Action recognition in videos using frequency analysis of critical point trajectories

Abstract : This paper focuses on human action recognition in video sequences. A method based on the optical flow estimation is presented, where critical points of the flow field are extracted. Multi-scale trajectories are generated from those points and are characterized in the frequency domain. Finally, a sequence is described by fusing this frequency information with motion orientation and shape information. Experiments show that this method has recognition rates among the highest in the state of the art on the KTH dataset. Contrary to recent dense sampling strategies, the proposed method only requires critical points of motion flow field, thus permitting a lower computation time and a better sequence description. Results and perspectives are then discussed.
Type de document :
Communication dans un congrès
IEEE International Conference on Image Processing (ICIP 2014), Oct 2014, Paris, France. p. 1445-1449, 2014
Liste complète des métadonnées


https://hal.archives-ouvertes.fr/hal-01004795
Contributeur : Renaud Péteri <>
Soumis le : mercredi 11 juin 2014 - 16:21:35
Dernière modification le : jeudi 9 février 2017 - 16:58:52
Document(s) archivé(s) le : mardi 11 avril 2017 - 06:41:56

Fichier

BEAUDRY_Cyrille_Action_recogni...
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01004795, version 1

Collections

MIA

Citation

Cyrille Beaudry, Renaud Péteri, Laurent Mascarilla. Action recognition in videos using frequency analysis of critical point trajectories. IEEE International Conference on Image Processing (ICIP 2014), Oct 2014, Paris, France. p. 1445-1449, 2014. <hal-01004795>

Partager

Métriques

Consultations de
la notice

392

Téléchargements du document

295