You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi @zhixinwang, will you be interested in adding support for Argoverse dataset to your code base. If yes, please let me know I can send a pull request for the same.
Thank you.
The text was updated successfully, but these errors were encountered:
@ParthasarathySubburaj Very Good. Appreciate your contribution to support other datasets.
If you have already trained successfully and get reasonable results on Argoverse dataset ? You can make a pull request. Make sure not to destroy the original KITTI results.
Hi @zhixinwang, I have written an adaptor that converts argoverse data to KITTI format that is inspired by this repo, where I fixed a few issues w.r.t the transformation and have trained a model for only a few epochs with the images from the front camera only (just 1/7 th of the data) due to my computational limitations. Initial results were pretty good but definitely not close to the results that you got for KIITI, I guess the reason for it is mainly due to the fact that in argoverse the LIDAR data intensity is not as great as KIITI, please refer to the same images for more info. But anyone with access to better computation resources could definitely try training the model on the entire dataset and experiment with your model's hyperparameters. If you are fine with it then I can give a PR for the same.
Hi @zhixinwang, will you be interested in adding support for Argoverse dataset to your code base. If yes, please let me know I can send a pull request for the same.
Thank you.
The text was updated successfully, but these errors were encountered: