Shrec'16 Track: Retrieval of Human Subjects from Depth Sensor Data

Abstract : In this paper we report the results of the SHREC 2016 contest on "Retrieval of human subjects from depth sensor data". The proposed task was created in order to verify the possibility of retrieving models of query human subjects from single shots of depth sensors, using shape information only. Depth acquisition of different subjects were realized under different illumination conditions, using different clothes and in three different poses. The resulting point clouds of the partial body shape acquisitions were segmented and coupled with the skeleton provided by the OpenNI software and provided to the participants together with derived triangulated meshes. No color information was provided. Retrieval scores of the different methods proposed were estimated on the submitted dissimilarity matrices and the influence of the different acquisition conditions on the algorithms were also analyzed. Results obtained by the participants and by the baseline methods demonstrated that the proposed task is, as expected, quite difficult, especially due the partiality of the shape information and the poor accuracy of the estimated skeleton, but give useful insights on potential strategies that can be applied in similar retrieval procedures and derived practical applications. Categories and Subject Descriptors (according to ACM CCS): I.4.8 [IMAGE PROCESSING AND COMPUTER VISION]: Scene Analysis—Shape
Type de document :
Communication dans un congrès
Eurographics Workshop on 3D Object Retrieval, May 2016, Lisbon, Portugal. pp.1-6, 2016, 〈http://eg3dor2016.ist.utl.pt/programme.html〉
Liste complète des métadonnées

Littérature citée [11 références]  Voir  Masquer  Télécharger

https://hal.archives-ouvertes.fr/hal-01314067
Contributeur : Santiago Velasco-Forero <>
Soumis le : mardi 10 mai 2016 - 17:39:24
Dernière modification le : mardi 12 septembre 2017 - 11:40:48
Document(s) archivé(s) le : mercredi 16 novembre 2016 - 00:01:19

Fichier

shrec16.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01314067, version 1

Collections

Citation

A Giachetti, F Fornasa, F Parezzan, A Saletti, L Zambaldo, et al.. Shrec'16 Track: Retrieval of Human Subjects from Depth Sensor Data. Eurographics Workshop on 3D Object Retrieval, May 2016, Lisbon, Portugal. pp.1-6, 2016, 〈http://eg3dor2016.ist.utl.pt/programme.html〉. 〈hal-01314067〉

Partager

Métriques

Consultations de
la notice

198

Téléchargements du document

251