Trajectories Comparing Based on Matching ...
Document type :
Autre communication scientifique (congrès sans actes - poster - séminaire...): Communication dans un congrès avec actes
Title :
Trajectories Comparing Based on Matching and Distance Evaluation Within Stiefel and Grassmann Manifolds
Author(s) :
Elaoud, Amani [Auteur]
Université de Tunis El Manar [UTM]
Barhoumi, Walid [Auteur]
Université de Tunis El Manar [UTM]
Drira, Hassen [Auteur]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Zagrouba, Ezzeddine [Auteur]
Laboratoire de recherche en Génie Logiciel, Applications distribuées, Systèmes décisionnels et Imagerie intelligente [Manouba] [RIADI]
Université de Tunis El Manar [UTM]
Barhoumi, Walid [Auteur]
Université de Tunis El Manar [UTM]
Drira, Hassen [Auteur]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Zagrouba, Ezzeddine [Auteur]
Laboratoire de recherche en Génie Logiciel, Applications distribuées, Systèmes décisionnels et Imagerie intelligente [Manouba] [RIADI]
Conference title :
International Conference INFORMATION SYSTEMS IADIS
City :
Lisbonne
Country :
Portugal
Start date of the conference :
2017
English keyword(s) :
Trajectories
Human Motion
Grassmann
Stiefel
Kinect
manifolds
matching
skeleton
Human Motion
Grassmann
Stiefel
Kinect
manifolds
matching
skeleton
HAL domain(s) :
Informatique [cs]
Informatique [cs]/Vision par ordinateur et reconnaissance de formes [cs.CV]
Informatique [cs]/Vision par ordinateur et reconnaissance de formes [cs.CV]
English abstract : [en]
In this paper, we are interested in comparing human trajectories using skeleton information provided by a consumer RGB-D sensor. In fact, 3D human joints given by skeletons offer an important information for human motion ...
Show more >In this paper, we are interested in comparing human trajectories using skeleton information provided by a consumer RGB-D sensor. In fact, 3D human joints given by skeletons offer an important information for human motion analysis. In this context, the use of manifolds has grown considerably in the computer vision community in recent years. The main contribution of this study resides in working jointly with two manifolds. The matching of the trajectories is performed in Stiefel manifold and dissimilarity measure is carried out in Grassmann manifold. Indeed, trajectories of motions are provided by the projection in the Stiefel manifold. Then, the Stiefel distance is used within the dynamic time warping in order to define the appropriate matching between a reference trajectory and a test one. This allows avoiding that the rotation within the motion will be ignored, as it is the case with the Grassmann manifold. Then, the dissimilarity is evaluated using the Grassmann distance to compare motions while being invariant against rotation. Realized experiments on standard challenging datasets prove that the proposed method, for the comparison of human trajectories with different sizes, performs accurately compared to existing manifold-based methods.Show less >
Show more >In this paper, we are interested in comparing human trajectories using skeleton information provided by a consumer RGB-D sensor. In fact, 3D human joints given by skeletons offer an important information for human motion analysis. In this context, the use of manifolds has grown considerably in the computer vision community in recent years. The main contribution of this study resides in working jointly with two manifolds. The matching of the trajectories is performed in Stiefel manifold and dissimilarity measure is carried out in Grassmann manifold. Indeed, trajectories of motions are provided by the projection in the Stiefel manifold. Then, the Stiefel distance is used within the dynamic time warping in order to define the appropriate matching between a reference trajectory and a test one. This allows avoiding that the rotation within the motion will be ignored, as it is the case with the Grassmann manifold. Then, the dissimilarity is evaluated using the Grassmann distance to compare motions while being invariant against rotation. Realized experiments on standard challenging datasets prove that the proposed method, for the comparison of human trajectories with different sizes, performs accurately compared to existing manifold-based methods.Show less >
Language :
Anglais
Peer reviewed article :
Oui
Audience :
Internationale
Popular science :
Non
Collections :
Source :
Files
- https://hal.archives-ouvertes.fr/hal-01703958/document
- Open access
- Access the document
- https://hal.archives-ouvertes.fr/hal-01703958/document
- Open access
- Access the document
- https://hal.archives-ouvertes.fr/hal-01703958/document
- Open access
- Access the document
- document
- Open access
- Access the document
- amani_AC.pdf
- Open access
- Access the document