A Plane Segmentation System Based on Affine Homography and Optical Flow

Planes are important geometric features and can be used in a wide range of vision tasks like scene reconstruction, path planning and robot navigation. This work aims to illustrate a plane segmentation system based on homography computation and optical flow estimation. Firstly, using two image frames from a monocular sequence, a set of match pairs of interest points is obtained. An algorithm was developed to cluster interest points belonging to the same plane based on the reprojection error of the affine homography. From the calculated homographies, the planar flow is computed for each image pixel. Following this, the estimated optical flow is used to expand and identify each detected plane. An optical flow estimation method based on color information is presented. Tests were performed in different sequences of indoor and outdoor images and results validate the proposed system.