Motion Trend Patterns for Action Modelling and Recognition - ENSTA Paris - École nationale supérieure de techniques avancées Paris Accéder directement au contenu
Communication Dans Un Congrès Année : 2013

Motion Trend Patterns for Action Modelling and Recognition

Thanh Phuong Nguyen
Antoine Manzanera
Matthieu Garrigues

Résumé

A new method for action modelling is proposed, which com-bines the trajectory beam obtained by semi-dense point tracking and a local binary trend description inspired from the Local Binary Patterns (LBP). The semi dense trajectory approach represents a good trade-off between reliability and density of the motion field, whereas the LBP component allows to capture relevant elementary motion elements along each trajectory, which are encoded into mixed descriptors called Motion Trend Patterns (MTP). The combination of those two fast operators al-lows a real-time, on line computation of the action descriptors, composed of space-time blockwise histograms of MTP values, which are classified using a fast SVM classifier. An encoding scheme is proposed and com-pared with the state-of-the-art through an evaluation performed on two academic action video datasets.
Fichier principal
Vignette du fichier
MTP.pdf (1.97 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01118278 , version 1 (18-02-2015)

Identifiants

Citer

Thanh Phuong Nguyen, Antoine Manzanera, Matthieu Garrigues. Motion Trend Patterns for Action Modelling and Recognition. Computer Analysis of Images and Patterns (CAIP), Aug 2013, York, United Kingdom. pp.360 - 367, ⟨10.1007/978-3-642-40261-6_43⟩. ⟨hal-01118278⟩
225 Consultations
75 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More