Computer vision module overview

ViSP contains a computer vision module that allows to compute a pose or an homography from points using a robust scheme. It embed also a wrapper over OpenCV able to detect, to describe or to match keypoints. This module contains also the necessary material for camera calibration.

The source code documentation shows which are the corresponding classes part of computer vision module.

This module uses optional 3rd party libraries.

Pose computation from points

This video shows an example of pose computation thanks to the vpPose class. Here pose estimation is performed from 4 points whose coordinates are obtained from tre tracking of the blobs.	The four corners of the postcard are used as the reference points for the pose computation. In order to compute their coordinates in the image, the borders of the postcard are tracked thanks to four vpMeLine. Their intersections give the required 2D coordinates. The postcard frame is displayed in the image in order to check if the computed pose is correct. [source code]

Homography estimation

From matched keypoints, a robust homography estimation is first performed to estimate the displacement of the postcard. Then a pose estimation from the location of the 4 corners is computed thanks to ViSP.