Conférence
Notice
Langue :
Anglais
Crédits
INRIA (Institut national de recherche en informatique et automatique) (Publication), Région PACA (Production), VSP - Vidéo Sud Production (Réalisation), INRIA (Institut national de recherche en informatique et automatique) (Production), UNS (Publication), CNRS - Centre National de la Recherche Scientifique (Publication), Cordelia Schmid (Intervention)
Conditions d'utilisation
Droit commun de la propriété intellectuelle
DOI : 10.60527/z7xb-en87
Citer cette ressource :
Cordelia Schmid. Inria. (2011, 1 décembre). Action recognition from video: some recent results. [Vidéo]. Canal-U. https://doi.org/10.60527/z7xb-en87. (Consultée le 25 avril 2025)
logo Inria
Inria - Institut national de recherche en sciences et technologies du numérique
S'abonner
Contacter

Action recognition from video: some recent results

Réalisation : 1 décembre 2011 - Mise en ligne : 27 janvier 2012
  • document 1 document 2 document 3
  • niveau 1 niveau 2 niveau 3
Descriptif

While recognition in still images has received a lot of attention over the past years, recognition in videos is just emerging. In this talk I will present some recent results.
Bags of features have demonstrated good performance for action recognition in videos. We briefly review the underlying principles and introduce trajectory-based video features, which have shown to outperform the state of the art. These features are obtained by dense point sampling in each frame and tracking them based on displacement information from a dense optical flow field. Trajectory descriptors are obtained from motion boundary histograms, which are robust to camera motion.
We then show how to integrate temporal structure into a bag-of-features model based on so-called actom sequences. We localize actions based on sequences of atomic actions, i.e., represent the temporal structure by sequences of histograms of actom-anchored visual features. This representation is flexible, sparse and discriminative. The resulting model is shown to significantly improve performance over existing methods for temporal action localization. Finally, we show how to move towards more structured representations by explicitly modeling human-object interactions. We learn how to represent human actions as interactions between persons and objects. We localize in space and track over time both the object and the person, and represent an action as the trajectory of the object with respect to the person position, i.e., our human-object interaction features capture the relative trajectory of the object with respect to the human. This is shown to improve over existing methods for action localization.

Intervention
Thème
Discipline :
Documentation
video.ppt
Colloquium Jacques Morgenstern

Le but du colloquium est d’offrir une vision d’ensemble des recherches les plus actives et les plus prometteuses dans le domaine des Sciences et Technologies de l’Information et de la Communication (STIC). Nouveaux thèmes scientifiques

Avec les mêmes intervenants et intervenantes

Sur le même thème