Tracking an object and get it's coordinates
 [1 posts]
3 years
I am totally new to computer vision world.
I just finished a course in image proccessing in opencv and now moved to RR.
I am able now to do calibration, and apply lucaskanade algorithm to click on an object on screen and get it's X,Y using OpenCV.

1- Any Idea on how to achieve this on RR?
2- Does there is better algorithm for clicking an object and tracking it and get it's x,y?
Thanks in advanced
Steven Gentner from United States  [1370 posts] 3 years

Perhaps you can explain the context of your project a bit more. There are many ways to track an object (the Object Recognition module for example) but it sounds like you want a more interactive mode where you simply click on the image and track whatever is under the mouse at the time?. Once this object moves out of view, you'd click again on another object as it appears? Is this behavior correct?

The reason for my question is that traditionally we try to build a model of one or a few objects that are then tracked without any user interaction (most robots ideally function independent of human interaction) but your task sounds a bit different.

There may also be other modules that can help to isolate trackable portions of the image which can aid in the selection of good trackable objects.


This forum thread has been closed due to inactivity (more than 4 months) or number of replies (more than 50 messages). Please start a New Post and enter a new forum thread with the appropriate title.

 New Post   Forum Index