Similarity Learning For Motion Estimation

Short overview of the Chapter—Motion estimation necessitates an appropriate choice of similarity function. Because generic similarity functions derived from simple assumptions are insufficient to model complex yet structured appearance variations in motion estimation, we propose to learn a discriminative similarity function to match images under varying appearances by casting image matching into a binary classification problem. We use the LogitBoost algorithm to learn the classifier based on an annotated database that exemplifies the structured appearance variations: An image pair in correspondence is positive and an image pair out of correspondence is negative. To leverage the additional distance structure of negatives, we present a location-sensitive cascade training procedure that bootstraps negatives for later stages of the cascade from the regions closer to the positives, which enables viewing a large number of negatives and steering the training process to yield lower training and test errors. We apply the learned similarity function to estimating the motion for the endocardial wall of left ventricle in echocardiography and to performing visual tracking. We obtain improved performances when comparing the learned similarity function with conventional ones.