search

StanfordORB / Stanford-ORB

Official Repository of the Stanford-ORB dataset
73 stars 2 forks source link

Availability of trained models #3

Closed Xiaoming-Zhao closed 8 months ago

Xiaoming-Zhao commented 9 months ago

Hi, thanks a lot for building this amazing dataset.

The paper states that "all data, trained models, and the evaluation protocol will be released" but I am unable to find those trained models on the webpage or 3dorb repo. I just want to make sure that I did not miss anything.

I understand that the cleanup is cumbersome and could take a lot of time. If this is the case, I am wondering whether it is possible to first release one trained model on all scenes and their corresponding results so that we can easily play around with the evaluation scripts with no need to train the model or even run the inference ourselves.

The reference model and results can help the user gain confidence that any modifications of the evaluation scripts and future comparisons are fair and valid since the user can easily make sure that they can reproduce the results.

Thanks a lot in advance.

zfkuang commented 8 months ago

Please email us if you'd like to download the training models of the baselines.