Ask Your Question

3D object Tracking

asked 2016-10-20 10:13:22 -0600

I have a task where i am asked to track parcels(carton boxes) of different dimensions moving on a conveyor

I am using Asus Xtion pro camera mounted on top of a conveyor in any inclined angle.r

I have read so many model-based methods where we need to create a model of the object to be detected and then we perform keypoints extraction,feature mapping and so many other concepts between the scene and the model. Since i am using Boxes of different dimensions, i definitely need a model of each of them to match with the scene. I don't want a model based tracker because, i need to create separate models of a box of all dimensions that my tracker will track.

My question : Is there a model free object tracker that will detect boxes in the scene,calculate its dimensions and gives their 6DOF and track them? My target object is just a box and i want to eliminate all other things in the scene.

Thanks in advance

edit retag flag offensive close merge delete

1 answer

Sort by ยป oldest newest most voted

answered 2016-10-20 11:06:06 -0600

Tetragramm gravatar image

Well, it's certainly possible to do what you want, but how easy it is depends on what you know or can assume about the problem.

No matter what, you need to know either the distance to part of the object, or the size of the object. Otherwise you're going to not have enough information to get the full 6DOF. From what you say, though, the camera is fixed. So you know where the conveyor is relative to the camera, and hence where the bottom of the box is.

From there it's just detecting rectangular objects travelling along the conveyor and some simple geometry. Not easy, but straightforward.

Your first step is likely to calibrate your camera and determine it's position relative to the conveyor. If you can figure out the 3D location of a single point travelling down the conveyor, you're most of the way there.

edit flag offensive delete link more


I would tell u what i have done till now: 1) I am using two cameras and hence did internal calibration of two cameras. 2) Using alvar markers, i set up a world frame on the conveyor and then transformed the views of these two cameras from their respective camera frame to the world coordinate frame that i created with the alvar markers and finally stitched the images from the two cameras in world frame. 3) did point cloud pre processing like downsampling, pass through and clustering. Now the resultant view gives me point clouds with only the boxes with some noise.,

Next i need to measure the box dimesions and then perform tracking of the box. I tried bounding box method with PCL but it does not suit because it just draws a box on top of the object irrespective of the pose of the object.

ramanan1991 gravatar imageramanan1991 ( 2016-10-20 13:47:54 -0600 )edit

as i said the boxes can be of different dimensions, i can't use a model to perform feature matching. I am really confused if i should be using point cloud library or opencv for this.

is there a solution?

ramanan1991 gravatar imageramanan1991 ( 2016-10-20 13:48:08 -0600 )edit

Ah, well if you've got the point cloud, you're done with OpenCV. I don't know much about PCL, but THIS is vaguely related.

Tetragramm gravatar imageTetragramm ( 2016-10-20 17:37:06 -0600 )edit This was really interesting, but i don't know if it will fit my needs

ramanan1991 gravatar imageramanan1991 ( 2016-10-24 07:39:57 -0600 )edit

Probably not. That would be for tracking before you generate the point cloud.

I'm pretty sure PCL has what you need. I suggest taking a look at their tutorials. Region Growing plus some simple post processing ought to find boxes pretty easily.

Tetragramm gravatar imageTetragramm ( 2016-10-24 17:14:38 -0600 )edit

Question Tools

1 follower


Asked: 2016-10-20 10:13:22 -0600

Seen: 1,351 times

Last updated: Oct 20 '16