A Statistical-Topological Feature Combination for Recognition of Isolated Hand Gestures from Kinect Based Depth Images

Abstract : Reliable hand gesture recognition is an important problem for automatic sign language recognition for the people with hearing and speech disabilities. In this paper, we create a new benchmark database of multi-oriented, isolated ASL numeric images using recently launched Kinect V2. Further, we design an effective statistical-topological feature combinations for recognition of the hand gestures using the available V1 sensor dataset and also over the new V2 dataset. For V1, our best accuracy is 98.4% which is comparable with the best one reported so far and for V2 we achieve an accuracy of 92.2% which is first of its kind.
Type de document :
Communication dans un congrès
IWCIA, Jun 2017, Plovdiv, Bulgaria. 10256, LNCS. 〈10.1007/978-3-319-59108-7〉
Liste complète des métadonnées

Littérature citée [22 références]  Voir  Masquer  Télécharger

https://hal.inria.fr/hal-01517817
Contributeur : Phuc Ngo <>
Soumis le : mercredi 3 mai 2017 - 16:35:05
Dernière modification le : jeudi 11 janvier 2018 - 06:25:24
Document(s) archivé(s) le : vendredi 4 août 2017 - 13:23:34

Fichier

paper29.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Citation

Soumi Paul, Hayat Nasser, Mita Nasipuri, Phuc Ngo, Subhadip Basu, et al.. A Statistical-Topological Feature Combination for Recognition of Isolated Hand Gestures from Kinect Based Depth Images. IWCIA, Jun 2017, Plovdiv, Bulgaria. 10256, LNCS. 〈10.1007/978-3-319-59108-7〉. 〈hal-01517817〉

Partager

Métriques

Consultations de la notice

208

Téléchargements de fichiers

89