Processing bag files

Preparing an image

Update submodules:

git submodule update --init --recursive

Prepare docker image:

docker build -t hd_maps -f Dockerfile.desktop .

Now run container (using PowerShell), please adjust D:/2022-08-19T12_49_37.772111 to your dataset. This directory will be visible as /data inside a container.

docker run -it --rm --mount type=bind,source="D:/2022-08-19T12_49_37.772111",target=/data hd_maps

Work in the container - export trajectory

Inside container process dataset to find odometry using FAST-LIO:

python src/rosbag_to_2000/launch.py --dir /data/ --rate 2

In your dataset directory D:/2022-08-19T12_49_37.772111 the file processed_lio.bag should appear.

Next, export data to binary format for further processing

rosrun rosbag_to_2000 rosbag_livox_to_dat --dir /data/

After last step the D:\2022-08-19T12_49_37.772111\data directory should appear with exported data.

Work in the container - export images

Inside container process dataset to export images.

python src/bag_to_images.py -i /data/log_*.bag -o /data/images

In your dataset directory D:/2022-08-19T12_49_37.772111 the directory images should appear.