Motion recognition using non parametric image motion models estimated from temporal and multiscale cooccurrence statistics

Ronan Fablet (1, 2) , Patrick Bouthemy (2)

1 SC - Département Signal et Communications
2 IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires

A new approach for motion characterization in image sequences is presented. It relies on the probabilistic modeling of temporal and scale co-occurrence distributions of local motion-related measurements directly computed over image sequences. Temporal multiscale Gibbs models allow us to handle both spatial and temporal aspects of image motion content within a unified statistical framework. Since this modeling mainly involves the scalar product between co-occurrence values and Gibbs potentials, we can formulate and address several fundamental issues: model estimation according to the ML criterion (hence, model training and learning) and motion classification. We have conducted motion recognition experiments over a large set of real image sequences comprising various motion types such as temporal texture samples, human motion examples, and rigid motion situations

Mots clés: Image classification; Image sequences; Image texture; Motion estimation; Nonparametric statistics; Spatiotemporal phenomena; Nonparametric motion analysis; motion recognition; multiscale analysis; Gibbs models; co-occurrences; ML criterion

Domaines: Traitement du signal et de l'image [eess.SP]

Ronan Fablet, Patrick Bouthemy. Motion recognition using non parametric image motion models estimated from temporal and multiscale cooccurrence statistics. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003, 25 (12), pp.1619 - 1624. ⟨10.1109/TPAMI.2003.1251155⟩. ⟨hal-02341756⟩