Trying to implement object detection using color and depth segmentation using opencv python

asked 2020-02-07 10:28:57 -0500

titli gravatar image

I have got a task to implement object detection. My input is an RGB image,

  1. I am first instructed to do homography transformation,
  2. followed by the transformation of the pixels in RGB colour space to CIE lab colour space.
  3. and next steps are included in the paper I am instructed to follow.

The link of the paper is:this

Where I lack in understanding is, while applying homography I need to specify a source and a destination. If I consider any RGB image as a source then what will be the destination?
It will be really helpful if I can get the codes for at least step 1 and 2 to start properly, Thanking you in advance, Titli

edit retag flag offensive close merge delete

Comments

  1. where does the homography come from ? (input:bgr or depth image, output: scaled / translated image in the same colorspace)
  2. cvtColor() (might need to convert to float first)
berak gravatar imageberak ( 2020-02-10 04:07:58 -0500 )edit

I did not get you

titli gravatar imagetitli ( 2020-02-10 08:14:19 -0500 )edit