Lucas-Kanade 20 Years On: A Unifying Framework
International Journal of Computer Vision
| Hi-index | 0.00 |
Planar tracking on video streams is a subject of interest in the augmented reality, robotics, human computer interaction for games or computer vision fields. Traditional approaches to tracking are based on finding correspondences in successive images. This can be achieved by computing optical flow or by matching a sparse collection of features. The tracking method presented in this paper belongs to the first group of methods. It is based on minimizing the sum-of-squared differences (SSD) between a previously stored image of the target (the template) and the current image of it.