-
 
No version for distro humble. Known supported distros are highlighted in the buttons above.
No version for distro iron. Known supported distros are highlighted in the buttons above.
No version for distro jazzy. Known supported distros are highlighted in the buttons above.
No version for distro rolling. Known supported distros are highlighted in the buttons above.

Package Summary

Tags No category tags.
Version 0.13.1
License GPLv2
Build type CATKIN
Use RECOMMENDED

Repository Summary

Checkout URI https://github.com/lagadic/vision_visp.git
VCS Type git
VCS Version noetic-devel
Last Updated 2024-08-23
Dev Status MAINTAINED
CI status Continuous Integration : 0 / 0
Released RELEASED
Tags No category tags.
Contributing Help Wanted (0)
Good First Issues (0)
Pull Requests to Review (0)

Package Description

Online automated pattern-based object tracker relying on visual servoing. visp_auto_tracker wraps model-based trackers provided by ViSP visual servoing library into a ROS package. The tracked object should have a QRcode, Flash code, or April tag pattern. Based on the pattern, the object is automaticaly detected. The detection allows then to initialise the model-based trackers. When lost of tracking achieves a new detection is performed that will be used to re-initialize the tracker. This computer vision algorithm computes the pose (i.e. position and orientation) of an object in an image. It is fast enough to allow object online tracking using a camera.

Additional Links

Maintainers

  • Fabien Spindler

Authors

  • Filip Novotny

visp_auto_tracker

visp_auto_tracker wraps model-based trackers provided by ViSP visual servoing library into a ROS package. The tracked object should have a QRcode, Flash code or April tag pattern. Based on the pattern, the object is automaticaly detected. The detection allows then to initialise the model-based trackers. When lost of tracking achieves a new detection is performed that will be used to re-initialize the tracker.

This computer vision algorithm computes the pose (i.e. position and orientation) of an object in an image. It is fast enough to allow object online tracking using a camera.

This package is composed of one node called ‘visp_auto_tracker’. The node tries first to detect the QRcode, the Flash or the April tag code associated to the object. Once the detection is performed, the node tracks the object. When a lost of tracking occurs the node tries to detect once again the object and then restart a tracking.

The viewer comming with visp_tracker package can be used to monitor the tracking result.

Setup

This package contains submodules. It can be compiled like any other ROS package using catkin_make.

Prerequisities

visp_auto_tracker depends on visp_bridge and visp_tracker packages available from https://github.com/lagadic/vision_visp (indigo-devel branches).

visp_auto_tracker depends also on libdmtx-dev and libzbar-dev system dependencies. To install them run:

$ sudo apt-get install libdmtx-dev libzbar-dev

How to get and build visp_tracker

Supposed you have a catkin work space just run:

$ cd ~/catkin_ws/src 
$ git clone -b indigo-devel https://github.com/lagadic/vision_visp.git
$ cd ..
$ catkin_make --pkg visp_auto_tracker

Documentation

The documentation is available on the project ROS homepage.

For more information, refer to the ROS tutorial.

CHANGELOG

Changelog for package visp_auto_tracker

0.13.1 (2022-02-20)

0.13.0 (2022-02-16)

0.12.1 (2021-04-22)

  • CMake fixes
  • Contributors: Fabien Spindler

0.12.0 (2021-04-21)

  • Fix warnings detected by catkin_lint
  • Fix warning detected during packaging
  • Update README with Noetic support
  • Use python3 on neotic
  • Fix compat with Ubuntu 20.04
  • Moves tracker_ref_frame from extern to param (#97)
  • Contributors: Fabien Spindler

0.11.0 (2019-05-27)

  • Update with Melodic
  • Fix memory leak
  • Introduce vpMbGenericTracker and visp3 headers in visp_auto_tracker
  • Introduce visp3 headers
  • Code indentation
  • Introduce vpMbGenericTracker
  • Introduce vpHandEyeCalibration class
  • Fix build for melodic (Closes #88)
  • Fix compilation on 17.04 by adding missing boost/format.hpp inclusion
  • Contributors: Fabien Spindler

0.10.0 (2017-02-10)

  • Fix catkin_lint warnings level 2
  • kinetic-0.9.3
  • Prepare changelogs
  • Contributors: Fabien Spindler

0.7.0 (2014-03-12)

  • Add flashcode_mbt as subdirectory.
  • Merge visp_auto_tracker as our subdirectory
  • Contributors: Thomas Moulard

0.7.2 (2014-04-07)

  • Fix various dependency issues in the CMakeLists.txt and package.xml files
  • lipstick

    Aesthetic changes

  • [visp_auto_tracker] Add missing dependency
  • Add missing dependency to ViSP
  • Fix errors detected with catkin_lint
  • Contributors: Fabien Spindler, Thomas Moulard

0.7.3 (2014-04-10)

  • Fix to install models folder
  • indigo-0.7.2
  • Prepare changelogs
  • Contributors: Fabien Spindler

0.8.0 (2015-04-01)

  • Remove catkin_lint issues and warnings
  • Fix doc url location
  • Update launch files
  • Comment viewer node. Only interesting when visp_auto_tracker debug_display param is False
  • Improve pose computation using Lagrange and Dementhon methods as initialisation
  • Increase quality value to avoid keypoint detection on the black part of the target
  • Introduce code_message parameter as a comment to show how to use.
  • New parameter "code_message" to specify the target to track from the code message. If this parameter is not set, the code that is tracked is the largest in the image.
  • Add frame_size parameter to the tracker client and viewer nodes that allow to specify the lenght (in meter) of the frame axis
  • Publish the code message of the tracked target on /visp_auto_tracker/code_message topic. If no target is tracked, publish an empty message. Remove fps printing
  • Fix to publish covariance matrix associated to the pose estimation
  • Fix to publish an empty list of moving edges and klt points on their respective topics when the tracker fails
  • Makes sure klt points that are sent via ros are from visible polygons.
  • Show how to use the viewer with visp_auto_tracker
  • Consider also .cao file to describe the cad model of the object to track with the hybrid model-based tracker
  • Remove useless include
  • Fix issue https://github.com/lagadic/vision_visp/issues/45 to take into account model_name parameter set in the launch file.
  • Changes done to use flash code detectors introduced in ViSP 2.10.0 rather than the one in visp_auto_tracker/flashcode_mbt/detectors.
  • Fix to avoid an OpenCV exception when the object roi is outside the image
  • Fix to run the viewer node
  • Fix to publish camera_info with the images
  • Fix compat with ViSP 2.10.0
  • Update and fix content of README files
  • Download the tutorial-qrcode.bag bag file from a new location that allows the download without SSL certificate
  • Install visp_auto_tracker/models folder
  • 0.7.3
  • Prepare changelogs
  • Prepare changelogs
  • Contributors: Aurelien Yol, Fabien Spindler

0.9.0 (2015-12-20)

  • Fix catkin_lint error and issues
  • Compat with ViSP 3.0.0
  • Fix to build with ViSP 2.10.0 when VISP_BUILD_DEPRECATED=OFF
  • jade-0.8.0
  • Prepare changelogs
  • Contributors: Aurelien Yol, Fabien Spindler

0.9.1 (2015-12-21)

  • Revert build_depend visp removal that is mandatory.
  • jade-0.9.0
  • Prepare changelogs
  • Contributors: Fabien Spindler

0.9.3 (2016-05-20)

  • Fix CMAKE_CXX_FLAGS as separated list
  • kinetic-0.9.2
  • Contributors: Fabien Spindler

Wiki Tutorials

This package does not provide any links to tutorials in it's rosindex metadata. You can check on the ROS Wiki Tutorials page for the package.

Launch files

  • launch/tracklive_firewire.launch
    • -*- xml -*- This tutorial relies on a live video sequence acquired with a firewire camera in front of a QR code planar target. The model corresponding to this target is given into the models directory of this package. Camera parameters are read from models/calibration.ini file. They need to be the one of your camera. Here we use the viewer coming with visp_tracker package to display the pose estimation results. See http://www.ros.org/wiki/visp_auto_tracker for more information.
  • launch/tracklive_usb.launch
    • -*- xml -*- This tutorial relies on a live video sequence acquired with an usb camera in front of a QR code planar target. The model corresponding to this target is given into the models directory of this package. Camera parameters are set as rosparam parameters. They need to be changed to be the one of your camera. See http://www.ros.org/wiki/visp_auto_tracker for more information.
  • launch/tutorial.launch
    • -*- xml -*- This tutorial relies on a recorded video sequence where the camera is fixed in front of QR code planar target. The model corresponding to this target is given into the models directory of this package. See http://www.ros.org/wiki/visp_auto_tracker for more information.

Messages

No message files found.

Services

No service files found

Plugins

No plugins found.

Recent questions tagged visp_auto_tracker at Robotics Stack Exchange

No version for distro galactic. Known supported distros are highlighted in the buttons above.

Package Summary

Tags No category tags.
Version 0.13.0
License GPLv2
Build type CATKIN
Use RECOMMENDED

Repository Summary

Checkout URI https://github.com/lagadic/vision_visp.git
VCS Type git
VCS Version melodic-devel
Last Updated 2022-02-21
Dev Status MAINTAINED
CI status Continuous Integration : 0 / 0
Released RELEASED
Tags No category tags.
Contributing Help Wanted (0)
Good First Issues (0)
Pull Requests to Review (0)

Package Description

Online automated pattern-based object tracker relying on visual servoing. visp_auto_tracker wraps model-based trackers provided by ViSP visual servoing library into a ROS package. The tracked object should have a QRcode, Flash code, or April tag pattern. Based on the pattern, the object is automaticaly detected. The detection allows then to initialise the model-based trackers. When lost of tracking achieves a new detection is performed that will be used to re-initialize the tracker. This computer vision algorithm computes the pose (i.e. position and orientation) of an object in an image. It is fast enough to allow object online tracking using a camera.

Additional Links

Maintainers

  • Fabien Spindler

Authors

  • Filip Novotny

visp_auto_tracker

visp_auto_tracker wraps model-based trackers provided by ViSP visual servoing library into a ROS package. The tracked object should have a QRcode, Flash code or April tag pattern. Based on the pattern, the object is automaticaly detected. The detection allows then to initialise the model-based trackers. When lost of tracking achieves a new detection is performed that will be used to re-initialize the tracker.

This computer vision algorithm computes the pose (i.e. position and orientation) of an object in an image. It is fast enough to allow object online tracking using a camera.

This package is composed of one node called ‘visp_auto_tracker’. The node tries first to detect the QRcode, the Flash or the April tag code associated to the object. Once the detection is performed, the node tracks the object. When a lost of tracking occurs the node tries to detect once again the object and then restart a tracking.

The viewer comming with visp_tracker package can be used to monitor the tracking result.

Setup

This package contains submodules. It can be compiled like any other ROS package using catkin_make.

Prerequisities

visp_auto_tracker depends on visp_bridge and visp_tracker packages available from https://github.com/lagadic/vision_visp (indigo-devel branches).

visp_auto_tracker depends also on libdmtx-dev and libzbar-dev system dependencies. To install them run:

$ sudo apt-get install libdmtx-dev libzbar-dev

How to get and build visp_tracker

Supposed you have a catkin work space just run:

$ cd ~/catkin_ws/src 
$ git clone -b indigo-devel https://github.com/lagadic/vision_visp.git
$ cd ..
$ catkin_make --pkg visp_auto_tracker

Documentation

The documentation is available on the project ROS homepage.

For more information, refer to the ROS tutorial.

CHANGELOG

Changelog for package visp_auto_tracker

0.13.0 (2022-02-21)

0.12.0 (2021-03-11)

0.11.1 (2019-07-04)

0.11.0 (2019-05-27)

  • Update with Melodic
  • Fix memory leak
  • Introduce vpMbGenericTracker and visp3 headers in visp_auto_tracker
  • Introduce visp3 headers
  • Code indentation
  • Introduce vpMbGenericTracker
  • Introduce vpHandEyeCalibration class
  • Fix build for melodic (Closes #88)
  • Fix compilation on 17.04 by adding missing boost/format.hpp inclusion
  • Contributors: Fabien Spindler

0.10.0 (2017-02-10)

  • Fix catkin_lint warnings level 2
  • kinetic-0.9.3
  • Prepare changelogs
  • Contributors: Fabien Spindler

0.7.0 (2014-03-12)

  • Add flashcode_mbt as subdirectory.
  • Merge visp_auto_tracker as our subdirectory
  • Contributors: Thomas Moulard

0.7.2 (2014-04-07)

  • Fix various dependency issues in the CMakeLists.txt and package.xml files
  • lipstick

    Aesthetic changes

  • [visp_auto_tracker] Add missing dependency
  • Add missing dependency to ViSP
  • Fix errors detected with catkin_lint
  • Contributors: Fabien Spindler, Thomas Moulard

0.7.3 (2014-04-10)

  • Fix to install models folder
  • indigo-0.7.2
  • Prepare changelogs
  • Contributors: Fabien Spindler

0.8.0 (2015-04-01)

  • Remove catkin_lint issues and warnings
  • Fix doc url location
  • Update launch files
  • Comment viewer node. Only interesting when visp_auto_tracker debug_display param is False
  • Improve pose computation using Lagrange and Dementhon methods as initialisation
  • Increase quality value to avoid keypoint detection on the black part of the target
  • Introduce code_message parameter as a comment to show how to use.
  • New parameter "code_message" to specify the target to track from the code message. If this parameter is not set, the code that is tracked is the largest in the image.
  • Add frame_size parameter to the tracker client and viewer nodes that allow to specify the lenght (in meter) of the frame axis
  • Publish the code message of the tracked target on /visp_auto_tracker/code_message topic. If no target is tracked, publish an empty message. Remove fps printing
  • Fix to publish covariance matrix associated to the pose estimation
  • Fix to publish an empty list of moving edges and klt points on their respective topics when the tracker fails
  • Makes sure klt points that are sent via ros are from visible polygons.
  • Show how to use the viewer with visp_auto_tracker
  • Consider also .cao file to describe the cad model of the object to track with the hybrid model-based tracker
  • Remove useless include
  • Fix issue https://github.com/lagadic/vision_visp/issues/45 to take into account model_name parameter set in the launch file.
  • Changes done to use flash code detectors introduced in ViSP 2.10.0 rather than the one in visp_auto_tracker/flashcode_mbt/detectors.
  • Fix to avoid an OpenCV exception when the object roi is outside the image
  • Fix to run the viewer node
  • Fix to publish camera_info with the images
  • Fix compat with ViSP 2.10.0
  • Update and fix content of README files
  • Download the tutorial-qrcode.bag bag file from a new location that allows the download without SSL certificate
  • Install visp_auto_tracker/models folder
  • 0.7.3
  • Prepare changelogs
  • Prepare changelogs
  • Contributors: Aurelien Yol, Fabien Spindler

0.9.0 (2015-12-20)

  • Fix catkin_lint error and issues
  • Compat with ViSP 3.0.0
  • Fix to build with ViSP 2.10.0 when VISP_BUILD_DEPRECATED=OFF
  • jade-0.8.0
  • Prepare changelogs
  • Contributors: Aurelien Yol, Fabien Spindler

0.9.1 (2015-12-21)

  • Revert build_depend visp removal that is mandatory.
  • jade-0.9.0
  • Prepare changelogs
  • Contributors: Fabien Spindler

0.9.3 (2016-05-20)

  • Fix CMAKE_CXX_FLAGS as separated list
  • kinetic-0.9.2
  • Contributors: Fabien Spindler

Wiki Tutorials

This package does not provide any links to tutorials in it's rosindex metadata. You can check on the ROS Wiki Tutorials page for the package.

Launch files

  • launch/tracklive_firewire.launch
    • -*- xml -*- This tutorial relies on a live video sequence acquired with a firewire camera in front of a QR code planar target. The model corresponding to this target is given into the models directory of this package. Camera parameters are read from models/calibration.ini file. They need to be the one of your camera. Here we use the viewer coming with visp_tracker package to display the pose estimation results. See http://www.ros.org/wiki/visp_auto_tracker for more information.
  • launch/tracklive_usb.launch
    • -*- xml -*- This tutorial relies on a live video sequence acquired with an usb camera in front of a QR code planar target. The model corresponding to this target is given into the models directory of this package. Camera parameters are set as rosparam parameters. They need to be changed to be the one of your camera. See http://www.ros.org/wiki/visp_auto_tracker for more information.
  • launch/tutorial.launch
    • -*- xml -*- This tutorial relies on a recorded video sequence where the camera is fixed in front of QR code planar target. The model corresponding to this target is given into the models directory of this package. See http://www.ros.org/wiki/visp_auto_tracker for more information.

Messages

No message files found.

Services

No service files found

Plugins

No plugins found.

Recent questions tagged visp_auto_tracker at Robotics Stack Exchange