Skip to content

Latest commit

 

History

History
50 lines (38 loc) · 2.11 KB

getting_started.md

File metadata and controls

50 lines (38 loc) · 2.11 KB
Raw

Installation

Follow https://github.com/fundamentalvision/BEVFormer/blob/master/docs/install.md to prepare the environment.

Preparing Dataset

  1. Download the gts and annotations.json we provided. You can download our imgs.tar.gz or using the original sample files of the nuScenes dataset.

  2. Download the CAN bus expansion data and maps HERE.

  3. Organize your folder structure as below：

Occupancy3D
├── projects/
├── tools/
├── ckpts/
│   ├── r101_dcn_fcos3d_pretrain.pth
├── data/
│   ├── can_bus/
│   ├── occ3d-nus/
│   │   ├── maps/
│   │   ├── samples/     # You can download our imgs.tar.gz or using the original sample files of the nuScenes dataset
│   │   ├── v1.0-trainval/
│   │   ├── gts/
│   │   │── annotations.json
  1. Generate the info files for training and validation:
python tools/create_data.py occ --root-path ./data/occ3d-nus --out-dir ./data/occ3d-nus --extra-tag occ --version v1.0-trainval --canbus ./data --occ-path ./data/occ3d-nus

Training

./tools/dist_train.sh projects/configs/bevformer/bevformer_base_occ.py 8

Testing

./tools/dist_test.sh projects/configs/bevformer/bevformer_base_occ.py work_dirs/bevformer_base_occ/epoch_24.pth 8

You can evaluate the F-score at the same time by adding --eval_fscore.

Performance

model name weight mIoU others barrier bicycle bus car construction_vehicle motorcycle pedestrian traffic_cone trailer truck driveable_surface other_flat sidewalk terrain manmade vegetation
bevformer_base_occ Google Drive 23.67 5.03 38.79 9.98 34.41 41.09 13.24 16.50 18.15 17.83 18.66 27.7 48.95 27.73 29.08 25.38 15.41 14.46