Distances evolution analysis for online and off-line human object interaction recognition

Meng, Meng; Drira, Hassen; Boonaert, Jacques

Type de document :

Compte-rendu et recension critique d'ouvrage

Titre :

Distances evolution analysis for online and off-line human object interaction recognition

Auteur(s) :

Meng, Meng [Auteur]
Tianjin University of Science and Technology [TUST]
Drira, Hassen [Auteur]

Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Boonaert, Jacques [Auteur]
Centre for Digital Systems [CERI SN - IMT Nord Europe]

Titre de la revue :

Image and Vision Computing

Éditeur :

Elsevier

Date de publication :

2017-12

ISSN :

0262-8856

Mot(s)-clé(s) en anglais :

Human object interaction
rate invariance
shape analysis
temporal modeling

Discipline(s) HAL :

Informatique [cs]/Vision par ordinateur et reconnaissance de formes [cs.CV]

Résumé en anglais : [en]

Human action recognition in 3D sequences is one of the most challenging and active areas of research in the computer vision domain. However designing automatic systems that are robust to significant variability due to ...
Lire la suite >Human action recognition in 3D sequences is one of the most challenging and active areas of research in the computer vision domain. However designing automatic systems that are robust to significant variability due to object combinations and high complexity of human motions are more challenging in addition to the typical requirements such as rotation, translation, and scale invariance is challenging task. In this paper, we propose a spatio-temporal modeling of human-object interaction videos for on-line and off-line recognition. The inter joint distances and the object are considered as low-level features for online classification. For off-line recognition, we propose rate-invariant classification of full video and early recognition. A shape analysis of trajectories of the inter-joint and object-joints distances is proposed for this end. The experiments conducted following state-of-the-art settings using MSR Daily Activity 3D Dataset and On-line RGBD Action Dataset and on a new Multi-view dataset for human object interaction demonstrate that the proposed approach is effective and discrimina-tive for human object interaction classification as demonstrated here.Lire moins >

Langue :

Anglais

Vulgarisation :

Non

Collections :