Detecting Low-Quality Reference Time Series in Stream Recognition - Archive ouverte HAL
Communication Dans Un Congrès Année : 2016

Detecting Low-Quality Reference Time Series in Stream Recognition

Résumé

On-line supervised spotting and classification of subsequences can be performed by comparing some distance between the stream and previously learnt time series. However, learning a few incorrect time series can trigger disproportionately many false alarms. In this paper, we propose a fast technique to prune bad instances away and automatically select appropriate distance thresholds. Our main contribution is to turn the ill-defined spotting problem into a collection of single well-defined binary classification problems, by segmenting the stream and by ranking subsets of instances on those segments very quickly. We further demonstrate our technique's effectiveness on a gesture recognition application.
Fichier principal
Vignette du fichier
ICPR16_1388_FI.pdf (1.58 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01435197 , version 1 (13-01-2017)

Identifiants

  • HAL Id : hal-01435197 , version 1

Citer

Marc Dupont, Pierre-François Marteau, Nehla Ghouaiel. Detecting Low-Quality Reference Time Series in Stream Recognition. International Conference on Pattern Recognition (ICPR), IAPR, Dec 2016, Cancun, Mexico. ⟨hal-01435197⟩
310 Consultations
186 Téléchargements

Partager

More