Multi-view pose estimation with mixtures-of-parts and adaptive viewpoint selection

Abstract : We propose a new method for human pose estimation which leverages information from multiple views to impose a strong prior on articulated pose. The novelty of the method concerns the types of coherence modelled. Consistency is maximised over the different views through different terms modelling classical geometric information (coherence of the resulting poses) as well as appearance information which is modelled as latent variables in the global energy function. Moreover, adequacy of each view is assessed and their contributions are adjusted accordingly. Experiments on the HumanEva and UMPM datasets show that the proposed method significantly decreases the estimation error compared to single-view results.
Type de document :
Pré-publication, Document de travail
8 pages, 7 figures, 4 tables. Second revision to the paper, as submitted to IET Computer Vision o.. 2017
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01593547
Contributeur : Christian Wolf <>
Soumis le : mardi 26 septembre 2017 - 14:12:16
Dernière modification le : lundi 10 décembre 2018 - 17:47:39

Lien texte intégral

Identifiants

  • HAL Id : hal-01593547, version 1
  • ARXIV : 1709.08527

Collections

Citation

Emre Dogan, Gonen Eren, Christian Wolf, Eric Lombardi, Atilla Baskurt. Multi-view pose estimation with mixtures-of-parts and adaptive viewpoint selection. 8 pages, 7 figures, 4 tables. Second revision to the paper, as submitted to IET Computer Vision o.. 2017. 〈hal-01593547〉

Partager

Métriques

Consultations de la notice

138