The Issues of 3D Hand Gesture and Posture Recognition Using the Kinect

Mohamed-Ikbel Boulabiar (1, 2) , Gilles Coppin (1, 3) , Franck Poirier (4)

1 LUSSI - Département Logique des Usages, Sciences sociales et Sciences de l'Information
2 Lab-STICC_TB_CID_IHSEV
3 Lab-STICC_TB_CID_IHSEV
4 Lab-STICC - Laboratoire des sciences et techniques de l'information, de la communication et de la connaissance

Besides the emergence of many input devices and sensors, they are still unable to provide good and simple recognition of human postures and gestures. The recognition using simple algorithms implemented on top of these devices (like the Kinect) enlarges use cases for these gestures and postures to newer domains and systems. Our methods cuts the needed computation and allow the integration of other algorithms to run in parallel. We present a system able to track the hand in 3D, log its position and surface information during the time, and recognize hand postures and gestures. We present our solution based on simple geometric algorithms, other tried algorithms, and we discuss some concepts raised from our tests.

Mots clés: Hand; Interaction; Kinect; 3D; Posture; Gesture

Domaines: Interface homme-machine [cs.HC]

Mohamed-Ikbel Boulabiar, Gilles Coppin, Franck Poirier. The Issues of 3D Hand Gesture and Posture Recognition Using the Kinect. HCI International 2014, Jun 2014, Heraklion, Crete, Greece. pp 205-214, ⟨10.1007/978-3-319-07230-2_20⟩. ⟨hal-01141915⟩