Shamma's profile - activity

The Nao's cameras work just like any other monocular camera. But what do you mean with "calculate real world coordinate"? Without any additional information (like a ground plane) you cannot derive 3D coordinates from 2D image coordinates. For that you need stereo vision which the Nao's camera's can't provide. Monocular SLAM (visual odometry, structure from motion, ...) could be a solution for that, but is a research topic of its own.

You can still calibrate your Nao's cameras just like any other webcam, it only lacks a proper ROS-enabled driver that outputs the correct topics for the automatic calibration. I would suggest to record 10 static frames with the checkerboard in various positions and then feed them into the OpenCV camera calibration.

Rviz is a visualizer, if you are looking for a simulator in row, then you should look into gazebo, see the wiki here http://www.ros.org/wiki/simulator_gazebo. OpenRAVE also provides a good simulator and is well documented here with row support available as well.

It would help if you clarify your needs further.

here what I would like to do, and I thought camera calibration would do it for me:

1. I want to define a zero point in the environment or what they call it a reference point in which that point will have (0,0,0) values
2. I want to take some picture of the environment where I have objects set on some places
3. I want to measure the coordinates of the object in real life "I will do that manually with a meter"
4. I want to feed a method "I don't know what method" with these values so it will be able to compare between real life coordinate and pixel coordinate anytime it needs to assume the real life coordinate

any idea how can I do that ?

Regards,

Shamma

hello there,

I would like to use Camera calibration that is already implemented in ROS, to get the images of my robot Nao, and then use the information of the calibration to calculate the real world coordinate..

http://www.ros.org/wiki/camera_calibr...

I don't know how would I do it with the robot camera, they have example of Monocular Camera...

Any help?

EDIT

here what I would like to do, and I thought camera calibration would do it for me:

I want to define a zero point in the environment or what they call it a reference point in which that point will have (0,0,0) values I want to take some picture of the environment where I have objects set on some places I want to measure the coordinates of the object in real life "I will do that manually with a meter" I want to feed a method "I don't know what method" with these values so it will be able to compare between real life coordinate and pixel coordinate anytime it needs to assume the real life coordinate any idea how can I do that ?

Regards,

Shamma

So what I am tying to so basically is that I am developing an interface for my robot "Nao" and I want to add a simulator to my interface to simulate his movement to the user like the things provided in Rviz, the only problem in Rviz is that I couldn't find a way to integrate it into my code

I am going to check the things you posted and get back at you ...thank you :D

hello all,

I am developing a project in which I need to have a simulator of the robot for the user, so I would like to integrate the simulator display of Rviz in my interface,

is there anyway in ROS in which I can input the commands by commandline for example or which I an manipulate their code to get it inside my gui ?

Thank you

Yes, you can do it with opencv's drawing functions. Use cvbridge to create cv::Mat from sensor_msg::image without copying.

If you use pyqt anyway, why not use the drawing functions of qt? You can convert the images to QImage (e.g. as it is done here (c++) from an opencv image) and use a QPainter, constructed with the image as parameter to use drawEllipse.

hello there,

I am using ROS to develop a system in my research. So, I am retrieving video images from the robot and I want to draw a circle on the screen to the object that the robot recognizes. I have the code for the object recognition and the position of the object with respect to the video , I was thinking of using OpenCV, but I don't know if they are compatible with pyqt and python?

any help

thank you

You're free to use whatever software you like. Keep in mind that PyQt is GPL, which places restrictions on the software you can use with it. An alternative is PySide, which recently reached 1.0 status.

hello All,

I saw the threads that the people post about pyqt, and I though maybe my question will be different and I will get the accurate answer that I want..

So now I already created a GUI with pyqt which takes data from the robot's camera and display it on the screen. Now, I want to display some information about the robot on the screen as well and I want to control him... I found some code which is written in ROS and I just want to create a GUI for it to display the information on the screen and I am going to add some other stuff.. the question here "is it ok to you pyqt?" or shall I look up for another GUI library ??

thank you