Question on KITTI training

ucbdrive / hd3

Code for Hierarchical Discrete Distribution Decomposition for Match Density Estimation (CVPR 2019)

BSD 3-Clause "New" or "Revised" License

204 stars 31 forks source link

Question on KITTI training #14

Closed AimAlex closed 5 years ago

AimAlex commented 5 years ago

Hi. In the stereo matching task, you trained the model on KITTI for 2000 epochs. But do you have a validation dataset? How could you know which epoch is the best model? Are you using the latest model to test?

I got similar result on the training set, and used the latest model to run the test. However the result is not as good as yours on the test dataset.

Thank you.

yzcjtr commented 5 years ago

There's no validation set on KITTI. We just use the model of the latest epoch for testing. How is your result on the test set of KITTI? Did you use the context module?

AimAlex commented 5 years ago

屏幕快照 2019-08-01 上午11 12 21

This is my result and I used the context module.

yzcjtr commented 5 years ago

Looks good. I think the result is plausible, considering different random seeds and the scarcity of training data on KITTI. Besides, numbers in stereo matching on the KITTI benchmark is saturated, making subtle differences not that important.