Figure-ground organization of 3-D scenes

摘要

Systems and methods for processing a pair of 2-D images are described. In one example, a stereoscopic set of images is converted into a collection of regions that represent individual 3-D objects in the pair of images. In one embodiment, the system recovers the 3-D point P for each point p that appears in both images. It estimates the 3-D orientation of the floor plane, and the image capture planes and their height from the floor. The system then identifies the collection B of points P that do not represent points on the floor and generates a projection C of B onto a plane parallel to the floor. It blurs the projection C and identifies peaks in the blurred image, then fits symmetric figures to the points in C around the identified peaks. The system projects the 3-D figures associated with the symmetric figures back onto the 2-D images.

主权项

1. A method for locating objects in a pair of two-dimensional images of a three-dimensional scene taken by an imaging device having two image capture planes, where the scene includes a floor, the method comprising:
recovering a 3-D point P for each of a plurality of points p that appear in both of the two-dimensional images; estimating a 3-D orientation of the floor, a 3-D orientation of the image capture planes, and a height of the image capture planes from the floor; generating a projection C onto a plane parallel to the floor of the collection of points P that do not represent points on the floor; partitioning the projection C into regions representing objects; fitting one or more symmetric figures to the points in C around the identified regions; and generating an output signal that encodes the one or more symmetric figures.