Incremental learning of temporally-coherent Gaussian mixture models

Ognjen Arandjelovic*, Roberto Cipolla

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contribution

9 Citations (Scopus)

Abstract

In this paper we address the problem of learning Gaussian Mixture Models (GMMs) incrementally. Unlike previous approaches which universally assume that new data comes in blocks representable by GMMs which are then merged with the current model estimate, our method works for the case when novel data points arrive one- by-one, while requiring little additional memory. We keep only two GMMs in the memory and no historical data. The current fit is updated with the assumption that the number of components is fixed which is increased (or reduced) when enough evidence for a new component is seen. This is deducedfrom the change from the oldest fit of the same complexity, termed the Historical GMM, the concept of which is central to our method. The performance of the proposed method is demonstrated qualitatively and quantitatively on several synthetic data sets and video sequences of faces acquired in realistic imaging conditions.

Original languageEnglish
Title of host publicationTechnical Paper - Society of Manufacturing Engineers
VolumeTP06PUB22
Publication statusPublished - 2006
EventBritish Machine Vision Conference - Oxford, United Kingdom
Duration: 5 Sept 20058 Sept 2005

Conference

ConferenceBritish Machine Vision Conference
Country/TerritoryUnited Kingdom
CityOxford
Period5/09/058/09/05

Keywords

  • Density
  • Estimation
  • Gaussian
  • Incremental
  • Mixture
  • Temporal

Fingerprint

Dive into the research topics of 'Incremental learning of temporally-coherent Gaussian mixture models'. Together they form a unique fingerprint.

Cite this