-
 

hri_face_detect package from hri_face_detect repo

hri_face_detect

Package Summary

Tags No category tags.
Version 2.0.10
License Apache License 2.0
Build type AMENT_CMAKE
Use RECOMMENDED

Repository Summary

Checkout URI https://github.com/ros4hri/hri_face_detect.git
VCS Type git
VCS Version humble-devel
Last Updated 2024-10-11
Dev Status DEVELOPED
CI status No Continuous Integration
Released UNRELEASED
Tags No category tags.
Contributing Help Wanted (0)
Good First Issues (0)
Pull Requests to Review (0)

Package Description

Google Mediapipe-based face detection for ROS4HRI

Additional Links

No additional links.

Maintainers

  • Séverin Lemaignan
  • Luka Juricic

Authors

  • Séverin Lemaignan
  • Luka Juricic

hri_face_detect

A ROS4HRI-compiant ROS node to perform fast face detection using YuNet face detector and Mediapipe Face Mesh. The former performs well at greater distances (depending on image resolution and image scaling applied) and extracts 5 keypoints. The latter works only at close distances and extracts all the ROS4HRI-defined landmarks.

Installing dependencies

If you are running this package outside PAL, first install all the dependencies:

pip install -r requirements.txt

ROS API

Parameters

All parameters are loaded in the lifecycle configuration transition.

  • processing_rate (int, default: 30): Image processing logic execution rate in Hertz.

  • face_mesh (bool, default: true): It enables the additional Mediapipe Face Mesh detection.

  • confidence_threshold (double, default: 0.75): Candidate face detections with confidence lower that this threshold are not published.

  • image_scale (double, default: 0.5): The YuNet face detector accepts input image of dynamic size. This parameter controls the rescale factor applied to the input image before running the YuNet face detector. Lower image scale results in less processing time required and lower detection confidences. The output data (e.g., RoI) is invariant with this parameter and always refers to the original input image size.

  • filtering_frame (string, default: “camera_color_optical_frame”): The reference frame the estimated face pose should be transformed to before performing the filtering operations. Due to the proximity between the camera frame and the detected faces, and considering that cameras can be mounted on frequently moving robot’s components (e.g., robot’s head), directly filtering a face pose expressed in camera optical frame might reduce the filtering quality.

  • deterministic_ids (bool, default: false): If true the face ids start from “f00000” and increases by one for each new face. If false it is a random five letters sequence.

  • debug (bool, default: false): If true opens a windows showing the input image with face detections overlayed.

Topics

This package follows the ROS4HRI conventions (REP-155). If the topic message type is not indicated, the ROS4HRI convention is implied.

Subscribed

Published

  • /humans/faces/<faceID>/roi
  • /humans/faces/<faceID>/landmarks
  • /humans/faces/<faceID>/cropped
  • /humans/faces/<faceID>/aligned
  • /humans/faces/tracked
  • /diagnostics (diagnostic_msgs/msg/DiagnosticArray)

Execution

ros2 launch hri_face_detect face_detect.launch rgb_camera:=<input camera namespace>

Example

For an example of usage, execute in different terminals:

  • USB camera:
    1. apt install ros-humble-usb-cam
    2. ros2 run usb_cam usb_cam_node_exe
  • HRI face detect:
    1. apt install ros-humble-hri-face-detect
    2. ros2 launch hri_face_detect face_detect.launch.py
  • RViz with HRI plugin:
    1. apt install ros-humble-rviz2
    2. apt install ros-humble-hri-rviz
    3. rviz2

In RViz, add the ‘Humans’ plugin to see the detected faces with the relative keypoints.

CHANGELOG

Changelog for package hri_face_detect

2.0.10 (2024-09-16)

  • use python3-mediapipe-pip on non-PAL environments While here, add missing dep on cv_bridge
  • Contributors: Séverin Lemaignan

2.0.9 (2024-09-09)

  • fixed quality of service camera-related subscribers
  • Contributors: lorenzoferrini

2.0.8 (2024-08-19)

  • rename diagnostics msg to match documentation (and diagnostic_aggregator) categories
  • Contributors: Séverin Lemaignan

2.0.7 (2024-07-04)

  • linting
  • Contributors: Séverin Lemaignan

2.0.6 (2024-07-04)

  • launch: use get_pal_configuration from launch_pal
  • expose the 'deterministic_ids' in the _with_args launch file
  • Contributors: Séverin Lemaignan

2.0.5 (2024-05-24)

  • Removed quotes from topic values
  • Contributors: Raquel Ros

2.0.4 (2024-05-23)

  • fix configuration node name
  • fix get _pal_configuration
  • Contributors: Luka Juricic

2.0.3 (2024-05-08)

  • [launch] impl logic to load overlaid parameters This commit:
    • add (and install) default parameters values in config/00-defaults.yml
    • rename (unchanged) launch/face_detect.launch.py to launch/face_detect_with_args.launch.py

    - add a new launch/face_detect.launch.py that implements the PAPS-007 logic to fetch possible parameters, remappings and node arguments via ament_index

  • Contributors: Séverin Lemaignan

2.0.2 (2024-04-24)

  • add pal module
  • Contributors: Luka Juricic

2.0.1 (2024-02-06)

  • fix image calibration K name to lowercase k. This caused the publication of TF frames to be broken.
  • assign default score detection in case of mesh detection
  • add usage example to README
  • Contributors: Luka Juricic

2.0.0 (2024-01-18)

  • port to ROS 2 Humble
  • change license to apache2
  • change folder structure
  • Contributors: Luka Juricic

1.5.3 (2023-11-27)

  • rework filtering frame validation Now, does not pre-validate the filtering frame: simply try to transform to the filtering frame if it is provided, and continue without using filtering frame if it is not available. Helps in the case hri_face_detect starts before the robot's TF is fully published -> the filtering will start in the correct frame as soon as it become available.
  • Contributors: Séverin Lemaignan

1.5.2 (2023-10-27)

  • port facedetection external cmake to project one
  • add tkinter dependency
  • Contributors: Luka Juricic

1.5.1 (2023-10-24)

  • fix library external project dependency
  • Contributors: Luka Juricic

1.5.0 (2023-10-18)

  • change detector to Yunet
    • large refactor
    • remove record node
    • use timer based logic to process the most recent image only
    • add Yunet detector as standalone C library
    • add Yunet python bindings
    • use Yunet as always on detector
    • use Mediapipe face mesh detector to refine near faces
    • update documentation
  • Contributors: Luka Juricic

1.4.9 (2023-07-05)

  • change RoI message type to hri_msgs/NormalizedRegionOfInterest2D
  • Contributors: Luka Juricic

1.4.8 (2023-06-22)

  • added filtering_frame parameter the user can now decide which frame to use to filter the faces position
  • Contributors: lorenzoferrini

1.4.7 (2023-05-23)

  • face pose filtering using the one-euro filter
  • Contributors: lorenzoferrini

1.4.6 (2023-05-12)

  • add diagnostics
  • Migrate to new python3-mediapipe rosdep key
  • Contributors: Séverin Lemaignan, lukajuricic, mathiasluedtke

1.4.5 (2023-03-08)

  • ensure mediapipe is not called from 2 threads in parallel This was causing mediapipe internal timestamp issues
  • Contributors: Séverin Lemaignan

1.4.4 (2022-10-06)

  • fix FacialLandmark object initialisation When face_mesh=False, the arguments for the FacialLandmarks objects initialisation were not correctly disposed, as the first element in a FacialLandmarks message is supposed to be a Header.
  • Contributors: lorenzoferrini

1.4.3 (2022-08-31)

  • more update to hri_msgs-0.8.0
  • Contributors: Séverin Lemaignan

1.4.2 (2022-08-31)

  • update to hri_msgs-0.8.0
  • Contributors: Séverin Lemaignan

1.4.1 (2022-08-02)

  • ensure face id are strings starting with a letter
  • [cosmetic] code formatting
  • pep8 code formatting
  • add tool to record faces
  • Contributors: Séverin Lemaignan

1.4.0 (2022-04-29)

  • large refactor of the code In particular:
    • added a Face class to maintain the state of a detected face
    • reworked how detection results are returned, to simplify code
  • publish aligned versions of the face under /humans/faces/<id>/aligned (aligned faces are rotated such as the eyes are always horizontal)
  • warn about faces height and width having to be equal
  • store various face publishers as dict to ease future extension
  • Delegated face estimation process to function.
  • publish empty list of faces upon closing to clean up state
  • update launch file to match hri_fullbody arguments names
  • [doc] node suitable for production
  • Contributors: Séverin Lemaignan, lorenzoferrini

1.3.1 (2022-03-01)

  • Use tf frame from source image
  • Contributors: lorenzoferrini

1.3.0 (2022-03-01)

  • changing the frames name syntax from face<id> to face<id> and gaze<id> to gaze<id> for compliance with ROS4HRI spec
  • [minor] adding default value for camera topics in launch
  • Documentation update
  • Fixed the default number of detectable faces to 10
  • Facial Landmark msg implementation Fully implemented facial landmark msg publishing for both basic face detection and face mesh detection
  • Contributors: lorenzoferrini

1.2.0 (2022-02-14)

  • mediapipe Face-mesh based face detection It is now possible to decide between two Mediapipe different solutions for face detection: face_detection and face_mesh. Since the overall performance (taking into account cpu, memory and detection results) appears to be better in the latter case, face_mesh detection will be the default option.
  • add missing deps
  • [minor] launch file modified according to new features available It is now possible to specify the solution to use (face_detection/face_mesh) and the maximum number of faces detectable by the face_mesh model as launch file parameters
  • max_num_faces as initialization parameter for FaceDetector class
  • [WiP] Correcting face orientation and introducing gaze frame Face and gaze frame orientation according to ROS4HRI convention.
  • Facial landmarks publishing Now publishing the facial landmarks according to the ROS4HRI definition, on the topic /humans/faces/<body_id>/landmarks. Additionally, the face frame is published now as face_<body_id> and the debug code has been removed.
  • first rough implementation of PnP head pose estimation
  • Contributors: Séverin Lemaignan, lorenzoferrini

1.1.0 (2022-01-18)

  • publish cropped faces under subtopic /cropped
  • add _preallocate_topics parameter (instead of hard-coded constant)
  • code formatting
  • RegionOfInterestStamped -> regionOfInterest to match changes in hri_msgs 0.2.1
  • Contributors: Séverin Lemaignan

1.0.1 (2021-11-09)

  • Added the dependency on python-mediapipe
  • Publish an Empty msg on /hri_detect_face/ready when ready to start This is eg required for automated testing, to ensure the node is fully ready before publishing the first frames.
  • added minimal node setup
  • Added basic readme
  • Simple, rough node using Google Mediapipe to perform fast face detection
  • Contributors: Séverin Lemaignan

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

No launch files found

Messages

No message files found.

Services

No service files found

Plugins

No plugins found.

Recent questions tagged hri_face_detect at Robotics Stack Exchange

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.
No version for distro noetic. Known supported distros are highlighted in the buttons above.
No version for distro ardent. Known supported distros are highlighted in the buttons above.
No version for distro bouncy. Known supported distros are highlighted in the buttons above.
No version for distro crystal. Known supported distros are highlighted in the buttons above.
No version for distro eloquent. Known supported distros are highlighted in the buttons above.
No version for distro dashing. Known supported distros are highlighted in the buttons above.
No version for distro galactic. Known supported distros are highlighted in the buttons above.
No version for distro foxy. Known supported distros are highlighted in the buttons above.
No version for distro lunar. Known supported distros are highlighted in the buttons above.
No version for distro jade. Known supported distros are highlighted in the buttons above.
No version for distro indigo. Known supported distros are highlighted in the buttons above.
No version for distro hydro. Known supported distros are highlighted in the buttons above.
No version for distro kinetic. Known supported distros are highlighted in the buttons above.
No version for distro melodic. Known supported distros are highlighted in the buttons above.