Time-of-Flight based multi-sensor fusion strategies for hand gesture recognition

Thomas Kopinski (1) , Darius Malysiak (1) , Alexander Gepperth (2, 3) , Uwe Handmann (1)

1 University of Applied Sciences [Bottrop]
2 Flowers - Flowing Epigenetic Robots and Systems
3 RV - Robotique et Vision

Building upon prior results, we present an alterna-tive approach to efficiently classifying a complex set of 3D hand poses obtained from modern Time-Of-Flight-Sensors (TOF). We demonstrate it is possible to achieve satisfactory results in spite of low resolution and high noise (inflicted by the sensors) and a demanding outdoor environment. We set up a large database of pointclouds in order to train multilayer perceptrons as well as support vector machines to classify the various hand poses. Our goal is to fuse data from multiple TOF sensors, which observe the poses from multiple angles. The presented contribution illustrates that real-time capability can be maintained with such a setup as the used 3D descriptors, the fusion strategy as well as the online confidence measures are computationally efficient.

Mots clés: gesture recognition; support vector machines; neural networks; efficient classification; tof sensors

Domaines: Apprentissage [cs.LG]

Thomas Kopinski, Darius Malysiak, Alexander Gepperth, Uwe Handmann. Time-of-Flight based multi-sensor fusion strategies for hand gesture recognition. IEEE International Symposium on Computational Intelligence and Informatics, Nov 2014, Budapest, Hungary. ⟨hal-01098695⟩