Browse by author
Lookup NU author(s): Professor Jonathon Chambers
Full text for this publication is not currently held within this repository. Alternative links are provided below where available.
© Springer International Publishing AG 2017. Sequential Monte Carlo probability hypothesis density (SMC-PHD) filtering has been recently exploited for audio-visual (AV) based tracking of multiple speakers, where audio data are used to inform the particle distribution and propagation in the visual SMC-PHD filter. However, the performance of the AV-SMC-PHD filter can be affected by the mismatch between the proposal and the posterior distribution. In this paper, we present a new method to improve the particle distribution where audio information (i.e. DOA angles derived from microphone array measurements) is used to detect new born particles and visual information (i.e. histograms) is used to modify the particles with particle flow (PF). Using particle flow has the benefit of migrating particles smoothly from the prior to the posterior distribution. We compare the proposed algorithm with the baseline AV-SMC-PHD algorithm using experiments on the AV16.3 dataset with multi-speaker sequences.
Author(s): Liu Y, Wang W, Chambers J, Kilic V, Hilton A
Publication type: Conference Proceedings (inc. Abstract)
Publication status: Published
Conference Name: 13th International Conference on Latent Variable Analysis and Signal Separation (LVA/ICA)
Year of Conference: 2017
Online publication date: 15/02/2017
Acceptance date: 02/04/2016
Publisher: Springer Verlag
Library holdings: Search Newcastle University Library for this item
Series Title: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)