2D/3D Pose Estimation and Action Recognition using Multitask Deep Learning - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2018

2D/3D Pose Estimation and Action Recognition using Multitask Deep Learning

David Picard
Hedi Tabia

Résumé

Action recognition and human pose estimation are closely related but both problems are generally handled as distinct tasks in the literature. In this work, we propose a multitask framework for jointly 2D and 3D pose estimation from still images and human action recognition from video sequences. We show that a single architecture can be used to solve the two problems in an efficient way and still achieves state-of-the-art results. Additionally , we demonstrate that optimization from end-to-end leads to significantly higher accuracy than separated learning. The proposed architecture can be trained with data from different categories simultaneously in a seamlessly way. The reported results on four datasets (MPII, Human3.6M, Penn Action and NTU) demonstrate the effectiveness of our method on the targeted tasks.
Fichier principal
Vignette du fichier
camera_ready.pdf (1.7 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01815703 , version 1 (14-06-2018)

Identifiants

  • HAL Id : hal-01815703 , version 1

Citer

Diogo C Luvizon, David Picard, Hedi Tabia. 2D/3D Pose Estimation and Action Recognition using Multitask Deep Learning. The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Jun 2018, Salt Lake City, United States. ⟨hal-01815703⟩
136 Consultations
439 Téléchargements

Partager

Gmail Facebook X LinkedIn More