Simple C++ tool for converting the nuScenes dataset from Aptiv.
The tool loads the json metadata and then the sample files for each scene. The sample are converted in a suitable ROS msg and written to a bag. TF tree is also written.
Probably the original dataset is also collected by Aptiv using ROS, so most data has the same format.
Bounding boxes in Rviz:
The master branch should compile on Ubuntu 19.04. It requires C++17 and Boost 1.66.
The ubuntu_1604 branch uses C++11 and has been tested on Ubuntu 16.04.
The ubuntu_1804 branch has been tested on Ubuntu 18.04.
The tool is a normal ROS package. Place it under a workspace and build it with catkin.
Command-line arguments:
--dataroot: The path to the directory that contains the 'maps', 'samples' and 'sweeps'.
--version: (optional) The sub-directory that contains the metadata .json files. Default = "v1.0-mini"
Converting the 'mini' dataset:
Convert one scene to a bag file, saved in a new directory:
Scene '0061' will be saved to 'nuscenes_bags/61.bag'
rosrun nuscenes2bag nuscenes2bag --scene_number 0061 --dataroot /path/to/nuscenes_mini_meta_v1.0/ --out nuscenes_bags/
Convert the entire dataset to bag files:
This processes 4 scenes simultaneously, however the scene numbers are not processed in numerical order.
rosrun nuscenes2bag nuscenes2bag --dataroot /path/to/nuscenes_mini_meta_v1.0/ --out nuscenes_bags/ --jobs 4
Converting other datasets:
Convert a dataset with the metadata in a sub-directory called 'v2.0':
rosrun nuscenes2bag nuscenes2bag --dataroot /path/to/nuscenes_data_v2.0/ --version v2.0 --out nuscenes_bags/ --jobs 4
Currently work in progress
- Image support
- Pointcloud support
- Radar support
- EgoPose and TF support
- Show progress
- Use range library
- Better usability
Create an Github issue for suggestion, bug and requests.
Built using: