A new motion detection algorithm based on Sigma-Delta background estimation

Antoine Manzanera (1) , Julien Richefeu (1)

1 RV - Robotique et Vision

Motion detection using a stationary camera can be done by estimating the static scene (background). In that purpose, we propose a new method based on a simple recursive non linear operator, the Sigma-Delta filter. Used along with a spatiotemporal regularization algorithm, it allows robust, computationally efficient and accurate motion detection. To deal with complex scenes containing a wide range of motion models with very different time constants, we propose a generalization of the basic model to multiple Sigma-Delta estimation.

Mots clés: Motion detection; Background estimation; Recursive filtering

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]; Traitement des images [eess.IV]

Antoine Manzanera, Julien Richefeu. A new motion detection algorithm based on Sigma-Delta background estimation. Pattern Recognition Letters, 2007, 28 (3), ⟨10.1016/j.patrec.2006.04.007⟩. ⟨hal-01222650⟩