Store/print pixel-information after Object Detection is applied

asked 2017-05-05 11:23:04 -0500

sollet gravatar image

I have a project where we need to gather the depth-information of an object in a video stream. Is there a way to apply object-detection to isolate the object in question, then return the Depth-based pixel-information for certian "known" points on my desired object, on a frame-by-frame basis?

edit retag flag offensive close merge delete

Comments

depends on what object you want to detect

electron gravatar imageelectron ( 2017-05-05 16:05:12 -0500 )edit

you'll have a hard time, trying to get depth information at all from a single camera/video src, it's next to impossible.

berak gravatar imageberak ( 2017-05-06 07:35:12 -0500 )edit

We need to track a beam moving through a defined volume-space... Is there a minimum number of RGB-D cameras we would need?

sollet gravatar imagesollet ( 2017-05-08 07:59:44 -0500 )edit

sorry, from a single RGB cam. with RGBD, you already have a depth channel

berak gravatar imageberak ( 2017-05-08 08:34:08 -0500 )edit