search

XiandaGuo / OpenStereo

OpenStereo: A Comprehensive Benchmark for Stereo Matching and Strong Baseline
473 stars 51 forks source link

Mistake in paper table7 #95

Closed Guptajakala closed 7 months ago

Guptajakala commented 7 months ago

image The results in the table do not seem to be on the same test set.

In STTR, it's evaluated on the flyingthings subset of sceneflow. In IGEV, it's evaluated on the whole sceneflow

BTW, what is the set used in StereoBase?

XiandaGuo commented 7 months ago

Thank you for your interest in our work. The evaluation of the STTR model is limited to the FlyingThings subset, while the evaluation of all other models encompasses the whole SceneFlow test dataset.