Recovering Camera Motion and Mobile Objects in Video Documents postscript, pdf, Damien Paulin, Dinesh Kumar, Raghav Bhaskar and Georges M. Quénot, International Workshop on Content-Based Multimedia Indexing, University of Brescia, Italy, September 19-21, 2001.

This paper presents a system that performs the recovery of camera motion parameters and the segmentation of mobile objects in video documents for content indexing.
Two different methods are used for the recovery of the camera motion (relatively to the main background), the first for a camera maintained at a fixed location with rotational and zoom degrees of freedom, and the second for a camera of arbitrary motion but assuming a fixed focal length. The first method is based on the search of an optimal projective transform between consecutive images combined with an iterative background / mobile objects segmentation process. The second method is based on a paraperspective factorization method for shape and motion recovery. Both methods rely on the use of a dense and high-quality matching between consecutive images (optical flow). The system also attempts to classify shots or sub-segments of shots into one of the following categories of ``no motion'', ``non mobile camera motion'', ``mobile camera motion'' or ``other type of motion''. Further subcategorization can be done for each recovered type. Results are presented using sequences extracted from document 8 of the ISIS GDR-PRC GT10/AIM corpus.