Download nuScenes V1.0 full dataset data and CAN bus expansion data HERE. Prepare nuscenes data by running
Download CAN bus expansion
# download 'can_bus.zip'
unzip can_bus.zip
# move can_bus to data dir
Prepare nuScenes data
We genetate custom annotation files which are different from mmdet3d's
python tools/generate_data.py --dataset nuscenes --root-path ./data/nuscenes --out-dir ./data/nuscenes
Using the above code will generate all_cam_nuscenes_map_{train,val}.pkl, which contain local vectorized map annotations.
Folder structure
TIOFUSION
├── datasets/
├── model/
├── tools/
├── configs/
├── ckpts/
│ ├── resnet50-11ad3fa6.pth
├── data/
│ ├── can_bus/
│ ├── nuscenes/
│ │ ├── maps/
│ │ ├── samples/
│ │ ├── sweeps/
│ │ ├── v1.0-test/
| | ├── v1.0-trainval/
| | ├── all_cam_nuscenes_map_train.pkl
| | ├── all_cam_nuscenes_map_val.pkl
Download the Openlane_v1 300 Sensor Dataset here.
Prepare openlane data
We genetate custom annotation files which are different from mmdet3d's
python tools/generate_data.py --dataset openlane --root-path ./data/lane3d_300/ --dataset_base_dir ./data/images/ --out-dir ./data/
Using the above code will generate openlane_map_{train,val}.pkl, which contain local vectorized map annotations.
Folder structure
TIOFUSION
├── datasets/
├── model/
├── tools/
├── configs/
├── ckpts/
│ ├── resnet50-11ad3fa6.pth
├── data/
│ ├── images/
│ ├── lane3d_300/
│ │ ├── test/
│ │ ├── training/
│ │ ├── validation/
│ ├── openlane_map_train.pkl
│ ├── openlane_map_val.pkl