Abstract: This paper shows that how to detect object using more efficient techniques in real time detection. We successively evaluate the features used in sliding window detection process to decide about object presence/ absence also contain knowledge about object deformation. We exploit these detection features to estimate the object deformation. Estimated deformation is then immediately applied to not yet evaluated features to align them with the observed image data. For increasing the efficiency we only check the windows which are near to the window in which object is detected in previous slide rather than detecting every window everytime. By using sliding window detection process we filter frame and increase tracking speed.

 

Keywords: Features, classifier, training data, sequence detection, sliding window, object deformation