This is not an officially supported Google product.
- Go to https://jrdb.erc.monash.edu/#downloads
- Create a User or login.
- Download and extract JRDB 2022 Full Train Dataset to
<data_path>/train_dataset
. - Download and extract JRDB 2022 Full Test Dataset to
<data_path>/test_dataset
. - Download and extract Train Detections from the JRDB 2019 section to
<data_path>/detections
.
Download and extract this leaderboard 3D tracking result to <data_path>/test_dataset/labels/PiFeNet/
. Such that you have <data_path>/test_dataset/labels/PiFeNet/00XX.txt
.
Download and extract this leaderboard 3D tracking result to <data_path>/test_dataset/labels/ss3d_mot/
. Such that you have <data_path>/test_dataset/labels/ss3d_mot/00XX.txt
. This was the best available leaderboard tracker at the time the method was developed.
Download the compressed Odometry data file here.
Extract the files and move them to <data_path>/processed/
such that you have <data_path>/processed/odoemtry/train
, <data_path>/processed/odoemtry/test
.
Alternatively you can extract the robot odometry from the raw rosbags yourself via extract_robot_odometry_from_rosbag.py
.
Download the compressed Keypoints data file here.
Extract the files and move them to <data_path>/processed/
such that you have <data_path>/processed/labels/labels_3d_keypoints/train/
, <data_path>/processed/labels/labels_3d_keypoints/test/
.
Run
python jrdb_train_detections_to_tracks.py --input_path=<data_path>
You should end up with a dataset folder of the following structure
- <data_path>
- train_dataset
- calibration
- detections
- images
- labels
- pointclouds
- test_dataset
- calibration
- images
- labels
- pointclouds
- processed
- labels
- labels_3d_keypoints
- train
- test
- labels_detections_3d
- odoemtry
- train
- test
python jrdb_preprocess_train.py --input_path=<data_path> --output_path=<output_path> --max_distance_to_robot=50.0
python jrdb_preprocess_test.py --input_path=<data_path> --output_path=<output_path> --max_distance_to_robot=50.0 --tracking_method=PiFeNet --tracking_confidence_threshold=0.01
Please note that this can take multiple hours due to the processing of the scene's
pointclouds. If you do not need the pointclouds you can speed up the processing
by passing --process_pointclouds=False
for both.
python jrdb_preprocess_train.py --input_path=<data_path> --output_path=<output_path> --max_distance_to_robot=15.0
python jrdb_preprocess_test.py --input_path=<data_path> --output_path=<output_path> --max_distance_to_robot=15.0 --tracking_method=ss3d_mot
Please note that this can take multiple hours due to the processing of the scene's
pointclouds. If you do not need the pointclouds you can speed up the processing
by passing --process_pointclouds=False
for both.