ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange
Ask Your Question
0

Face Recognition with kinect

asked 2014-04-23 10:27:58 -0500

smart engineer gravatar image

updated 2014-04-23 10:32:06 -0500

Hi , I am trying to make a face Recognition with kinect ... i have fuerte ROS working on ubuntu 12.04 ?? f

edit retag flag offensive close merge delete

2 Answers

Sort by ยป oldest newest most voted
0

answered 2014-04-23 12:56:41 -0500

Carlos Hdz gravatar image

updated 2014-04-23 12:58:48 -0500

Hello.

1) In order to do that, first check the "openni" tutorial. This tutorial demonstrates how to open a Kinect in ROS, introduces ROS tools for visualization and configuration, etc: http://wiki.ros.org/openni_launch/Tutorials/QuickStart

2) Then, also check how to use a bridge between the kinect image topic and Opencv. This tutorial describes how to interface ROS and OpenCV by converting ROS images into OpenCV images, and vice versa, using cv_bridge. Included is a sample node that can be used as a template for your own node: http://wiki.ros.org/cv_bridge/Tutorials/ConvertingBetweenROSImagesAndOpenCVImagesPython

3) Read the OpenCV documentation on how to recognize faces: http://docs.opencv.org/trunk/modules/contrib/doc/facerec/index.html http://docs.opencv.org/trunk/modules/contrib/doc/facerec/tutorial/facerec_video_recognition.html

3) Try some example that you can found on the web, like this (I haven't tried it): https://code.google.com/p/ros-by-example/source/browse/trunk/rbx_vol_1/rbx1_vision/nodes/face_detector.py?r=116

I hope this links can help you, good luck!

edit flag offensive delete link more
1

answered 2014-04-23 16:31:24 -0500

Pi Robot gravatar image

If you want to do true face recognition rather than simple face detection, and you want to use ROS, check out the most excellent cob_people_detection packages. These packages take full advantage of the depth information from the Kinect to first look for "head-like" blobs which are likely to contain faces. You can then train the system to recognize different faces by giving each face a name and taking a number of snapshots from different angles.

--patrick

edit flag offensive delete link more

Comments

when I try to execute cob_people_detection and try to know requirements for the package""" rosdep cob_people_perception-fuerte_dev"" the following requirements appear :: """ rosdep check <stacks-and-packages>... check if the dependencies of package(s) have been met. rosdep install <stacks-and-packages>... generate a bash script and then execute it. rosdep db generate the dependency database and print it to the console. rosdep init initialize rosdep sources in /etc/ros/rosdep. May require sudo. rosdep keys <stacks-and-packages>... list the rosdep keys that the packages depend on. rosdep resolve <rosdeps> resolve <rosdeps> to system dependencies rosdep update update the local rosdep database based on the rosdep sources. rosdep what-needs <rosdeps>... print a list of packages that declare a rosdep on (at least one of) <rosdeps> rosdep where-defined <rosdeps>... print a list of yaml files that declare a rosdep on (at least one of) <rosdeps> """???

smart engineer gravatar image smart engineer  ( 2014-04-28 04:06:00 -0500 )edit

Question Tools

1 follower

Stats

Asked: 2014-04-23 10:27:58 -0500

Seen: 2,385 times

Last updated: Apr 23 '14