Fast and Accurate Registration of Visible and Infrared Videos

In this work, we propose a novel fast and accurate method based on keypoints and temporal information to solve the registration problem on planar scenes with moving objects for infrared-visible stereo pairs. A keypoint descriptor and a temporal buffer (reservoir) filled with matched keypoints are used in order to find the homography transformation for the registration. Inside a given frame, the problem of registration is formulated as correspondences between noisy polygon vertices. Sections of polygons are matched locally to find the corresponding vertices inside a frame. These correspondences are then accumulated temporally using a reservoir of matches for homography calculation. Results show that our method outperforms two recent state-of-the-art global registration methods by a large margin in almost all tested videos.