Augmented Reality Experiment (Chapter I – Fig. 1.8)

On the left the result with the Kalman filter, on the right with the SVK.


The same video at a lower frame rate (1 fps)


Mean-shift tracking through occlusions (Chapter I - Fig. 1.9)

Red: Mean-shift. Blue: Mean-shift + Kalman. Cyan: Mean-shift + SVK.


Dollar Sequence (Chapter II - Fig. 2.7)


Faceocc2 Sequence (Chapter II - Fig. 2.9)


Coke Sequence (Chapter II - Fig. 2.11)


CAVIAR_1 sequence (Chapter III - Fig. 3.4)


CAVIAR_2 sequence (Chapter III - Fig. 3.5)


CAVIAR_3 sequence (Chapter III - Fig. 3.6)


ISSIA_P sequence (Chapter III - Fig. 3.8)


Object Recognition in Clutter and Occlusions (Chapter IV)
This video demonstrate object recognition by means of CSHOT point-to-point correspondences and Hough voting in RGB-D data provided by a Kinect sensor .