You can directly follow HERE for OpenScene data pre-processing, or follow the step-by-step instructions below, which is the same as above link.
Download and Unzip Data
-
Download and unzip all the sensor data from nuPlan. For ViDAR, we use both the
Camera
sensor data and theLiDAR
sensor data. We also provide the links below:mini-set
openscene_sensor_mini_camera (84 GB): OpenDriveLab / Hugging Facemini-set
openscene_sensor_mini_lidar (60 GB): OpenDriveLab / Hugging Facetrainval-set
openscene_sensor_trainval_camera (1.1 TB): OpenDriveLabtrainval-set
openscene_sensor_trainval_lidar (822 GB): OpenDriveLabtest-set
openscene_sensor_test_camera (120 GB): OpenDriveLab / Hugging Facetest-set
openscene_sensor_test_lidar (87 GB): OpenDriveLab / Hugging Faceprivate-test
openscene_sensor_private_test_wm (15 GB): Hugging Face
-
Download and unzip our pre-processed
meta_data
at the following links:mini-set
openscene_metadata_mini.tgz (509.6 MB): Google Drive / Hugging Facetrainval-set
openscene_metadata_trainval.tgz (6.6 GB): Google Drive / Hugging Facetest-set
openscene_metadata_test.tgz (454 MB): Google Drive / Hugging Faceprivate-test-set
openscene_metadata_private_test_wm.tgz (7.3 MB): Hugging Face
After downloading and unzipping all the data, soft linking those source data files to the ViDAR/data
folder as the following structure.
ViDAR
├── projects/
├── tools/
├── pretrained/
│ └── r101_dcn_fcos3d_pretrain.pth
└── data/
└── openscene-v1.1
├── meta_datas
| ├── mini
│ │ ├── 2021.05.12.22.00.38_veh-35_01008_01518.pkl
│ │ ├── 2021.05.12.22.28.35_veh-35_00620_01164.pkl
│ │ ├── ...
│ │ └── 2021.10.11.08.31.07_veh-50_01750_01948.pkl
| ├── trainval
| ├── test
| └── private_test_wm
│ └── private_test_wm.pkl
|
└── sensor_blobs
├── mini
│ ├── 2021.05.12.22.00.38_veh-35_01008_01518
│ │ ├── CAM_F0
│ │ │ ├── c082c104b7ac5a71.jpg
│ │ │ ├── af380db4b4ca5d63.jpg
│ │ │ ├── ...
│ │ │ └── 2270fccfb44858b3.jpg
│ │ ├── CAM_B0
│ │ ├── CAM_L0
│ │ ├── CAM_L1
│ │ ├── CAM_L2
│ │ ├── CAM_R0
│ │ ├── CAM_R1
│ │ ├── CAM_R2
│ │ └── MergedPointCloud
│ │ ├── 0079e06969ed5625.pcd
│ │ ├── 01817973fa0957d5.pcd
│ │ ├── ...
│ │ └── fffb7c8e89cd54a5.pcd
│ ├── 2021.06.09.17.23.18_veh-38_00773_01140
│ ├── ...
│ └── 2021.10.11.08.31.07_veh-50_01750_01948
├── trainval
├── test
└── private_test_wm
Prepare OpenScene data
# mini / trainval / test
python tools/collect_nuplan_data.py mini
python tools/collect_nuplan_data.py trainval
python tools/collect_nuplan_data.py test
Download nuScenes V1.0 full dataset data and CAN bus expansion data HERE. Prepare nuscenes data by running the following steps.
Download CAN bus expansion
# download 'can_bus.zip'
unzip can_bus.zip
# move can_bus to data dir
Prepare nuScenes data
python tools/create_data.py nuscenes --root-path ./data/nuscenes --out-dir ./data/nuscenes --extra-tag nuscenes --version v1.0 --canbus ./data
Using the above code will generate nuscenes_infos_temporal_{train,val}.pkl
.
Folder structure
ViDAR
├── projects/
├── tools/
├── pretrained/
│ ├── r101_dcn_fcos3d_pretrain.pth
├── data/
│ ├── can_bus/
│ ├── nuscenes/
│ │ ├── maps/
│ │ ├── samples/
│ │ ├── sweeps/
│ │ ├── v1.0-test/
| | ├── v1.0-trainval/
| | ├── nuscenes_infos_temporal_train.pkl
| | ├── nuscenes_infos_temporal_val.pkl