Siamese Neural Network based Similarity Metric for Inertial Gesture Classification and Rejection - Archive ouverte HAL Access content directly
Conference Papers Year : 2015

Siamese Neural Network based Similarity Metric for Inertial Gesture Classification and Rejection

Abstract

In this paper, we tackle the task of symbolic gesture recognition using inertial MicroElectroMechanicals Systems (MEMS) present in Smartphones. We propose to build a non-linear similarity metric based on a Siamese Neural Network (SNN), trained using a new error function that models the relations between pairs of similar and dissimilar samples in order to structure the network output space. Experiments performed on different datasets regrouping up to 22 individuals and 18 gesture classes, targeting the most likely real case applications, show that this structure allows for an improved classification and a higher rejection quality over the conventional MultiLayer Perceptron (MLP) and Dynamic Time Warping (DTW) similarity metric.
Fichier principal
Vignette du fichier
berlemont_fg2015.pdf (375.5 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01179993 , version 1 (24-07-2015)

Identifiers

Cite

Samuel Berlemont, Grégoire Lefebvre, Stefan Duffner, Christophe Garcia. Siamese Neural Network based Similarity Metric for Inertial Gesture Classification and Rejection. International Conference on Automatic Face and Gesture Recognition, May 2015, Ljubljana, Slovenia. pp.1-6, ⟨10.1109/FG.2015.7163112⟩. ⟨hal-01179993⟩
334 View
1214 Download

Altmetric

Share

Gmail Facebook X LinkedIn More