I want to apply object detection models, maybe like an YOLO5 on the RGB colour images that I am getting from the camera frames and then I want the point cloud of only that part of the image that has objects. I tried using the examples where I found the following code:
points = frames.convert_to_points(camera_param)
here the points is a 1d list of x, y and z coordiantes of all the pixels in the image. How do I get the coordinates of the required object only?