Human Motion Prediction Using Manifold-Aware ...
Document type :
Autre communication scientifique (congrès sans actes - poster - séminaire...): Communication dans un congrès avec actes
Title :
Human Motion Prediction Using Manifold-Aware Wasserstein GAN
Author(s) :
Chopin, Baptiste [Auteur]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Otberdout, Naima [Auteur]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Daoudi, Mohamed [Auteur]
Ecole nationale supérieure Mines-Télécom Lille Douai [IMT Nord Europe]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Bartolo, Angela [Auteur]
Laboratoire Sciences Cognitives et Sciences Affectives - UMR 9193 [SCALab]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Otberdout, Naima [Auteur]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Daoudi, Mohamed [Auteur]
Ecole nationale supérieure Mines-Télécom Lille Douai [IMT Nord Europe]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Bartolo, Angela [Auteur]
Laboratoire Sciences Cognitives et Sciences Affectives - UMR 9193 [SCALab]
Conference title :
IEEE conference series on Automatic Face and Gesture Recognition
City :
Jodhpur (virtual)
Country :
Inde
Start date of the conference :
2021-12-15
HAL domain(s) :
Informatique [cs]/Intelligence artificielle [cs.AI]
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]
Human motion prediction aims to forecast future human poses given a prior pose sequence. The discontinuity of the predicted motion and the performance deterioration in long-term horizons are still the main challenges ...
Show more >Human motion prediction aims to forecast future human poses given a prior pose sequence. The discontinuity of the predicted motion and the performance deterioration in long-term horizons are still the main challenges encountered in current literature. In this work, we tackle these issues by using a compact manifold-valued representation of human motion. Specifically, we model the temporal evolution of the 3D human poses as trajectory, what allows us to map human motions to single points on a sphere manifold. To learn these non-Euclidean representations, we build a manifold-aware Wasserstein generative adversarial model that captures the temporal and spatial dependencies of human motion through different losses. Extensive experiments show that our approach outperforms the state-of-the-art on CMU MoCap and Human 3.6M datasets. Our qualitative results show the smoothness of the predicted motions. The pretrained models and the code are provided at the following link.Show less >
Show more >Human motion prediction aims to forecast future human poses given a prior pose sequence. The discontinuity of the predicted motion and the performance deterioration in long-term horizons are still the main challenges encountered in current literature. In this work, we tackle these issues by using a compact manifold-valued representation of human motion. Specifically, we model the temporal evolution of the 3D human poses as trajectory, what allows us to map human motions to single points on a sphere manifold. To learn these non-Euclidean representations, we build a manifold-aware Wasserstein generative adversarial model that captures the temporal and spatial dependencies of human motion through different losses. Extensive experiments show that our approach outperforms the state-of-the-art on CMU MoCap and Human 3.6M datasets. Our qualitative results show the smoothness of the predicted motions. The pretrained models and the code are provided at the following link.Show less >
Language :
Anglais
Peer reviewed article :
Oui
Audience :
Internationale
Popular science :
Non
Collections :
Source :
Files
- https://hal.archives-ouvertes.fr/hal-03384332/document
- Open access
- Access the document
- https://hal.archives-ouvertes.fr/hal-03384332/document
- Open access
- Access the document
- https://hal.archives-ouvertes.fr/hal-03384332/document
- Open access
- Access the document
- document
- Open access
- Access the document
- IEEE_FG_2021-3.pdf
- Open access
- Access the document