A Large Dataset to Train Convolutional Networks for Disparity, Optical Flow, and Scene Flow Estimation

Open paper
Benchmark Model Rank Results
stereo-depth-estimation-on-scene-flow-stereo-testDispNetCorr1D (DispNetC)#16EPE: 1.68
stereo-depth-estimation-on-kitti-2015-stereo-testDispNetCorr1D-K (DispNetC, KITTI 2015 fine-tuned) (extra data)#20D1-all: 4.34

Rank counts only results with a code link.