\begin{tabular}{c | c | c | c | c | c | c | c}
{\bf Method} & {\bf Setting} & {\bf Out-Noc} & {\bf Out-All} & {\bf Avg-Noc} & {\bf Avg-All} & {\bf Runtime} & {\bf Environment}\\ \hline
ViTAStereo & & 1.88 \% & 2.21 \% & 0.7 px & 0.8 px & 0.22 s & NVIDIA RTX 4090 (PyTorch)\\
VFM Adapter & & 1.96 \% & 2.41 \% & 0.7 px & 0.8 px & 1 s & 8 cores @ 3.5 Ghz (Python)\\
HART & & 1.99 \% & 2.49 \% & 0.8 px & 0.8 px & 0.34 s & NVIDIA Tesla A100 (Python)\\
StereoBase & & 2.09 \% & 2.65 \% & 0.8 px & 0.9 px & 0.29 s & GPU @ 1.5 Ghz (Python)\\
LoS & & 2.41 \% & 3.01 \% & 0.8 px & 0.9 px & 0.19 s & 1 core @ 2.5 Ghz (Python)\\
EGLCR-Stereo & & 2.51 \% & 3.17 \% & 0.9 px & 1.0 px & 0.45 s & 1 core @ 2.5 Ghz (C/C++)\\
MPFV-Stereo & & 2.56 \% & 3.17 \% & 0.8 px & 0.9 px & 0.31 s & 1 core @ 2.5 Ghz (Python)\\
MoCha-V2 & & 2.58 \% & 3.19 \% & 0.9 px & 0.9 px & 0.25 s & NVIDIA Tesla A100 (Python)\\
MoCha-Stereo & & 2.62 \% & 3.08 \% & 0.8 px & 0.9 px & 0.33 s & NVIDIA Tesla A6000 (PyTorch)\\
Selective-IGEV & & 2.66 \% & 3.05 \% & 0.9 px & 1.0 px & 0.24 s & 1 core @ 2.5 Ghz (Python)\\
IGEV-dynamic & & 2.69 \% & 3.46 \% & 0.9 px & 1.0 px & 0.2 s & 1 core @ 2.5 Ghz (C/C++)\\
Sn-stereo & & 2.72 \% & 3.36 \% & 0.9 px & 1.0 px & 0.35 s & GPU @ 1.5 Ghz (Python)\\
DMIO & & 2.72 \% & 3.36 \% & 0.9 px & 1.0 px & 0.3 s & 1 core @ 2.5 Ghz (Python)\\
SR Stereo & & 2.72 \% & 3.11 \% & 0.9 px & 1.0 px & 0.18 s & 1 core @ 2.5 Ghz (C/C++)\\
4D-IteraStereo & & 2.82 \% & 3.27 \% & 0.9 px & 1.0 px & 0.4 s & GPU @ 2.5 Ghz (Python)\\
IGEV-ICGNet & & 2.85 \% & 3.53 \% & 0.9 px & 1.0 px & 0.18 s & GPU @ 2.5 Ghz (C/C++)\\
GeoNet & & 2.86 \% & 3.57 \% & 0.9 px & 1.0 px & 0.22 s & 1 core @ 2.5 Ghz (C/C++)\\
RiskMin & & 2.87 \% & 3.74 \% & 0.9 px & 1.1 px & 0.20 s & GPU @ 2.5 Ghz (Python)\\
Fine-stereo & & 2.87 \% & 3.27 \% & 0.9 px & 1.0 px & 0.32 s & 1 core @ 2.5 Ghz (C/C++)\\
IGEV-Stereo(32) & & 2.92 \% & 3.35 \% & 0.9 px & 1.0 px & 0.32 s & NVIDIA RTX 3090 (PyTorch)\\
GSSNet & & 2.97 \% & 3.80 \% & 1.1 px & 1.2 px & 0.78 s & 1 core @ 2.5 Ghz (Python)\\
ADBM & & 2.97 \% & 3.43 \% & 0.9 px & 1.0 px & 0.4 s & 1 core @ 2.5 Ghz (C/C++)\\
Any-IGEV & & 3.03 \% & 3.39 \% & 1.0 px & 1.1 px & 0.32 s & 1 core @ 2.5 Ghz (C/C++)\\
MC-Stereo & & 3.06 \% & 3.93 \% & 1.0 px & 1.1 px & 0.40 s & GPU @ 2.5 Ghz (Python)\\
PCMAnet & & 3.10 \% & 4.45 \% & 1.0 px & 1.1 px & 0.27 s & 1 core @ 2.5 Ghz (C/C++)\\
ICGNet-abl & & 3.11 \% & 3.67 \% & 1.0 px & 1.0 px & 0.18s & 1 core @ 2.5 Ghz (C/C++)\\
Ms\_Igev & & 3.14 \% & 3.91 \% & 1.1 px & 1.1 px & 0.5 s & GPU @ 2.5 Ghz (Python)\\
MR\_Igev & & 3.15 \% & 3.78 \% & 0.9 px & 1.0 px & 0.5 s & 1 core @ 2.5 Ghz (C/C++)\\
IGEV-Stereo & & 3.16 \% & 3.57 \% & 1.0 px & 1.1 px & 0.18 s & NVIDIA RTX 3090 (PyTorch)\\
EGA-Stereo & & 3.18 \% & 4.20 \% & 1.0 px & 1.1 px & 0.41 s & 1 core @ 2.5 Ghz (Python)\\
DualNet-kd* & & 3.20 \% & 3.91 \% & 1.0 px & 1.1 px & 0.3 s & 1 core @ 2.5 Ghz (C/C++)\\
CAS++ & & 3.22 \% & 3.84 \% & 1.3 px & 1.3 px & .1 s & 1 core @ 2.5 Ghz (C/C++)\\
ff-stereo & & 3.22 \% & 3.84 \% & 1.3 px & 1.3 px & 0.15 s & GPU @ 3.0 Ghz (Python)\\
Selective-RAFT & & 3.27 \% & 3.49 \% & 1.1 px & 1.2 px & 0.45 s & 1 core @ 2.5 Ghz (Python)\\
MAStereo & & 3.33 \% & 4.14 \% & 1.1 px & 1.2 px & 0.07 s & GPU @ 2.5 Ghz (Python)\\
PCWNet & & 3.38 \% & 4.19 \% & 1.0 px & 1.2 px & 0.44 s & 1 core @ 2.5 Ghz (C/C++)\\
MAStereo & & 3.38 \% & 4.32 \% & 1.1 px & 1.2 px & 0.06 s & GPU @ 2.5 Ghz (Python)\\
GANet+ADL & & 3.43 \% & 4.39 \% & 1.2 px & 1.4 px & 0.67 s & NVIDIA RTX 3090 (PyTorch) \\
PCWNet-SCE & & 3.44 \% & 4.39 \% & 1.1 px & 1.3 px & 0.44 s & 1 core @ 2.5 Ghz (C/C++)\\
MVACVNet & & 3.45 \% & 4.40 \% & 1.0 px & 1.1 px & 0.01 s & GPU @ 2.5 Ghz (Python)\\
CGF-ACV & & 3.45 \% & 4.40 \% & 1.0 px & 1.1 px & 0.24 s & NVIDIA RTX 3090 (PyTorch)\\
UGNet & & 3.47 \% & 4.62 \% & 1.1 px & 1.3 px & 0.3 s & GPU @ 2.5 Ghz (Python)\\
HD^3-Stereo & & 3.49 \% & 4.88 \% & 1.0 px & 1.1 px & 0.14 s & NVIDIA Pascal Titan XP\\
WiCRI\_STEREO & & 3.53 \% & 4.42 \% & 1.0 px & 1.1 px & 1 s & 1 core @ 2.5 Ghz (C/C++)\\
OnestageStereo & & 3.53 \% & 4.64 \% & 1.1 px & 1.2 px & 1 s & GPU @ 2.5 Ghz (C/C++)\\
LEAStereo & & 3.56 \% & 4.32 \% & 1.1 px & 1.2 px & 0.3 s & GPU @ 2.5 Ghz (Python)\\
DCVSMNet & & 3.58 \% & 4.36 \% & 1.2 px & 1.3 px & 0.07 s & GPU @ 2.5 Ghz (Python)\\
DN+ACVNet & & 3.63 \% & 4.74 \% & 1.2 px & 1.4 px & 0.24 s & 1 core @ 2.5 Ghz (C/C++)\\
GEMAStereo & & 3.63 \% & 4.66 \% & 1.1 px & 1.2 px & 0.03 s & GPU @ 2.5 Ghz (Python)\\
Any-RAFT & & 3.65 \% & 4.27 \% & 1.2 px & 1.2 px & 0.32 s & 1 core @ 2.5 Ghz (C/C++)\\
NeXt-Stereo & & 3.69 \% & 4.31 \% & 1.1 px & 1.2 px & 0.06 s & GPU @ 2.0 Ghz (Python)\\
EdgeStereo-V2 & & 3.78 \% & 5.06 \% & 1.0 px & 1.2 px & 0.32 s & Nvidia GTX Titan Xp\\
AutoDispNet-CSS & & 3.78 \% & 4.47 \% & 1.0 px & 1.1 px & 0.9 s & 1 core @ 2.5 Ghz (C/C++)\\
ACVNet-DCA & & 3.80 \% & 4.58 \% & 1.1 px & 1.3 px & 0.2 s & 1 core @ 2.5 Ghz (C/C++)\\
MDA & & 3.87 \% & 5.13 \% & 1.2 px & 1.4 px & 0.32s & 1 core @ 2.5 Ghz (Python)\\
ICVP & & 3.94 \% & 4.74 \% & 1.2 px & 1.2 px & 0.17 s & GPU @ 1.5 Ghz (Python)\\
MAF-Stereo & & 3.99 \% & 5.20 \% & 1.2 px & 1.4 px & 0.07 s & GPU @ 2.5 Ghz (Python)\\
DVANet & & 3.99 \% & 5.30 \% & 1.2 px & 1.3 px & 0.28 s & NVIDIA 3090 (PyTorch)\\
DispNO & & 4.02 \% & 5.07 \% & 1.2 px & 1.4 px & 0.73 s & GPU @ 3.0 Ghz (Python)\\
HITNet & & 4.04 \% & 5.34 \% & 1.0 px & 1.2 px & 0.02 s & GPU @ 2.5 Ghz (Python + C/C++)\\
PDSNet & & 4.05 \% & 5.87 \% & 1.4 px & 1.6 px & 0.5 s & 1 core @ 2.5 Ghz (Python + C/C++)\\
IGEVStereo-DCA & & 4.06 \% & 5.17 \% & 1.2 px & 1.3 px & 0.4 s & 1 core @ 2.5 Ghz (C/C++)\\
NLCA-Net v2 & & 4.06 \% & 5.19 \% & 1.2 px & 1.3 px & 0.67 s & GPU @ >3.5 Ghz (Python)\\
LI-ACVNet & & 4.09 \% & 4.98 \% & 1.3 px & 1.4 px & 0.14 s & GPU @ 2.5 Ghz (Python)\\
PFSMNet & & 4.14 \% & 5.27 \% & 1.2 px & 1.3 px & 0.31 s & 1 core @ 2.5 Ghz (C/C++)\\
UCFNet & & 4.15 \% & 4.99 \% & 1.1 px & 1.2 px & 0.21 s & 1 core @ 2.5 Ghz (C/C++)\\
LaC+GANet & & 4.15 \% & 5.18 \% & 1.3 px & 1.5 px & 1.8 s & 1 core @ 2.5 Ghz (C/C++)\\
ADPNet & & 4.20 \% & 5.55 \% & 1.3 px & 1.5 px & 0.06 s & 1 core @ 2.5 Ghz (C/C++)\\
SegStereo & & 4.22 \% & 5.52 \% & 1.1 px & 1.3 px & 0.6 s & Nvidia GTX Titan Xp\\
CFNet & & 4.24 \% & 5.05 \% & 1.2 px & 1.3 px & 0.18 s & 1 core @ 2.5 Ghz (Python)\\
BGNet+ & & 4.26 \% & 5.80 \% & 1.2 px & 1.4 px & 0.02 s & GPU @ 2.5 Ghz (Python)\\
GANet-deep & & 4.32 \% & 5.66 \% & 1.2 px & 1.3 px & 1.8 s & GPU @ 2.5 Ghz (Python)\\
TBFE-Net & & 4.34 \% & 5.50 \% & 1.3 px & 1.4 px & 0.3 s & 1 core @ 2.5 Ghz (Python)\\
ProNet & & 4.35 \% & 5.56 \% & 1.2 px & 1.4 px & 0.33 s & GPU @ 2.5 Ghz (Python)\\
ERSCNet & & 4.35 \% & 6.20 \% & 1.2 px & 1.4 px & 0.28 s & GPU @ 2.5 Ghz (Python)\\
AMNet & & 4.44 \% & 5.62 \% & 1.2 px & 1.3 px & 0.9 s & GPU @ 2.5 Ghz (Python)\\
HCR & & 4.46 \% & 5.68 \% & 1.2 px & 1.3 px & 0.19 s & GPU @ 2.5 Ghz (Python)\\
DPCTF-S & & 4.47 \% & 5.70 \% & 1.2 px & 1.4 px & 0.11 s & GPU @ 2.5 Ghz (Python)\\
FusionStereo & & 4.49 \% & 5.82 \% & 1.2 px & 1.3 px & 16 s & 1 core @ 2.5 Ghz (Python)\\
LaC+GwcNet & & 4.55 \% & 5.89 \% & 1.5 px & 1.7 px & 0.65 s & GPU @ 2.5 Ghz (Python)\\
SCV-Stereo & & 4.57 \% & 5.96 \% & 1.2 px & 1.4 px & 0.08 s & GPU @ 2.5 Ghz (Python)\\
OptStereo & & 4.58 \% & 5.87 \% & 1.3 px & 1.4 px & 0.10 s & GPU @ 2.5 Ghz (Python)\\
GINet & & 4.58 \% & 5.86 \% & 1.3 px & 1.4 px & 0.25 s & 2 cores @ 2.5 Ghz (Python)\\
CoEx & & 4.61 \% & 6.00 \% & 1.3 px & 1.4 px & 0.027 s & RTX 2080Ti (Python)\\
NLCA-Net-3 & & 4.63 \% & 5.57 \% & 1.2 px & 1.4 px & 0.44 s & GPU @ 2.5 Ghz (Python)\\
CAL-Net & & 4.66 \% & 5.96 \% & 1.5 px & 1.7 px & 0.44 s & 4 cores @ 2.5 Ghz (Python)\\
VCNet & & 4.68 \% & 5.67 \% & 1.4 px & 1.5 px & 0.6 s & 1 core @ 2.5 Ghz (Python)\\
PGNet & & 4.70 \% & 6.13 \% & 1.3 px & 1.5 px & 0.7 s & 1 core @ 2.5 Ghz (python)\\
Abc-Net & & 4.73 \% & 5.92 \% & 1.4 px & 1.5 px & 0.72 s & 4 cores @ 2.5 Ghz (Python)\\
DANet-Stereo & & 4.75 \% & 5.91 \% & 1.3 px & 1.5 px & 2.7 s & GPU @ 2.5 Ghz (Python)\\
NMRF-Stereo & & 4.80 \% & 6.09 \% & 1.4 px & 1.6 px & 0.09 s & NVIDIA RTX 3090 (PyTorch)\\
UAIStereo & & 4.81 \% & 6.09 \% & 1.3 px & 1.4 px & 0.06 s & GPU @ 3.5 Ghz (Python)\\
FGDS-Net & & 4.87 \% & 6.05 \% & 1.3 px & 1.4 px & 0.3 s & 1 core @ 2.5 Ghz (Python)\\
SGNet & & 4.87 \% & 6.27 \% & 1.4 px & 1.5 px & 0.6 s & 1 core @ 2.5 Ghz (Python + C/C++)\\
CREStereo & & 4.93 \% & 5.55 \% & 1.4 px & 1.4 px & 0.40 s & GPU @ >3.5 Ghz (C/C++)\\
SSPCVNet & & 4.98 \% & 6.55 \% & 1.5 px & 1.7 px & 0.9 s & 1 core @ 2.5 Ghz (Python)\\
ICGNet-gwc & & 5.07 \% & 6.51 \% & 1.3 px & 1.4 px & 0.32 s & GPU @ 2.5 Ghz (Python)\\
iResNet-i2 & & 5.07 \% & 6.34 \% & 1.2 px & 1.3 px & 0.12 s & 1 core @ 2.5 Ghz (C/C++)\\
HSM & & 5.13 \% & 6.77 \% & 1.3 px & 1.6 px & 0.15 s & Titan X Pascal\\
ACVNet & & 5.18 \% & 6.48 \% & 1.4 px & 1.5 px & 0.2 s & NVIDIA RTX 3090 (PyTorch)\\
NLCA-Net & & 5.19 \% & 6.43 \% & 1.3 px & 1.4 px & 0.6 s & GPU @ 2.5 Ghz (C/C++)\\
AcfNet & & 5.20 \% & 6.44 \% & 1.8 px & 1.9 px & 0.48 s & 1 core @ 2.5 Ghz (Python)\\
AANet+ & & 5.25 \% & 6.66 \% & 1.3 px & 1.4 px & 0.06 s & NVIDIA V100 GPU\\
CFP-Net & & 5.26 \% & 6.90 \% & 1.4 px & 1.6 px & 0.95 s & 8 cores @ 2.5 Ghz (Python)\\
MABNet\_origin & & 5.29 \% & 6.33 \% & 1.3 px & 1.4 px & 0.38 s & Nvidia rtx2080ti (Python)\\
DFGA-Net & & 5.37 \% & 6.95 \% & 1.3 px & 1.5 px & 0.09 s & NVIDIA RTX 3090 (PyTorch)\\
yjlig & & 5.48 \% & 6.95 \% & 1.3 px & 1.5 px & 0.35 s & 1 core @ 2.5 Ghz (C/C++)\\
EAC-Stereo & & 5.48 \% & 6.95 \% & 1.3 px & 1.5 px & 0.38s & 1 core @ 2.5 Ghz (Python)\\
GANet-15 & & 5.49 \% & 7.10 \% & 1.3 px & 1.5 px & 0.36 s & 1 core @ 2.5 Ghz (C/C++)\\
FADNet & & 5.57 \% & 6.51 \% & 1.2 px & 1.3 px & 0.05 s & Tesla V100 (Python)\\
GwcNet-gc & & 5.69 \% & 6.70 \% & 1.3 px & 1.4 px & 0.32 s & GPU @ 2.0 Ghz (Java + C/C++)\\
MCVFNet & & 5.87 \% & 7.58 \% & 1.5 px & 1.7 px & 0.029 s & RTX 2080TI\\
PSMNet & & 5.89 \% & 7.29 \% & 1.4 px & 1.6 px & 0.41 s & Nvidia Titan Xp\\
Fast DS-CS & & 6.01 \% & 7.23 \% & 1.3 px & 1.5 px & 0.02 s & GPU @ 2.0 Ghz (Python + C/C++)\\
WSMCnet & & 6.08 \% & 7.50 \% & 2.1 px & 2.3 px & 0.39 s & GPU @ Nvidia GTX 1070 (Pytorch)\\
Displets & & 6.11 \% & 7.23 \% & 1.9 px & 2.3 px & 265 s & >8 cores @ 3.0 Ghz (Matlab + C/C++)\\
PVStereo & & 6.17 \% & 7.06 \% & 2.2 px & 2.3 px & 0.10 s & 1 core @ 2.5 Ghz (Python)\\
RTSnet & & 6.69 \% & 8.25 \% & 1.7 px & 1.9 px & 0.02 s & 1 core @ 2.5 Ghz (Python)\\
MSDC-Net & & 6.72 \% & 8.37 \% & 1.8 px & 2.0 px & 0.6 s & 1 core @ 2.5 Ghz (C/C++)\\
MMStereo & & 6.78 \% & 8.23 \% & 1.6 px & 1.7 px & 0.04 s & Nvidia Titan RTX (Python)\\
Displets v2 & & 6.92 \% & 8.02 \% & 2.0 px & 2.2 px & 265 s & >8 cores @ 3.0 Ghz (Matlab + C/C++)\\
DualNet-kd & & 7.06 \% & 8.48 \% & 1.7 px & 1.9 px & 0.3 s & 1 core @ 2.5 Ghz (C/C++)\\
WaveletStereo & & 7.38 \% & 9.27 \% & 1.7 px & 1.9 px & 0.27 s & 1 core @ 2.5 Ghz (C/C++)\\
AANet & & 7.89 \% & 8.91 \% & 1.7 px & 1.8 px & 0.06 s & GPU @ 2.5 Ghz (Python)\\
GC-NET & & 8.13 \% & 9.77 \% & 1.8 px & 2.0 px & 0.9 s & Nvidia GTX Titan X\\
SMFormer & & 8.16 \% & 9.72 \% & 1.7 px & 1.9 px & 0.36 s & 1 core @ 2.5 Ghz (C/C++)\\
VC-SF & fl mv & 9.11 \% & 9.64 \% & 2.7 px & 2.8 px & 300 s & 1 core @ 2.5 Ghz (C/C++)\\
PRSM & fl mv & 9.13 \% & 9.75 \% & 2.0 px & 2.1 px & 300 s & 1 core @ 2.5 Ghz (C/C++)\\
OSF & fl & 9.27 \% & 12.24 \% & 2.1 px & 2.6 px & 50 min & 1 core @ 3.0 Ghz (Matlab + C/C++)\\
CKDNet\_1.0 & & 9.34 \% & 11.52 \% & 2.5 px & 2.9 px & 1 s & 1 core @ 2.5 Ghz (C/C++)\\
CKDNet\_0.3 & & 10.41 \% & 12.89 \% & 2.3 px & 2.8 px & 1 s & 1 core @ 2.5 Ghz (C/C++)\\
RecResNet & & 10.64 \% & 13.40 \% & 2.1 px & 2.5 px & 0.3 s & GPU @ NVIDIA TITAN X (Tensorflow)\\
PCBP-SS & & 10.74 \% & 14.75 \% & 2.4 px & 3.9 px & 5 min & 4 cores @ 2.5 Ghz (Matlab + C/C++)\\
SsSMnet & & 10.87 \% & 13.21 \% & 3.1 px & 3.6 px & 0.8 s & Titan Xp\\
SSMNet & & 10.88 \% & 12.82 \% & 2.4 px & 2.6 px & 0.01 s & GPU @ 2.0 Ghz (Python)\\
JSOSM & & 10.90 \% & 13.35 \% & 2.4 px & 2.8 px & 105 s & 8 cores @ 2.5 Ghz (C/C++)\\
CKDNet\_0.5 & & 11.24 \% & 13.77 \% & 3.1 px & 3.8 px & 1 s & 1 core @ 2.5 Ghz (C/C++)\\
DispNetC & & 11.38 \% & 13.19 \% & 2.1 px & 2.3 px & 0.06 s & Nvidia GTX Titan X (Caffe)\\
FD-Fusion & & 11.38 \% & 13.56 \% & 2.6 px & 2.9 px & 0.01 s & 1 core @ 2.5 Ghz (C/C++)\\
P3SNet+ & & 11.66 \% & 14.13 \% & 2.1 px & 2.4 px & 0.01 s & GPU @ 2.5 Ghz (Python)\\
cfusion & mv & 11.76 \% & 12.43 \% & 2.6 px & 2.8 px & 70 s & GPU (Matlab + CUDA)\\
SPS-StFl & fl ms & 11.94 \% & 14.88 \% & 2.9 px & 3.6 px & 35 s & 1 core @ 3.5 Ghz (C/C++)\\
SGM-Net & & 12.18 \% & 15.62 \% & 3.0 px & 3.8 px & 67 s & Titan X\\
CoR & & 12.21 \% & 15.95 \% & 2.7 px & 4.1 px & 6 s & 6 cores @ 3.3 Ghz (Matlab + C/C++)\\
MABNet\_tiny & & 12.34 \% & 14.45 \% & 2.5 px & 2.8 px & 0.11 s & 1 core @ 2.5 Ghz (Python)\\
DDS-SS & & 12.48 \% & 15.39 \% & 2.5 px & 3.0 px & 1 min & 1 core @ 2.5 Ghz (Matlab + C/C++)\\
CNNF+SGM & & 12.66 \% & 16.12 \% & 2.9 px & 3.9 px & 71 s & TESLA K40C\\
PCBP & & 12.66 \% & 16.57 \% & 2.8 px & 4.4 px & 5 min & 4 cores @ 2.5 Ghz (Matlab + C/C++)\\
L-ResMatch & & 12.92 \% & 16.52 \% & 3.7 px & 5.2 px & 48 s & Titan X (Torch7, CUDA)\\
P3SNet & & 13.01 \% & 15.71 \% & 2.3 px & 2.8 px & 0.01 s & GPU @ 2.5 Ghz (Python)\\
SPS-St & & 13.11 \% & 16.11 \% & 3.1 px & 3.6 px & 2 s & 1 core @ 3.5 Ghz (C/C++)\\
Deep Embed & & 13.18 \% & 16.70 \% & 2.8 px & 3.8 px & 3 s & 1 core @ 2.5 Ghz (C/C++)\\
PBCP & & 13.40 \% & 16.75 \% & 3.1 px & 4.2 px & 68 s & Nvidia GTX Titan X\\
CBMV & & 13.61 \% & 17.01 \% & 3.7 px & 4.4 px & 250 s & 6 cores@3.0Ghz(Python,C/C++,CUDA TitanX)\\
SMCM & & 13.64 \% & 17.93 \% & 2.8 px & 4.8 px & 1800 s & Nvidia GTX 1080 (Caffe)\\
MC-CNN-acrt & & 13.76 \% & 17.17 \% & 3.2 px & 4.1 px & 67 s & Nvidia GTX Titan X (CUDA, Lua/Torch7)\\
MC-CNN-WS & & 14.12 \% & 17.77 \% & 3.1 px & 3.8 px & 1.35 s & 1 core 2.5 Ghz + K40 NVIDIA, Lua-Torch\\
StereoSLIC & & 14.32 \% & 17.59 \% & 2.8 px & 3.6 px & 2.3 s & 1 core @ 3.0 Ghz (C/C++)\\
PR-Sf+E & fl & 14.45 \% & 17.28 \% & 3.3 px & 4.0 px & 200 s & 4 cores @ 3.0 Ghz (Matlab + C/C++)\\
SACA & & 14.63 \% & 19.62 \% & 3.5 px & 7.1 px & 5 s & GPU @ 2.5 Ghz (Python)\\
SAFT-Stereo & & 14.77 \% & 16.72 \% & 3.2 px & 3.5 px & 0.007 s & NVIDIA GeForce RTX 4090\\
MC-CNN & & 14.93 \% & 18.22 \% & 3.5 px & 4.3 px & 100 s & Nvidia GTX Titan (CUDA, Lua/Torch7)\\
pSGM & & 14.96 \% & 18.98 \% & 3.3 px & 5.7 px & 7.92 s & 4 cores @ 3.5 Ghz (C/C++)\\
TCD-CRF & & 14.98 \% & 19.36 \% & 3.3 px & 6.6 px & 60 s & 4 cores @ 3.5 Ghz (C/C++)\\
DLP & & 15.26 \% & 19.42 \% & 3.2 px & 5.9 px & 60 s & 8 cores @ >3.5 Ghz (C/C++)\\
Content-CNN & & 15.53 \% & 18.91 \% & 3.6 px & 4.2 px & 0.7 s & Nvidia GTX Titan X (Torch)\\
PR-Sceneflow & fl & 15.59 \% & 18.28 \% & 3.3 px & 4.0 px & 150 sec & 4 core @ 3.0 Ghz (Matlab - C/C++)\\
CRD-Fusion & & 15.65 \% & 18.41 \% & 2.9 px & 3.5 px & 0.02 s & GPU @ 2.5 Ghz (Python)\\
MBM & & 16.38 \% & 19.53 \% & 3.6 px & 4.3 px & 0.2 s & 1 core @ 3.0 Ghz (C/C++)\\
CoR-Conf & & 16.59 \% & 20.08 \% & 4.0 px & 5.4 px & 6 s & 6 cores @ 3.3 Ghz (Matlab + C/C++)\\
UHP & & 16.67 \% & 19.67 \% & 2.9 px & 3.7 px & 0.02 s & GPU @ 2.5 Ghz (Python)\\
SGM-post & & 16.96 \% & 19.78 \% & 3.3 px & 3.8 px & 5 s & 4 cores @ 2.5 Ghz (C/C++)\\
Flow2Stereo & & 16.96 \% & 18.84 \% & 2.8 px & 3.1 px & 0.05 s & GPU @ 2.5 Ghz (Python)\\
CAT & & 17.38 \% & 20.46 \% & 3.3 px & 4.3 px & 10 s & 1 core @ 3.5 Ghz (C/C++)\\
RBM & & 17.90 \% & 20.65 \% & 3.6 px & 4.0 px & 0.2 s & 1 core @ 3.0 Ghz (C/C++)\\
AARBM & & 18.24 \% & 21.00 \% & 3.7 px & 4.2 px & 0.25 s & 1 core @ 3.0 Ghz (C/C++)\\
Ensemble & & 18.28 \% & 21.95 \% & 4.5 px & 6.9 px & 135 s & 2 cores @ >3.5 Ghz (Matlab)\\
SNCC & & 18.36 \% & 21.16 \% & 3.7 px & 4.2 px & 0.11 s & 1 core @ 3.1 Ghz (C/C++)\\
ATGV & & 18.43 \% & 22.49 \% & 3.3 px & 5.3 px & 6 min & >8 cores @ 3.0 Ghz (Matlab + C/C++)\\
CSPMS & & 18.59 \% & 22.41 \% & 4.5 px & 6.0 px & 6 s & 4 cores @ 2.5 Ghz (C/C++)\\
AABM & & 18.65 \% & 21.45 \% & 3.8 px & 4.4 px & 0.12 s & 1 core @ 3.1 Ghz (C/C++)\\
ALTGV & & 18.86 \% & 22.19 \% & 3.4 px & 4.2 px & 20 s & GPU @ 2.5 Ghz (C/C++)\\
AAFS & & 19.01 \% & 21.31 \% & 3.3 px & 3.8 px & 0.01 s & 1 core @ 2.5 Ghz (C/C++)\\
ITGV & & 19.05 \% & 22.52 \% & 4.9 px & 5.8 px & 7 s & 1 core @ 3.0 Ghz (Matlab + C/C++)\\
DispSegNet & & 20.15 \% & 22.96 \% & 3.7 px & 4.2 px & 0.9 s & GPU @ 2.5 Ghz (Python)\\
ARW & & 20.17 \% & 23.92 \% & 5.2 px & 6.8 px & 4.6s & 1 core @ 3.5 Ghz (MATLAB+C/C++)\\
iSGM & & 20.52 \% & 24.84 \% & 4.7 px & 8.6 px & 8 s & 2 cores @ 2.5 Ghz (C/C++)\\
mSGM-LDE & & 20.96 \% & 25.53 \% & 5.5 px & 9.3 px & 55 s & 2 cores @ 2.5 Ghz (C/C++)\\
Permutation Stereo & & 21.96 \% & 24.90 \% & 5.0 px & 5.8 px & 30 s & GPU @ 2.5 Ghz (Matlab)\\
wSGM & & 22.07 \% & 25.17 \% & 8.0 px & 9.0 px & 6s & 1 core @ 3.5 Ghz (C/C++)\\
ELAS & & 22.49 \% & 26.13 \% & 5.4 px & 6.3 px & 0.3 s & 1 core @ 2.5 Ghz (C/C++)\\
S+GF (Cen) & & 22.58 \% & 27.10 \% & 5.6 px & 10.1 px & 140 s & 1 core @ 3.0 Ghz (C/C++)\\
linBP & & 23.00 \% & 27.42 \% & 5.9 px & 9.6 px & 1.6 min & 1 core @ 3.0 Ghz (C/C++)\\
Toast2 & st & 23.11 \% & 26.12 \% & 5.0 px & 5.7 px & 0.03 s & 4 cores @ 3.5 Ghz (C/C++)\\
rSGM & & 23.20 \% & 26.71 \% & 6.7 px & 7.9 px & 0.2 s & 4 cores @ 2.6 Ghz (C/C++)\\
SGM & & 23.22 \% & 26.03 \% & 5.1 px & 5.8 px & 3.7 s & 1 core @ 3.0 Ghz (C/C++)\\
SSMW & & 23.53 \% & 26.59 \% & 5.2 px & 6.2 px & 2.5 min & 8 cores @ 2.5 Ghz (C/C++)\\
OASM-Net & & 23.79 \% & 28.18 \% & 4.6 px & 7.3 px & 0.73 s & GPU @ 2.5 Ghz (Python)\\
HSMA & & 24.04 \% & 28.43 \% & 5.6 px & 9.3 px & 44s & 1 core @ 3.0 Ghz (C/C++)\\
LAMC-DSΜ & & 24.72 \% & 28.38 \% & 5.1 px & 8.0 px & 10.8 min & 2 cores @ 2.5 Ghz (Matlab)\\
GF (Census) & & 26.71 \% & 30.99 \% & 8.8 px & 12.9 px & 120 s & 1 core @ 3.0 Ghz (C/C++)\\
Deep-Raw & & 27.70 \% & 31.87 \% & 9.8 px & 13.2 px & 1 s & 1 core @ 2.5 Ghz (C/C++)\\
BSM & & 27.72 \% & 31.35 \% & 7.5 px & 10.3 px & 2.5 min & 1 core @ 3.0 Ghz (C/C++)\\
IIW & & 28.34 \% & 31.82 \% & 8.5 px & 11.7 px & 5.5 s & 1 core @ 2.5 Ghz (C/C++)\\
GLDS & & 28.48 \% & 31.84 \% & 5.8 px & 7.5 px & 26 s & GPU @ 1.5 Ghz (C/C++)\\
ADSM & & 28.58 \% & 31.54 \% & 8.4 px & 10.6 px & 125 s & 1 core @ 2.0 Ghz (C/C++)\\
SM\_GPTM & & 29.68 \% & 32.77 \% & 11.7 px & 12.9 px & 6.5 s & 2 cores @ 2.5 Ghz (C/C++)\\
SymST-GP & & 30.18 \% & 33.86 \% & 12.1 px & 15.2 px & 0.254 s & Dual - Nvidia GTX Titan (CUDA)\\
HLSC\_mesh & & 31.04 \% & 34.50 \% & 9.0 px & 11.3 px & 800 s & 1 core @ 2.5 Ghz (Matlab + C/C++)\\
CrossCensus & & 31.35 \% & 34.22 \% & 8.0 px & 10.1 px & 30 s & 1 core @ 2.5 Ghz (C/C++)\\
GCSF & fl & 31.46 \% & 33.93 \% & 5.9 px & 6.7 px & 2.4 s & 1 core @ 2.5 Ghz (C/C++)\\
GCS & & 32.01 \% & 34.42 \% & 6.2 px & 7.0 px & 2.2 s & 1 core @ 2.5 Ghz (C/C++)\\
SDM & & 32.42 \% & 34.83 \% & 7.7 px & 8.6 px & 1 min & 1 core @ 2.5 Ghz (C/C++)\\
OCV-BM-post & & 32.83 \% & 35.71 \% & 8.6 px & 9.3 px & 0.1 s & 1 core @ 2.5 Ghz (C/C++)\\
MSMW & st & 33.69 \% & 35.97 \% & 9.3 px & 10.0 px & 3 min & 4 cores @ 2.5 Ghz (C/C++)\\
OCV-SGBM & & 34.00 \% & 36.78 \% & 13.1 px & 14.0 px & 1.1 s & 1 core @ 2.5 Ghz (C/C++)\\
CostFilter & & 35.43 \% & 37.95 \% & 11.8 px & 13.5 px & 4 min & 1 core @ 2.5 Ghz (Matlab)\\
VariableCros & & 37.67 \% & 40.73 \% & 13.8 px & 16.1 px & 30 s & 1 core @ 2.5 Ghz (Matlab)\\
GC+occ & & 42.30 \% & 45.12 \% & 14.9 px & 17.6 px & 6 min & 1 core @ 2.5 Ghz (C/C++)\\
ALE-Stereo & & 80.56 \% & 81.22 \% & 24.6 px & 25.4 px & 50 min & 1 core @ 3.0 Ghz (C/C++)\\
AVERAGE & & 84.41 \% & 85.32 \% & 26.2 px & 28.2 px & 0.01 s & 1 core @ 2.5 Ghz (C/C++)\\
MEDIAN & & 87.87 \% & 88.57 \% & 29.6 px & 31.5 px & 0.01 s & 1 core @ 2.5 Ghz (C/C++)
\end{tabular}