New College SFM (source code available)

The video shows the results of estimating visual odometry and independent motion on the New College data set (http://www.robots.ox.ac.uk/NewCollegeData/). Visual odometry is obtained by the algorithm of Badino and Kanade [1] that minimizes the reprojection error of tracked features. Features are tracked by the KLT algorithm. Independent motion is obtained by tracking feature position and velocity over time by means of a Kalman filter. Main window: the arrows show the prediction of position of the tracked features in 2 seconds. Color encodes the 3D Euclidean speed from green to red. Top-right: optical flow vectors of the tracked features. The color encodes the length of the optical flow vector. Middle-right: color encoded stereo disparities from green to red. Bottom-right: estimated traveled path by dead-reckoning (no loop closure) The source code is available on SourceForge.net: http://sourceforge.net/projects/qcv/ [1] Hernan Badino and Takeo Kanade. A Head-Wearable Short-Baseline Stereo System for the Simultaneous Estimation of Structure and Motion. In IAPR Conference on Machine Vision Applications (MVA), Nara, Japan, June 2011. Hernan Badino http://lelaps.de/