This repo contains the RGB-D sequences used for the paper "Sparse-Dense Motion Modelling and Tracking for Manipulation without Prior Object Models" in the ROS 1 bag format.
The bag files contain the colour images with intrinsics, the depth images registered to the colour frame and ground truth camera poses:
/depth_to_rgb/camera_info : sensor_msgs/CameraInfo
/depth_to_rgb/image_raw/compressed : sensor_msgs/CompressedImage
/depth_to_rgb/image_raw/filtered/compressed : sensor_msgs/CompressedImage
/diagnostics : diagnostic_msgs/DiagnosticArray
/rgb/camera_info : sensor_msgs/CameraInfo
/rgb/image_raw/compressed : sensor_msgs/CompressedImage
/tf : tf2_msgs/TFMessage
/tf_static : tf2_msgs/TFMessage
/vicon/kinect/kinect : geometry_msgs/TransformStamped
/vicon/markers : vicon_bridge/Markers
If you use this data, please cite:
@ARTICLE{Rauch2022,
author={Rauch, Christian and Long, Ran and Ivan, Vladimir and Vijayakumar, Sethu},
journal={IEEE Robotics and Automation Letters},
title={Sparse-Dense Motion Modelling and Tracking for Manipulation Without Prior Object Models},
year={2022},
volume={7},
number={4},
pages={11394-11401},
doi={10.1109/LRA.2022.3200177}
}Mirrors:
- https://conferences.inf.ed.ac.uk/MultiMotionFusion
- https://github.com/christian-rauch/MultiMotionFusion-data
- https://gitlab.com/christian-rauch/MultiMotionFusion-data
Use the ROS 1 rosbag command-line tool to publish the topics from the bag files:
rosbag play estimation/nx_estim1_manipulation.bagInstall rosbags:
pip3 install rosbagsand convert the bag files from ROS 1 to ROS 2:
find . -type f -name *.bag -exec rosbags-convert "{}" \;Then you can use the ROS 2 command-line tool to publish the topics:
ros2 bag play estimation/nx_estim1_manipulation