ROS index logo ROS Index
  1. Home
  2. Packages
  3. face_cropping
 
No version for distro humble. 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.

face_cropping package from opentera-webrtc-ros repo

face_cropping map_image_generator opentera_client_ros opentera_protobuf_messages opentera_webrtc_demos opentera_webrtc_robot_gui opentera_webrtc_ros opentera_webrtc_ros_msgs turtlebot3_beam_description

API Docs
Browse Code

Package Summary

Tags No category tags.
Version 0.0.0
License Apache-2.0
Build type AMENT_CMAKE
Use RECOMMENDED

Repository Summary

Checkout URI https://github.com/introlab/opentera-webrtc-ros.git
VCS Type git
VCS Version ros2
Last Updated 2025-02-14
Dev Status UNMAINTAINED
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

The face_cropping package

Additional Links

No additional links.

Maintainers

  • Dominic Létourneau

Authors

No additional authors.
face_cropping/README.md

face_cropping

Face Detection Models

  • haarcascade_frontalface_default.xml Haar feature-based cascade classifiers to detect faces. The model is from the OpenCV repository and the license is at the beginning of the file.
  • lbpcascade_frontalface_improved.xml LBP feature-based cascade classifiers to detect faces. The model is from the OpenCV repository and the license is at the beginning of the file.
  • small_yunet_.pt The models are derived from YuNet, but it is faster since it is optimized for big faces. Their backbones are trained on ImageNet using knowledge distillation. The face detectors are trained on WIDER FACE without the small faces. The training code is available in the dnn_training folder. SimOTA is used for the face assignation. The SimOTA code is from (MMDetection)[https://github.com/open-mmlab/mmdetection].

LibTorch

LibTorch and torchvision are required to use small_yunet_* models. LibTorch and torchvision are automatically downloaded on AMD64 computers. On Jetson computer, you need to install PyTorch built by NVIDIA and compile torchvision for C++.

To use the GPU, you need to set the option FACE_CROPPER_USE_CUDA to ON.

Resource usage

The CPU and memory usage are measured on an AMD RYZEN 7 3700X at 30 Hz. The APs are measured on a subset of WIDER FACE. The subset includes faces occupying at least 10% of the width and the height of the image.

Model LibTorch Required CPU (%, 100% = 1 core) Memory (MB) AP@0.25 AP@0.50 AP@0.75
haarcascade No 220.4 299.7 0.5973 0.5649 0.0314
lbpcascade No 154.7 299.1 0.4141 0.4014 0.0705
small_yunet_0.25_160 Yes 11.4 273.9 0.6896 0.4940 0.1298
small_yunet_0.25_320 Yes 16.7 276.4 0.7969 0.7019 0.2669
small_yunet_0.25_640 Yes 34.3 297.4 0.8266 0.7601 0.4237
small_yunet_0.5_160 Yes 12.7 274.4 0.7930 0.6834 0.3183
small_yunet_0.5_320 Yes 20.2 282.2 0.8622 0.8034 0.4840
small_yunet_0.5_640 Yes 50.1 315.8 0.8780 0.8466 0.6018

face_cropping_node

Implements a ROS node that subscribes to an image stream. It detects the biggest face in the received frame and crops it out.

It can be enabled and disabled from the robot GUI by pressing the face cropping button, by default it is enabled.

Parameters

  • face_detection_model (string): The face detection model (haarcascade, lbpcascade, small_yunet_0.25_160, small_yunet_0.25_320, small_yunet_0.25_640, small_yunet_0.5_160, small_yunet_0.5_320 or small_yunet_0.5_640).
  • use_gpu_if_available (bool): Indicates whether to use the GPU or not.
  • min_face_width (double): The minimum face width.
  • min_face_height (double): The minimum face height.
  • output_width (int): The output face image width.
  • output_height (int): The output face image height.
  • adjust_brightness (bool): Indicates whether to adjust the brightness or not.

Subscribed Topics

Published Topics

CHANGELOG
No CHANGELOG found.

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.

Package Dependencies

Deps Name
ament_cmake
ament_lint_auto
ament_lint_common
rclcpp
rclpy
sensor_msgs
opentera_webrtc_ros_msgs
cv_bridge
image_transport
ament_index_cpp

System Dependencies

No direct system dependencies.

Dependant Packages

No known dependants.

Launch files

  • launch/face_cropping.launch.xml
      • name [default: face_cropping_node]
      • input_camera_topic [default: /camera1/image_raw]
      • output_camera_topic [default: /face_camera/image_raw]
      • face_detection_model [default: lbpcascade]
      • use_gpu_if_available [default: false]
      • min_face_width [default: 100.0]
      • min_face_height [default: 100.0]
      • output_width [default: 480]
      • output_height [default: 720]
      • adjust_brightness [default: true]
  • launch/test_face_cropping.launch.xml
      • model [default: lbpcascade]
  • launch/face_cropping_webrtc.launch.xml
      • name [default: face_cropping_webrtc_node]
      • input_camera_topic [default: /camera1/image_raw]
      • output_camera_topic [default: /face_camera/image_raw]
      • face_detection_model [default: lbpcascade]
      • use_gpu_if_available [default: false]
      • min_face_width [default: 100.0]
      • min_face_height [default: 100.0]
      • output_width [default: 480]
      • output_height [default: 720]
      • adjust_brightness [default: true]
  • launch/test_face_cropping_webrtc.launch.xml
      • model [default: lbpcascade]

Messages

No message files found.

Services

No service files found

Plugins

No plugins found.

Recent questions tagged face_cropping at Robotics Stack Exchange

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 iron. 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.