GREAT-IGEV

• Original code: https://github.com/JarvisLee0423/GREAT-Stereo at 2a6b1efede8c
• Code licence: Apache-2.0 (permissive); files: LICENSE
• Weights licence: Apache-2.0 (assumed from code)
• Licence evidence: https://github.com/JarvisLee0423/GREAT-Stereo/blob/2a6b1efede8c3a9eec9540790c2700ea2f3dee7d/README.md#L558
• Training data: Freiburg SceneFlow (FlyingThings3D, Driving, Monkaa)
  • great-igev-kitti: Freiburg SceneFlow (FlyingThings3D, Driving, Monkaa), KITTI stereo 2012 / 2015 (SceneFlow, then fine-tuned on KITTI 2012 + 2015 (launchers/.../great_igev_trainer_kitti.sh; KITTI submission model))
  • great-igev-eth3d: Freiburg SceneFlow (FlyingThings3D, Driving, Monkaa), CREStereo synthetic stereo dataset (Megvii), InStereo2K, ETH3D two-view stereo, not stated by the authors (ETH3D submission model; original recipe (launchers, utils/stereo_matching/data_utils/data_fetchers.py): SceneFlow model fine-tuned on eth3d_finetune (ETH3D, InStereo2K, CREStereo); whether the eth3d_train mixture (+ TartanAir, Sintel) ran first is not stated)
  • great-igev-middlebury: Freiburg SceneFlow (FlyingThings3D, Driving, Monkaa), CREStereo synthetic stereo dataset (Megvii), InStereo2K, HR-VS (high-resolution virtual stereo, Yang et al. CVPR 2019), Middlebury stereo (2014), NVIDIA Falling Things (FAT), not stated by the authors (Middlebury submission model; original recipe (launchers, utils/stereo_matching/data_utils/data_fetchers.py): SceneFlow model fine-tuned on middlebury_finetune (CREStereo, InStereo2K, HR-VS, Middlebury 2005 / 2006 / 2014 / 2021 / MiddEval3 H + F, Falling Things); whether the middlebury_train mixture (+ TartanAir) ran first is not stated)
• Options: iters = 32 (>= 1), cuda_graphs = False (speed only)
• Kernel libraries: igev, igevpp, cutlass_conv, cudnn_hacc
• Variants: great-igev-sceneflow, great-igev-kitti, great-igev-eth3d, great-igev-middlebury
• Outputs: disparity
• Input size: any H x W >= 32, padded to a multiple of 32
• Tested up to (portable, fp16, RTX 5060 Ti 16 GB): great-igev-sceneflow not measured yet; great-igev-kitti not measured yet; great-igev-eth3d not measured yet; great-igev-middlebury not measured yet
• great-igev-kitti, great-igev-eth3d: same network and settings as great-igev-sceneflow, measured once (rows not repeated).
• great-igev-sceneflow original (fp32) 994x1482, original 994x1482, portable 994x1482, kernels 994x1482: measured on great-igev-eth3d (same architecture, other checkpoint)
• great-igev-sceneflow load / first call: original 0.44 s / 4.596 s (375x1242), portable 0.32 s / 1.102† s, kernels 0.23† s / 0.397† s
• great-igev-kitti load / first call: original 0.44 s / 4.596 s (375x1242), portable 0.32 s / 1.102† s, kernels 0.23† s / 0.397† s
• great-igev-eth3d load / first call: original 0.44 s / 4.596 s (375x1242), portable 0.32 s / 1.102† s, kernels 0.23† s / 0.397† s
• great-igev-middlebury load / first call: original 0.39 s / 4.586 s (375x1242), portable 3.91 s / 1.808† s, kernels 0.61† s / 0.589† s
• † measured in the speed sweep after the portable load (or an earlier size) in the same process: file cache and kernels are warm, so the time is lower than a fresh start.
• Protocol, median of 5 after 3 warm-up, uint8 CUDA input (synthetic for bench_inputs sizes), card defaults; gc before each load; peak = max over the first call and the timed calls

Measured

Original = the original code named under Details, same checkpoint and protocol.

Measured on NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, fastforks 468fba10, 2026-11-05.

Parity

CPU, float32: max |diff| <= 0.002 vs the original. GPU, float16 (the default), portable and kernels: mean |diff| <= 0.02 px and p99 <= 0.25 px.