Relevant LMA Features for Human Motion Recognition

Abstract : Motion recognition from videos is actually a very complex task due to the high variability of motions. This paper describes the challenges of human motion recognition, especially motion representation step with relevant features. Our descriptor vector is inspired from Laban Movement Analysis method. We propose discriminative features using the Random Forest algorithm in order to remove redundant features and make learning algorithms operate faster and more effectively. We validate our method on MSRC-12 and UTKinect datasets.
Type de document :
Communication dans un congrès
20th International Conference on Image Analysis and Processing (ICIAP 2018), Oct 2018, Paris, France. Proc. of the 20th International Conference on Image Analysis and Processing (ICIAP 2018), 2018, Proc. of the 20th International Conference on Image Analysis and Processing (ICIAP 2018)
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01971029
Contributeur : Frédéric Davesne <>
Soumis le : dimanche 6 janvier 2019 - 18:13:27
Dernière modification le : jeudi 7 février 2019 - 15:38:59

Fichier

ICIAP1-bestPaperAward.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01971029, version 1

Citation

Insaf Ajili, Malik Mallem, Jean-Yves Didier. Relevant LMA Features for Human Motion Recognition. 20th International Conference on Image Analysis and Processing (ICIAP 2018), Oct 2018, Paris, France. Proc. of the 20th International Conference on Image Analysis and Processing (ICIAP 2018), 2018, Proc. of the 20th International Conference on Image Analysis and Processing (ICIAP 2018). 〈hal-01971029〉

Partager

Métriques

Consultations de la notice

40

Téléchargements de fichiers

17