A probabilistic modeling approach to one-shot gesture recognition

Onderzoeksoutput: Bijdrage aan tijdschriftTijdschriftartikelAcademic

17 Downloads (Pure)


Gesture recognition enables a natural extension of the way we currently interact with devices. Commercially available gesture recognition systems are usually pre-trained and offer no option for customization by the user. In order to improve the user experience, it is desirable to allow end users to define their own gestures. This scenario requires learning from just a few training examples if we want to impose only a light training load on the user. To this end, we propose a gesture classifier based on a hierarchical probabilistic modeling approach. In this framework, high-level features that are shared among different gestures can be extracted from a large labeled data set, yielding a prior distribution for gestures. When learning new types of gestures, the learned shared prior reduces the number of required training examples for individual gestures. We implemented the proposed gesture classifier for a Myo sensor bracelet and show favorable results for the tested system on a database of 17 different gesture types. Furthermore, we propose and implement two methods to incorporate the gesture classifier in a real-time gesture recognition system.
Originele taal-2Engels
Aantal pagina's24
StatusGepubliceerd - 29 jun 2018

Vingerafdruk Duik in de onderzoeksthema's van 'A probabilistic modeling approach to one-shot gesture recognition'. Samen vormen ze een unieke vingerafdruk.

  • Activiteiten

    Annual machine learning conference of the Benelux (Benelearn 2017)

    Anouk van Diepen (Deelnemer)

    10 jun 2017

    Activiteit: Types deelname aan of organisatie van een evenementCongresWetenschappelijk


    Nominated for the KNVI Scriptie Prijs voor Informatica en Informatiekunde

    van Diepen, Anouk (Ontvanger), 2017

    Prijs: AndersWerk, activiteit of publicatie gerelateerde prijzen (lifetime, best paper, poster etc.)Wetenschappelijk

    Citeer dit