The smart furniture provide various services alone or by communication with other devices. The cost and time for building such devices is a barrier to the deployment of various novel applications. Since the smart devices are typically equipped with embedded or traditional computers, various sensors and multimodal I/O devices, it can provide various services in public spaces as well as in private spaces. The goal of our research was to introduce affordable techniques on how to build and model the acceptable multimodal HCIs and interactions with the environment all integrated in the mobile smart furniture device. In this paper we present a detailed analysis of the associated, promising state of the art techniques and the initial set of techniques we have used to realize the multimodal HCI interaction, followed by the descriptions of implemented control algorithms and some first-hand experimental evaluation analysis.

ROBOSOFA-Low Cost Multimodal I/O Fusion for Smart Furniture

A computing system receives user input via a touch-interface that involves motion of one or more touches relative to the touch-interface. Responsive to the user input including two or more concurrent touches of the touch-interface involving motion in a coordinate direction, the computing system compresses graphical content within a graphical user interface in the coordinate direction toward a reference datum line. Responsive to the user input including two or more concurrent touches of the touch-interface involving motion in an opposite direction from the coordinate direction, the computing system expands graphical content within the graphical user interface in the opposite direction and away from the reference datum line.

Multi-touch GUI featuring directional compression and expansion of graphical content

L'utilisation des surfaces tactiles est actuellement de plus en plus repandue dans divers domaines, cette technologie vise a se rapprocher le plus possible d'une utilisation simple et naturelle. L'arrivee du nouveau type de surfaces multipoints, permettant plusieurs point de contact a ouvert un nouveau champ de possibilites pour la composition et la manipulation des documents numeriques. Jusqu'ici ces interfaces n'autorisent souvent que des commandes basiques de manipulation, comme par exemple agrandir une image en utilisant deux doigts ou encore passer d'une page a l'autre en glissant le doigt sur la surface de l'ecran. Le travail effectue lors de ce stage vise l'exploration de ce domaine en abordant la reconnaissance de commandes gestuelles graphiques multipoints plus evoluees. Plusieurs approches pour la reconnaissance de forme existent dans la litterature, les methodes structurelles a base de graphe reposent sur une representation analytique des formes, modelisant les differentes parties de la forme ainsi que les relations existantes entre ces parties. Nous representons les gestes multipoints a reconnaitre, sous forme de graphes de relations spatiales et temporelles. Pour comparer et reconnaitre des symboles, les graphes sont affectes par les imperfections des outils de comparaison disponibles. Les techniques de projection de graphe permettent de pallier cette complexite en cherchant a repositionner les graphes dans un espace vectoriel et permettre ainsi l'utilisation d'outils statistiques. Notre approche est basee sur les techniques de projection de graphes. Nous proposons d'etendre une methode [44] de l'etat de l'art, qui projette un graphe en comptant le nombre d'occurrence de motifs predefinis, ces motifs sont des sous-structures de graphes construits independamment du systeme de reconnaissance pour lequel ils sont employes. Nous proposons d'extraire ces motifs directement de l'ensemble des graphes manipules, en selectionnant les sous-structures de graphe les plus frequentes dans la base de graphes considere. Ce qui permet un lien plus fort avec la nature des graphes manipule.

https://dumas.ccsd.cnrs.fr/dumas-00725332/document

Composition manuscrite de gestes graphiques sur des interfaces tactiles multipoints

Multi-touch interaction is a more free interaction means, however, recent related applications limit in the 2D area, such as interaction desktop, shop window. For the applications in virtual exhibition, this paper proposes a 3D multi-touch recognition method. The different gestures firstly are defined for gesture recognition. Then a character recognition algorithm based on directions is presented to achieve the input and erasing gesture of characters. Further, based on a quantitative touch feedback mechanism, the three-dimensional multi-touch interaction algorithm is presented. Finally, we implement the multi-touch interaction display in different conditions, including moving, rotating, zoom for three-dimensional scene.

Long Wang

Papers

3D Multi-touch recognition based virtual interaction