Some questions

[lhthq]

Thanks for you great work and open-source code. I have read your paper. And I meet some questions. Which sequences of both datasets are used for traning? Could you please provide the pretrained model and test data in the README format? Thank you.

[wangziyue-nubot]

Thank you for your attention. Following the setting of DiffLoc, we trained on the NCLT dataset using sequences 2012-01-22, 2012-02-02, 2012-02-18, 2012-05-11 and tested using sequences 2012-02-12, 2012-02-19, 2012-03-31. On the Oxford dataset, we trained using sequences 11-14-02-26, 14-12-05-52, 14-14-48-55, 18-15-20-12, and tested using sequences 17-13-26-39, 17-14-03-00, 18-14-14-42.

The generation of BEV and supported datasets can be found in Readme.

We will provide pretrained models soon.

[Tcode8]

Thank you for your excellent work. How can I obtain trainsplit.txt