- Published 1996 in Computer Vision and Image Understanding

method and show that the computations can be performed in time which is linear in the number of points for a dense This paper presents and analyzes a new constraint (the Collinear Point Constraint) for use in computing egomotion and flow field, thus satisfying the need for a fast solution. The detecting independently moving objects from image motion. detection of independently moving regions is based on a The constraint provides an elegant means of cancelling the theorem which we derive giving the minimal number of rotation of the observer while simultaneously providing evipoints needed for the detection of such regions. The theory dence for the location of the Translational Focus of Expansion. is compared to that of previous methods; experimental It also provides the minimal computation necessary for deverification of its performance using synthetic input is pretecting independent motion. 1996 Academic Press, Inc. sented. These two computations provide important inputs for an attention system (such as that described in [45, 5]) in

@article{Lobo1996ComputingEA,
title={Computing Egomotion and Detecting Independent Motion from Image Motion Using Collinear Points},
author={Niels da Vitoria Lobo and John K. Tsotsos},
journal={Computer Vision and Image Understanding},
year={1996},
volume={64},
pages={21-52}
}