Only released in EOL distros:
Package Summary
This package provides face detection.
- Maintainer status: developed
- Maintainer: Siddhartha Banerjee <banerjs.sid AT gmail DOT com>, Andrew Silva <andrew.silva AT gatech DOT edu>
- Author: Siddhartha Banerjee <banerjs.sid AT gmail DOT com>, Andrew Silva <andrew.silva AT gatech DOT edu>
- License: MIT
- Source: git https://github.com/GT-RAIL/rail_face_detection.git (branch: indigo-devel)
Contents
Two Minute Intro
This detector uses Caffe to perform face detection. It publishes faces found in images from a subscribed image topic. The face detector itself can be found here: https://github.com/kpzhang93/MTCNN_face_detection_alignment.
The message type coming back from the face detector is a Detections.msg which contains an array of Face.msg objects. Each Face.msg has:
top_left_x - X pixel value of top-left corner of bounding box
top_left_y - Y pixel value of top-left corner of bounding box
bot_right_x - X pixel value of bottom-right corner of bounding box
bot_right_y - Y pixel value of bottom-right corner of bounding box
nose_x - X coordinate of the nose
nose_y - Y coordinate of the nose
left_eye_x - X coordinate of the left eye
left_eye_y - Y coordinate of the left eye
right_eye_x - X coordinate of the right eye
right_eye_y - Y coordinate of the right eye
left_mouth_x - X coordinate of the left corner of the mouth
left_mouth_y - Y coordinate of the left corner of the mouth
right_mouth_x - X coordinate of the right corner of the mouth
right_mouth_y - Y coordinate of the right corner of the mouth
Nodes
This package contains a single ROS node - face_detector_node - which serves as an interface between a ROS system and the trained face recognition network.
face_detector_node
Topics
detector_node/faces
Type: face_detector/detections
Topic with face detections performed in the background by running on images as they come in the subscriber.
detector_node/debug/face_images
Type: face_detector/debug/face_images
Topic with face detections visualized on incoming images as they come in from the subscriber. Only published if debug=true.
Parameters
image_sub_topic_name
Type: string
Default: "/kinect/qhd/image_color_rect"
Image topic name to use for detections.
debug
Type: bool
Default: false
Enable or disable debug mode, which publishes incoming images with bounding boxes over faces
use_gpu
Type: bool
Default: true
Enable or disable gpu mode. Enabled by default. Significantly speeds up detection.
use_compressed_image
Type: bool
Default: false
Change to compressed image stream or not. Simply appends a "/compressed" to the image topic name. This lightens the load your local network if the images are being transmitted to the detector.
