang, C. Qiao, Y. Chen and T. Bai: SGNet: Semantics Guided Deep Stereo Matching. Proceedings of the Asian Conference on Computer Vision (ACCV) 2020.
99	ACVNet		code	7.03 %	8.67 %	1.4 px	1.5 px	0.2 s	NVIDIA RTX 3090 (PyTorch)
G. Xu, J. Cheng, P. Guo and X. Yang: Attention Concatenation Volume for Accurate and Efficient Stereo Matching. CVPR 2022.
100	ICGNet-gwc			7.07 %	8.85 %	1.3 px	1.4 px	0.32 s	GPU @ 2.5 Ghz (Python)
101	NLCA-Net		code	7.20 %	9.00 %	1.3 px	1.4 px	0.6 s	GPU @ 2.5 Ghz (C/C++)
Z. Rao, M. He, Y. Dai, Z. Zhu, B. Li and R. He: NLCA-Net: a non-local context attention network for stereo matching. APSIPA Transactions on Signal and Information Processing 2020.
102	AANet+		code	7.22 %	9.10 %	1.3 px	1.4 px	0.06 s	NVIDIA V100 GPU
H. Xu and J. Zhang: AANet: Adaptive Aggregation Network for Efficient Stereo Matching. CVPR 2020.
103	GDANet			7.29 %	9.15 %	1.5 px	1.7 px	0.04 s	1 core @ 2.5 Ghz (Python)
104	ASNet			7.32 %	9.15 %	1.5 px	1.7 px	0.17 s	GPU @ >3.5 Ghz (Python)
105	GASN			7.33 %	9.35 %	1.3 px	1.5 px	0.09 s	NVIDIA RTX 3090 (PyTorch)
106	HSM		code	7.38 %	9.40 %	1.3 px	1.6 px	0.15 s	Titan X Pascal
G. Yang, J. Manela, M. Happold and D. Ramanan: Hierarchical Deep Stereo Matching on High- Resolution Images. The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2019.
107	W-Stereo-a-r			7.39 %	9.24 %	1.4 px	1.5 px	0.07 s	1 core @ 2.5 Ghz (Python)
108	iResNet-i2		code	7.40 %	9.09 %	1.2 px	1.3 px	0.12 s	1 core @ 2.5 Ghz (C/C++)
Z. Liang, Y. Feng, Y. Guo, H. Liu, W. Chen, L. Qiao, L. Zhou and J. Zhang: Learning for disparity estimation through feature constancy. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition 2018.
109	yjlig			7.48 %	9.37 %	1.3 px	1.5 px	0.35 s	1 core @ 2.5 Ghz (C/C++)
110	EAC-Stereo		code	7.48 %	9.37 %	1.3 px	1.5 px	0.38s	1 core @ 2.5 Ghz (Python)
111	AFNet			7.50 %	9.04 %	1.3 px	1.4 px	0.25 s	1 core @ 2.5 Ghz (Python)
112	CFP-Net		code	7.52 %	9.58 %	1.4 px	1.6 px	0.95 s	8 cores @ 2.5 Ghz (Python)
Z. Zhu, M. He, Y. Dai, Z. Rao and B. Li: Multi-scale Cross-form Pyramid Network for Stereo Matching. arXiv preprint 2019.
113	SSPCVNet			7.56 %	9.43 %	1.5 px	1.7 px	0.9 s	1 core @ 2.5 Ghz (Python)
Z. Wu, X. Wu, X. Zhang, S. Wang and L. Ju: Semantic Stereo Matching With Pyramid Cost Volumes. The IEEE International Conference on Computer Vision (ICCV) 2019.
114	MABNet_origin		code	7.57 %	9.07 %	1.3 px	1.4 px	0.38 s	Nvidia rtx2080ti (Python)
J. Xing, Z. Qi, J. Dong, J. Cai and H. Liu: MABNet: A Lightweight Stereo Network Based on Multibranch Adjustable Bottleneck Module. .
115	PASNet			7.75 %	9.22 %	1.4 px	1.5 px	0.38 s	GPU @ 3.5 Ghz (Python)
116	GwcNet-gc		code	7.80 %	9.28 %	1.3 px	1.4 px	0.32 s	GPU @ 2.0 Ghz (Java + C/C++)
X. Guo, K. Yang, W. Yang, X. Wang and H. Li: Group-wise correlation stereo network. CVPR 2019.
117	FADNet		code	7.85 %	9.17 %	1.2 px	1.3 px	0.05 s	Tesla V100 (Python)
Q. Wang, S. Shi, S. Zheng, K. Zhao and X. Chu: FADNet: A Fast and Accurate Network for Disparity Estimation. arXiv preprint arXiv:2003.10758 2020.
118	GANet-15		code	7.87 %	9.85 %	1.3 px	1.5 px	0.36 s	1 core @ 2.5 Ghz (C/C++)
F. Zhang, V. Prisacariu, R. Yang and P. Torr: GA-Net: Guided Aggregation Net for End-to-end Stereo Matching. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2019.
119	PSMNet		code	8.36 %	10.18 %	1.4 px	1.6 px	0.41 s	Nvidia Titan Xp
J. Chang and Y. Chen: Pyramid Stereo Matching Network. arXiv preprint arXiv:1803.08669 2018.
120	Displets		code	8.40 %	9.89 %	1.9 px	2.3 px	265 s	>8 cores @ 3.0 Ghz (Matlab + C/C++)
F. Guney and A. Geiger: Displets: Resolving Stereo Ambiguities using Object Knowledge. Conference on Computer Vision and Pattern Recognition (CVPR) 2015.
121	WSMCnet		code	8.50 %	10.27 %	2.1 px	2.3 px	0.39 s	GPU @ Nvidia GTX 1070 (Pytorch)
Y. Wang, H. Wang, G. Yu, M. Yang, Y. Yuan and J. Quan: Stereo Matching Algorithm Based on Three-Dimensional Convolutional Neural Network. Acta Optica Sinica 2019.
122	PVStereo			8.54 %	9.96 %	2.2 px	2.3 px	0.10 s	1 core @ 2.5 Ghz (Python)
H. Wang, R. Fan, P. Cai and M. Liu: PVStereo: Pyramid voting module for end-to-end self-supervised stereo matching. IEEE Robotics and Automation Letters 2021.
123	Fast DS-CS		code	8.88 %	10.40 %	1.3 px	1.5 px	0.02 s	GPU @ 2.0 Ghz (Python + C/C++)
K. Yee and A. Chakrabarti: Fast Deep Stereo with 2D Convolutional Processing of Cost Signatures. WACV 2020 (to appear).
124	GAANet			8.98 %	11.01 %	1.8 px	2.0 px	0.08	2080tiGPU @ 2.5 Ghz (Python)
125	Displets v2		code	8.99 %	10.41 %	2.0 px	2.2 px	265 s	>8 cores @ 3.0 Ghz (Matlab + C/C++)
F. Guney and A. Geiger: Displets: Resolving Stereo Ambiguities using Object Knowledge. Conference on Computer Vision and Pattern Recognition (CVPR) 2015.
126	MSDC-Net			9.16 %	11.27 %	1.8 px	2.0 px	0.6 s	1 core @ 2.5 Ghz (C/C++)
Z. Rao, M. He, Y. Dai, Z. Zhu, B. Li and R. He: MSDC-Net: Multi-Scale Dense and Contextual Networks for Stereo Matching. 2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) 2019.
127	MMStereo			9.27 %	11.17 %	1.6 px	1.7 px	0.04 s	Nvidia Titan RTX (Python)
K. Shankar, M. Tjersland, J. Ma, K. Stone and M. Bajracharya: A Learned Stereo Depth System for Robotic Manipulation in Homes. .
128	DualNet-kd			9.40 %	11.25 %	1.7 px	1.9 px	0.3 s	1 core @ 2.5 Ghz (C/C++)
129	RTSnet		code	9.54 %	11.49 %	1.7 px	1.9 px	0.02 s	1 core @ 2.5 Ghz (Python)
H. Lee and Y. Shin: Real-Time Stereo Matching Network with High Accuracy. 2019 IEEE International Conference on Image Processing (ICIP) 2019.
130	WaveletStereo			9.84 %	12.17 %	1.7 px	1.9 px	0.27 s	1 core @ 2.5 Ghz (C/C++)
Anonymous: WaveletStereo: Learning wavelet coefficients for stereo matching. arXiv: Computer Vision and Pattern Recognition 2019.
131	AANet		code	10.51 %	11.97 %	1.7 px	1.8 px	0.06 s	GPU @ 2.5 Ghz (Python)
H. Xu and J. Zhang: AANet: Adaptive Aggregation Network for Efficient Stereo Matching. CVPR 2020.
132	GC-NET			10.80 %	12.80 %	1.8 px	2.0 px	0.9 s	Nvidia GTX Titan X
A. Kendall, H. Martirosyan, S. Dasgupta, P. Henry, R. Kennedy, A. Bachrach and A. Bry: End-to-End Learning of Geometry and Context for Deep Stereo Regression. Proceedings of the International Conference on Computer Vision (ICCV) 2017.
133	MDTE4			11.13 %	13.31 %	2.2 px	2.4 px	0.03 s	1 core @ 2.5 Ghz (C/C++)
134	VC-SF			11.58 %	12.29 %	2.7 px	2.8 px	300 s	1 core @ 2.5 Ghz (C/C++)
C. Vogel, S. Roth and K. Schindler: View-Consistent 3D Scene Flow Estimation over Multiple Frames. Proceedings of European Conference on Computer Vision. Lecture Notes in, Computer Science 2014.
135	PRSM		code	11.91 %	12.87 %	2.0 px	2.1 px	300 s	1 core @ 2.5 Ghz (C/C++)
C. Vogel, K. Schindler and S. Roth: 3D Scene Flow Estimation with a Piecewise Rigid Scene Model. ijcv 2015.
136	TSNnet_Teacher			12.06 %	14.38 %	2.5 px	2.9 px	0.01 s	1 core @ 2.5 Ghz (C/C++)
137	OSF		code	12.18 %	15.35 %	2.1 px	2.6 px	50 min	1 core @ 3.0 Ghz (Matlab + C/C++)
M. Menze and A. Geiger: Object Scene Flow for Autonomous Vehicles. Conference on Computer Vision and Pattern Recognition (CVPR) 2015.
138	TSNnet_naive			13.95 %	16.67 %	2.3 px	2.8 px	0.01 s	1 core @ 2.5 Ghz (C/C++)
139	SsSMnet			14.02 %	16.59 %	3.1 px	3.6 px	0.8 s	Titan Xp
Y. Zhong, Y. Dai and H. Li: Self-Supervised Learning for Stereo Matching with Self-Improving Ability. arXiv:1709.00930 2017.
140	JSOSM			14.16 %	16.89 %	2.4 px	2.8 px	105 s	8 cores @ 2.5 Ghz (C/C++)
X. Li and J. Liu: EFFICIENT STEREO MATCHING USING SEGMENT OPTIMIZATION. ICIP 2016.
141	RecResNet		code	14.25 %	17.32 %	2.1 px	2.5 px	0.3 s	GPU @ NVIDIA TITAN X (Tensorflow)
K. Batsos and P. Mordohai: RecResNet: A Recurrent Residual CNN Architecture for Disparity Map Enhancement. In International Conference on 3D Vision (3DV) 2018.
142	PCBP-SS			14.26 %	18.33 %	2.4 px	3.9 px	5 min	4 cores @ 2.5 Ghz (Matlab + C/C++)
K. Yamaguchi, D. McAllester and R. Urtasun: Robust Monocular Epipolar Flow Estimation. CVPR 2013.
143	TSNnet_student			14.33 %	17.03 %	3.1 px	3.8 px	0.01 s	1 core @ 2.5 Ghz (C/C++)
144	FD-Fusion		code	14.64 %	17.15 %	2.6 px	2.9 px	0.01 s	1 core @ 2.5 Ghz (C/C++)
M. Ferrera, A. Boulch and J. Moras: Fast Stereo Disparity Maps Refinement By Fusion of Data-Based And Model-Based Estimations. International Conference on 3D Vision (3DV) 2019.
145	SPS-StFl			14.74 %	18.00 %	2.9 px	3.6 px	35 s	1 core @ 3.5 Ghz (C/C++)
K. Yamaguchi, D. McAllester and R. Urtasun: Efficient Joint Segmentation, Occlusion Labeling, Stereo and Flow Estimation. ECCV 2014.
146	CoR		code	15.30 %	19.15 %	2.7 px	4.1 px	6 s	6 cores @ 3.3 Ghz (Matlab + C/C++)
A. Chakrabarti, Y. Xiong, S. Gortler and T. Zickler: Low-level Vision by Consensus in a Spatial Hierarchy of Regions. CVPR 2015.
147	cfusion		code	15.31 %	16.20 %	2.6 px	2.8 px	70 s	GPU (Matlab + CUDA)
V. Ntouskos and F. Pirri: Confidence driven TGV fusion. arXiv preprint arXiv:1603.09302 2016.
148	SGM-Net			15.31 %	18.97 %	3.0 px	3.8 px	67 s	Titan X
A. Seki and M. Pollefeys: SGM-Nets: Semi-Global Matching With Neural Networks. CVPR 2017.
149	P3SNet+		code	15.85 %	18.50 %	2.1 px	2.4 px	0.01 s	GPU @ 2.5 Ghz (Python)
A. Emlek and M. Peker: P3SNet: Parallel Pyramid Pooling Stereo Network. IEEE Transactions on Intelligent Transportation Systems 2023.
150	L-ResMatch		code	15.94 %	19.71 %	3.7 px	5.2 px	48 s	Titan X (Torch7, CUDA)
A. Shaked and L. Wolf: Improved Stereo Matching with Constant Highway Networks and Reflective Loss. arXiv preprint arxiv:1701.00165 2016.
151	DispNetC		code	16.04 %	18.15 %	2.1 px	2.3 px	0.06 s	Nvidia GTX Titan X (Caffe)
N. Mayer, E. Ilg, P. Häusser, P. Fischer, D. Cremers, A. Dosovitskiy and T. Brox: A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimation. CVPR 2016.
152	SPS-St		code	16.05 %	19.34 %	3.1 px	3.6 px	2 s	1 core @ 3.5 Ghz (C/C++)
K. Yamaguchi, D. McAllester and R. Urtasun: Efficient Joint Segmentation, Occlusion Labeling, Stereo and Flow Estimation. ECCV 2014.
153	CNNF+SGM			16.22 %	19.87 %	2.9 px	3.9 px	71 s	TESLA K40C
F. Zhang and B. Wah: Fundamental Principles on Learning New Features for Effective Dense Matching. IEEE Transactions on Image Processing 2018.
154	DDS-SS			16.23 %	19.39 %	2.5 px	3.0 px	1 min	1 core @ 2.5 Ghz (Matlab + C/C++)
D. Wei, C. Liu and W. Freeman: A Data-driven Regularization Model for Stereo and Flow. 3DTV-Conference, 2014 International Conference on 2014.
155	PCBP			16.28 %	20.22 %	2.8 px	4.4 px	5 min	4 cores @ 2.5 Ghz (Matlab + C/C++)
K. Yamaguchi, T. Hazan, D. McAllester and R. Urtasun: Continuous Markov Random Fields for Robust Stereo Estimation. ECCV 2012.
156	MABNet_tiny		code	16.50 %	18.83 %	2.5 px	2.8 px	0.11 s	1 core @ 2.5 Ghz (Python)
J. Xing, Z. Qi, J. Dong, J. Cai and H. Liu: MABNet: A Lightweight Stereo Network Based on Multibranch Adjustable Bottleneck Module. .
157	PBCP			16.78 %	20.29 %	3.1 px	4.2 px	68 s	Nvidia GTX Titan X
A. Seki and M. Pollefeys: Patch Based Confidence Prediction for Dense Disparity Map. British Machine Vision Conference (BMVC) 2016.
158	MC-CNN-acrt		code	17.09 %	20.70 %	3.2 px	4.1 px	67 s	Nvidia GTX Titan X (CUDA, Lua/Torch7)
J. Zbontar and Y. LeCun: Stereo Matching by Training a Convolutional Neural Network to Compare Image Patches. Submitted to JMLR .
159	CBMV		code	17.10 %	20.70 %	3.7 px	4.4 px	250 s	6 cores@3.0Ghz(Python,C/C++,CUDA TitanX)
K. Batsos, C. Cai and P. Mordohai: CBMV: A Coalesced Bidirectional Matching Volume for Disparity Estimation. 2018.
160	Deep Embed			17.14 %	20.78 %	2.8 px	3.8 px	3 s	1 core @ 2.5 Ghz (C/C++)
Z. Chen, X. Sun, Y. Yu, L. Wang and C. Huang: A Deep Visual Correspondence Embedding Model for Stereo Matching Costs. ICCV 2015.
161	SMCM			17.46 %	21.74 %	2.8 px	4.8 px	1800 s	Nvidia GTX 1080 (Caffe)
M. Yang, Y. Liu, Y. Cai and Z. You: Stereo matching based on classification of materials. Neurocomputing 2016.
162	P3SNet		code	17.48 %	20.35 %	2.3 px	2.8 px	0.01 s	GPU @ 2.5 Ghz (Python)
A. Emlek and M. Peker: P3SNet: Parallel Pyramid Pooling Stereo Network. IEEE Transactions on Intelligent Transportation Systems 2023.
163	MC-CNN-WS		code	17.70 %	21.54 %	3.1 px	3.8 px	1.35 s	1 core 2.5 Ghz + K40 NVIDIA, Lua-Torch
S. Tulyakov, A. Ivanov and F. Fleuret: Weakly supervised learning of deep metrics for stereo reconstruction. ICCV 2017.
164	PR-Sf+E			17.85 %	20.82 %	3.3 px	4.0 px	200 s	4 cores @ 3.0 Ghz (Matlab + C/C++)
C. Vogel, K. Schindler and S. Roth: Piecewise Rigid Scene Flow. International Conference on Computer Vision (ICCV) 2013.
165	StereoSLIC			18.22 %	21.60 %	2.8 px	3.6 px	2.3 s	1 core @ 3.0 Ghz (C/C++)
K. Yamaguchi, D. McAllester and R. Urtasun: Robust Monocular Epipolar Flow Estimation. CVPR 2013.
166	MC-CNN			18.45 %	21.96 %	3.5 px	4.3 px	100 s	Nvidia GTX Titan (CUDA, Lua/Torch7)
J. Zbontar and Y. LeCun: Computing the Stereo Matching Cost with a Convolutional Neural Network. Conference on Computer Vision and Pattern Recognition (CVPR) 2015.
167	Content-CNN			18.81 %	22.38 %	3.6 px	4.2 px	0.7 s	Nvidia GTX Titan X (Torch)
W. Luo, A. Schwing and R. Urtasun: Efficient Deep Learning for Stereo Matching. CVPR 2016.
168	pSGM			18.87 %	22.87 %	3.3 px	5.7 px	7.92 s	4 cores @ 3.5 Ghz (C/C++)
Y. Lee, M. Park, Y. Hwang, Y. Shin and C. Kyung: Memory-Efficient Parametric Semiglobal Matching. IEEE Signal Processing Letters 2018.
169	PR-Sceneflow			19.22 %	22.07 %	3.3 px	4.0 px	150 sec	4 core @ 3.0 Ghz (Matlab - C/C++)
C. Vogel, K. Schindler and S. Roth: Piecewise Rigid Scene Flow. International Conference on Computer Vision (ICCV) 2013.
170	TCD-CRF			19.22 %	23.53 %	3.3 px	6.6 px	60 s	4 cores @ 3.5 Ghz (C/C++)
S. Arjomand Bigdeli, G. Budweiser and M. Zwicker: Temporally Coherent Disparity Maps Using CRFs with Fast 4D Filtering. Proc. ACPR 2015.
171	CoR-Conf		code	19.38 %	23.00 %	4.0 px	5.4 px	6 s	6 cores @ 3.3 Ghz (Matlab + C/C++)
A. Chakrabarti, Y. Xiong, S. Gortler and T. Zickler: Low-level Vision by Consensus in a Spatial Hierarchy of Regions. CVPR 2015.
172	DLP			19.67 %	23.72 %	3.2 px	5.9 px	60 s	8 cores @ >3.5 Ghz (C/C++)
V. Nguyen, H. Nguyen and J. Jeon: Robust Stereo Data Cost With a Learning Strategy. IEEE Transactions on Intelligent Transportation Systems 2017.
173	CRD-Fusion		code	19.84 %	22.68 %	2.9 px	3.5 px	0.02 s	GPU @ 2.5 Ghz (Python)
X. Fan, S. Jeon and B. Fidan: Occlusion-Aware Self-Supervised Stereo Matching with Confidence Guided Raw Disparity Fusion. Conference on Robots and Vision 2022.
174	MBM			20.37 %	23.63 %	3.6 px	4.3 px	0.2 s	1 core @ 3.0 Ghz (C/C++)
N. Einecke and J. Eggert: A Multi-Block-Matching Approach for Stereo. IV 2015.
175	SepStereo			20.91 %	23.24 %	3.2 px	3.6 px	0.087s	GPU @ 2.0 Ghz (Pyhton)
176	UHP			21.28 %	24.34 %	2.9 px	3.7 px	0.02 s	GPU @ 2.5 Ghz (Python)
177	AAFS+			21.32 %	23.51 %	3.0 px	3.2 px	0.01 s	1 core @ 2.5 Ghz (Python)
178	Flow2Stereo			21.35 %	23.51 %	2.8 px	3.1 px	0.05 s	GPU @ 2.5 Ghz (Python)
P. Liu, I. King, M. Lyu and J. Xu: Flow2Stereo: Effective Self-Supervised Learning of Optical Flow and Stereo Matching. CVPR 2020.
179	SGM-post			21.71 %	24.70 %	3.3 px	3.8 px	5 s	4 cores @ 2.5 Ghz (C/C++)
Z. Zhong: Efficient Learning based Semi-Global Stereo Matching. 2015 submitted.
180	CAT			21.72 %	24.88 %	3.3 px	4.3 px	10 s	1 core @ 3.5 Ghz (C/C++)
J. Ha, J. Jeon, G. Bae, S. Jo and H. Jeong: Cost Aggregation Table: Cost Aggregation Method Using Summed Area Table Scheme for Dense Stereo Correspondence. Advances in Visual Computing 2014.
181	CSPMS			22.41 %	26.24 %	4.5 px	6.0 px	6 s	4 cores @ 2.5 Ghz (C/C++)
J. Cho and M. Humenberger: Fast PatchMatch Stereo Matching Using Multi-Scale Cost Fusion for Automotive Applications. IV 2015.
182	ITGV			22.43 %	26.01 %	4.9 px	5.8 px	7 s	1 core @ 3.0 Ghz (Matlab + C/C++)
R. Ranftl, S. Gehrig, T. Pock and H. Bischof: Pushing the Limits of Stereo Using Variational Stereo Estimation. IV 2012.
183	RBM			22.51 %	25.39 %	3.6 px	4.0 px	0.2 s	1 core @ 3.0 Ghz (C/C++)
N. Einecke and J. Eggert: Block-Matching Stereo with Relaxed Fronto-Parallel Assumption. IV 2014.
184	ATGV			22.85 %	26.81 %	3.3 px	5.3 px	6 min	>8 cores @ 3.0 Ghz (Matlab + C/C++)
R. Ranftl, T. Pock and H. Bischof: Minimizing TGV-based Variational Models with Non-Convex Data terms. ICSSVM 2013.
185	AARBM			22.86 %	25.76 %	3.7 px	4.2 px	0.25 s	1 core @ 3.0 Ghz (C/C++)
N. Einecke and J. Eggert: Block-Matching Stereo with Relaxed Fronto-Parallel Assumption. IV 2014.
186	PASMNet_192		code	23.02 %	26.94 %	4.5 px	5.7 px	0.06 s	GPU @ 2.5 Ghz (Python)
187	SNCC			23.03 %	25.94 %	3.7 px	4.2 px	0.11 s	1 core @ 3.1 Ghz (C/C++)
N. Einecke and J. Eggert: A Two-Stage Correlation Method for Stereoscopic Depth Estimation. DICTA 2010.
188	Ensemble			23.04 %	26.70 %	4.5 px	6.9 px	135 s	2 cores @ >3.5 Ghz (Matlab)
A. Spyropoulos and P. Mordohai: Ensemble Classifier for Combining Stereo Matching Algorithms. International Conference on 3D Vision (3DV) 2015.
189	AABM			23.33 %	26.25 %	3.8 px	4.4 px	0.12 s	1 core @ 3.1 Ghz (C/C++)
N. Einecke and J. Eggert: Stereo Image Warping for Improved Depth Estimation of Road Surfaces. IV 2013.
190	ALTGV			23.47 %	26.85 %	3.4 px	4.2 px	20 s	GPU @ 2.5 Ghz (C/C++)
G. Kuschk and D. Cremers: Fast and Accurate Large-scale Stereo Reconstruction using Variational Methods. ICCV Workshop on Big Data in 3D Computer Vision 2013.
191	ARW		code	24.54 %	28.29 %	5.2 px	6.8 px	4.6s	1 core @ 3.5 Ghz (MATLAB+C/C++)
S. Lee, J. Lee, J. Lim and I. Suh: Robust Stereo Matching using Adaptive Random Walk with Restart Algorithm. Image and vision computing (accepted) 2015.
192	iSGM			24.67 %	28.86 %	4.7 px	8.6 px	8 s	2 cores @ 2.5 Ghz (C/C++)
S. Hermann and R. Klette: Iterative Semi-Global Matching for Robust Driver Assistance Systems. ACCV 2012.
193	AAFS			24.86 %	27.26 %	3.3 px	3.8 px	0.01 s	1 core @ 2.5 Ghz (C/C++)
J. Chang, P. Chang and Y. Chen: Attention-Aware Feature Aggregation for Real-time Stereo Matching on Edge Devices. Proceedings of the Asian Conference on Computer Vision 2020.
194	mSGM-LDE			24.97 %	29.33 %	5.5 px	9.3 px	55 s	2 cores @ 2.5 Ghz (C/C++)
V. Nguyen, D. Nguyen, S. Lee and J. Jeon: Local Density Encoding for Robust Stereo Matching. TCSVT 2014.
195	DispSegNet			25.00 %	27.94 %	3.7 px	4.2 px	0.9 s	GPU @ 2.5 Ghz (Python)
J. Zhang, K. Skinner, R. Vasudevan and M. Johnson-Roberson: DispSegNet: Leveraging Semantics for End- to-End Learning of Disparity Estimation From Stereo Imagery. IEEE Robotics and Automation Letters 2019.
196	wSGM			25.99 %	29.17 %	8.0 px	9.0 px	6s	1 core @ 3.5 Ghz (C/C++)
R. Spangenberg, T. Langner and R. Rojas: Weighted Semi-Global Matching and Center-Symmetric Census Transform for Robust Driver Assistance. CAIP 2013.
197	ELAS		code	26.75 %	30.41 %	5.4 px	6.3 px	0.3 s	1 core @ 2.5 Ghz (C/C++)
A. Geiger, M. Roser and R. Urtasun: Efficient Large-Scale Stereo Matching. ACCV 2010.
198	Permutation Stereo			26.86 %	29.87 %	5.0 px	5.8 px	30 s	GPU @ 2.5 Ghz (Matlab)
P. Brousseau and S. Roy: A Permutation Model for the Self- Supervised Stereo Matching Problem. 2022 19th Conference on Robots and Vision (CRV) 2022.
199	S+GF (Cen)		code	27.10 %	31.36 %	5.6 px	10.1 px	140 s	1 core @ 3.0 Ghz (C/C++)
K. Zhang, Y. Fang, D. Min, L. Sun, S. Yang, S. Yan and Q. Tian: Cross-Scale Cost Aggregation for Stereo Matching. CVPR 2014.
200	SGM			27.39 %	30.31 %	5.1 px	5.8 px	3.7 s	1 core @ 3.0 Ghz (C/C++)
H. Hirschmueller: Stereo Processing by Semi-Global Matching and Mutual Information. PAMI 2008.
201	rSGM		code	27.46 %	30.98 %	6.7 px	7.9 px	0.2 s	4 cores @ 2.6 Ghz (C/C++)
R. Spangenberg, T. Langner, S. Adfeldt and R. Rojas: Large Scale Semi-Global Matching on the CPU. IV 2014.
202	linBP			27.77 %	31.94 %	5.9 px	9.6 px	1.6 min	1 core @ 3.0 Ghz (C/C++)
W. Khan, V. Suaste, D. Caudillo and R. Klette: Belief Propagation Stereo Matching Compared to iSGM on Binocular or Trinocular Video Data. IV 2013.
203	SSMW			27.84 %	30.96 %	5.2 px	6.2 px	2.5 min	8 cores @ 2.5 Ghz (C/C++)
X. Li, J. Liu, G. Chen and H. Fu: Efficient Methods Using Slanted Support Windows for Slanted Surfaces. IET Computer Vision, http://ietdl.org/t/5QsTxb 2016.
204	Toast2			27.85 %	30.88 %	5.0 px	5.7 px	0.03 s	4 cores @ 3.5 Ghz (C/C++)
B. Ranft and T. Strau\ss: Modeling Arbitrarily Oriented Slanted Planes for Efficient Stereo Vision based on Block Matching. Intelligent Transportation Systems (ITSC), 2014 IEEE 17th International Conference on 2014.
205	OASM-Net			28.28 %	32.45 %	4.6 px	7.3 px	0.73 s	GPU @ 2.5 Ghz (Python)
A. Li and Z. Yuan: Occlusion Aware Stereo Matching via Cooperative Unsupervised Learning. Proceedings of the Asian Conference on Computer Vision, ACCV 2018.
206	HSMA			28.76 %	32.90 %	5.6 px	9.3 px	44s	1 core @ 3.0 Ghz (C/C++)
O. Zeglazi, M. Rziza, A. Amine and C. Demonceaux: A hierarchical stereo matching algorithm based on adaptive support region aggregation method. Pattern Recognition Letters 2018.
207	LAMC-DSΜ			29.18 %	32.73 %	5.1 px	8.0 px	10.8 min	2 cores @ 2.5 Ghz (Matlab)
C. Stentoumis, L. Grammatikopoulos, I. Kalisperakis, E. Petsa and G. Karras: A local adaptive approach for dense stereo matching in architectural scene reconstruction. ISPRS 2013.
208	GF (Census)		code	30.81 %	34.85 %	8.8 px	12.9 px	120 s	1 core @ 3.0 Ghz (C/C++)
A. Hosni, C. Rhemann, M. Bleyer, C. Rother and M. Gelautz: Fast Cost-Volume Filtering for Visual Correspondence and Beyond. TPAMI 2013. K. Zhang, Y. Fang, D. Min, L. Sun, S. Yang, S. Yan and Q. Tian: Cross-Scale Cost Aggregation for Stereo Matching. CVPR 2014.
209	Deep-Raw			31.55 %	35.52 %	9.8 px	13.2 px	1 s	1 core @ 2.5 Ghz (C/C++)
Z. Chen, X. Sun, Y. Yu, L. Wang and C. Huang: A Deep Visual Correspondence Embedding Model for Stereo Matching Costs. ICCV 2015.
210	BSM		code	32.50 %	35.96 %	7.5 px	10.3 px	2.5 min	1 core @ 3.0 Ghz (C/C++)
K. Zhang, J. Li, Y. Li, W. Hu, L. Sun and S. Yang: Binary stereo matching. Pattern Recognition (ICPR), 2012 21st International Conference on 2012.
211	ADSM			32.69 %	35.60 %	8.4 px	10.6 px	125 s	1 core @ 2.0 Ghz (C/C++)
O. Zeglazi, M. Rziza, A. Amine and C. Demonceaux: Accurate dense stereo matching for road scenes. 2017 IEEE International Conference on Image Processing, ICIP 2017, Beijing, China, September 17-20, 2017 .
212	IIW			33.04 %	36.39 %	8.5 px	11.7 px	5.5 s	1 core @ 2.5 Ghz (C/C++)
A. Murarka and N. Einecke: A meta-technique for increasing density of local stereo methods through iterative interpolation and warping. Canadian Conference on Computer and Robot Vision 2014.
213	GLDS		code	33.15 %	36.36 %	5.8 px	7.5 px	26 s	GPU @ 1.5 Ghz (C/C++)
K. Oguri and Y. Shibata: A new stereo formulation not using pixel and disparity models. 2018.
214	SymST-GP			33.68 %	37.24 %	12.1 px	15.2 px	0.254 s	Dual - Nvidia GTX Titan (CUDA)
R. Ralha, G. Falcao, J. Amaro, V. Mota, M. Antunes, J. Barreto and U. Nunes: Parallel refinement of slanted 3D reconstruction using dense stereo induced from symmetry. Journal of Real-Time Image Processing 2016.
215	SM_GPTM			33.78 %	36.81 %	11.7 px	12.9 px	6.5 s	2 cores @ 2.5 Ghz (C/C++)
C. Cigla and A. Alatan: An Improved Stereo Matching Algorithm with Ground Plane and Temporal Smoothness Constraints. ECCV Workshops 2012.
216	HLSC_mesh			35.58 %	38.92 %	9.0 px	11.3 px	800 s	1 core @ 2.5 Ghz (Matlab + C/C++)
S. Hadfield, K. Lebeda and R. Bowden: Stereo reconstruction using top-down cues. Computer Vision and Image Understanding 2016.
217	CrossCensus			35.97 %	38.80 %	8.0 px	10.1 px	30 s	1 core @ 2.5 Ghz (C/C++)
K. Zhang, J. Lu and G. Lafruit: Cross-Based Local Stereo Matching Using Orthogonal Integral Images. Circuits and Systems for Video Technology, IEEE Transactions on 2009.
218	OCV-BM-post		code	37.03 %	39.85 %	8.6 px	9.3 px	0.1 s	1 core @ 2.5 Ghz (C/C++)
G. Bradski: The OpenCV Library. Dr. Dobb's Journal of Software Tools 2000.
219	GCSF		code	37.08 %	39.53 %	5.9 px	6.7 px	2.4 s	1 core @ 2.5 Ghz (C/C++)
J. Cech, J. Sanchez-Riera and R. Horaud: Scene Flow Estimation by Growing Correspondence Seeds. CVPR 2011.
220	SDM		code	37.23 %	39.63 %	7.7 px	8.6 px	1 min	1 core @ 2.5 Ghz (C/C++)
J. Kostkova: Stratified dense matching for stereopsis in complex scenes. BMVC 2003.
221	OCV-SGBM		code	37.45 %	40.21 %	13.1 px	14.0 px	1.1 s	1 core @ 2.5 Ghz (C/C++)
H. Hirschmueller: Stereo processing by semiglobal matching and mutual information. PAMI 2008.
222	MSMW		code	37.60 %	39.98 %	9.3 px	10.0 px	3 min	4 cores @ 2.5 Ghz (C/C++)
A. Buades and G. Facciolo: On the performance of local methods for stereovision. 2013 submitted.
223	GCS		code	37.63 %	40.05 %	6.2 px	7.0 px	2.2 s	1 core @ 2.5 Ghz (C/C++)
J. Cech and R. Sara: Efficient Sampling of Disparity Space for Fast And Accurate Matching. BenCOS 2007.
224	CostFilter		code	39.41 %	41.95 %	11.8 px	13.5 px	4 min	1 core @ 2.5 Ghz (Matlab)
C. Rhemann, A. Hosni, M. Bleyer, C. Rother and M. Gelautz: Fast Cost-Volume Filtering for Visual Correspondence and Beyond. CVPR 2011.
225	VariableCros			42.39 %	45.26 %	13.8 px	16.1 px	30 s	1 core @ 2.5 Ghz (Matlab)
K. Zhang, J. Lu and G. Lafruit: Cross-Based Local Stereo Matching Using Orthogonal Integral Images. Circuits and Systems for Video Technology, IEEE Transactions on 2009.
226	GC+occ		code	46.25 %	48.96 %	14.9 px	17.6 px	6 min	1 core @ 2.5 Ghz (C/C++)
V. Kolmogorov and R. Zabih: Computing Visual Correspondence with Occlusions using Graph Cuts. ICCV 2001.
227	ALE-Stereo		code	83.80 %	84.37 %	24.6 px	25.4 px	50 min	1 core @ 3.0 Ghz (C/C++)
L. Ladicky, P. Sturgess, C. Russell, S. Sengupta, Y. Bastanlar, W. Clocksin and P. Torr: Joint Optimisation for Object Class Segmentation and Dense Stereo Reconstruction. BMVC 2010.
228	AVERAGE			88.11 %	88.80 %	26.2 px	28.2 px	0.01 s	1 core @ 2.5 Ghz (C/C++)
229	MEDIAN			90.79 %	91.33 %	29.6 px	31.5 px	0.01 s	1 core @ 2.5 Ghz (C/C++)

• HCI/Bosch Robust Vision Challenge: Optical flow and stereo vision challenge on high resolution imagery recorded at a high frame rate under diverse weather conditions (e.g., sunny, cloudy, rainy). The Robert Bosch AG provides a prize for the best performing method.
• Image Sequence Analysis Test Site (EISATS): Synthetic image sequences with ground truth information provided by UoA and Daimler AG. Some of the images come with 3D range sensor information.
• Middlebury Stereo Evaluation: The classic stereo evaluation benchmark, featuring four test images in version 2 of the benchmark, with very accurate ground truth from a structured light system. 38 image pairs are provided in total.
• Daimler Stereo Dataset: Stereo bad weather highway scenes with partial ground truth for freespace
• Make3D Range Image Data: Images with small-resolution ground truth used to learn and evaluate depth from single monocular images.
• Lubor Ladicky's Stereo Dataset: Stereo Images with manually labeled ground truth based on polygonal areas.

When using this dataset in your research, we will be happy if you cite us:
@INPROCEEDINGS{Geiger2012CVPR,
  author = {Andreas Geiger and Philip Lenz and Raquel Urtasun},
  title = {Are we ready for Autonomous Driving? The KITTI Vision Benchmark Suite},
  booktitle = {Conference on Computer Vision and Pattern Recognition (CVPR)},
  year = {2012}
}