Shaoshuai Shi

It consumes about 630GB in my server (the directory `waymo_processed_data_v0_5_0`).

It contains 100% samples.

Sorry, we are not sure since there are so many commits from v0.2 to v0.5 and it depends on what model/dataset you are going to use. Maybe you can check...

Hi A, Currently, we only have one multimodal model for KITTI dataset, and you can get some information from this config: https://github.com/open-mmlab/OpenPCDet/blob/master/tools/cfgs/kitti_models/voxel_rcnn_car_focal_multimodal.yaml. The multimodal models with multiple cameras are still...

If you run the model, you can find that it actually has the option to use the image like this: https://github.com/open-mmlab/OpenPCDet/blob/master/pcdet/models/backbones_3d/focal_sparse_conv/focal_sparse_conv.py#L203-L205 > Do you refer to multiple cameras as multiple...

This model is commited by their authors, and by default it is trained with multimodal (the training/inferene is same with other models).

For now, only the model I mentioned above uses point clouds+RGB as inputs.

Thank you very much, that's a good point, maybe we can add an assert function in the python iou3d calculation codes. So the users do not need to re-compile the...

Sorry, currently we do not have a cloud space to store the pre-trained models within China. Maybe you can train it by yourself. If you only need a single model,...

The results in this repo are reported on validation set, and the codes of original paper are almost same with this repo, yet training with different times may have variations...