Transductive parameter transfer, bags of dense trajectories and MILES for no-audio multimodal speech detection

Laura Cabrera-Quiros, Ekin Gedik, Hayley Hung

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

9 Downloads (Pure)


This paper presents the algorithms that task organisers deployed for the automatic Human Behaviour Analysis (HBA) task of the MediaEval 2018. HBA task aims to investigate alternate modalities of video and body-worn acceleration for the detection of speaking status. For unimodal estimation from acceleration, a transfer learning approach, Transductive Parameter Transfer (TPT), which is shown to perform satisfactorily in a similar setting[4] is employed. For the estimation from the video modality, bags of Dense Trajectories were used in a multiple instance learning approach (MILES) [2]. Finally, late fusion is used for combining the outputs from both modalities. The multi-modal approach resulted in a mean AUC of 0.658, outperforming the performance of both single modality approaches. Copyright held by the owner/author(s).

Original languageEnglish
Title of host publicationMediaEval 2018 Multimedia Benchmark Workshop
Subtitle of host publicationWorking Notes Proceedings of the MediaEval 2018 Workshop Sophia Antipolis, France, 29-31 October 2018
EditorsMartha Larson, Claire-Hélène Demarty, Piyush Arora
Number of pages3
Publication statusPublished - 1 Jan 2018
Externally publishedYes
Event2018 MediaEval Workshop - Sophia Antipolis, France
Duration: 29 Oct 201831 Oct 2018

Publication series

NameCEUR Workshop Proceedings
ISSN (Print)1613-0073


Conference2018 MediaEval Workshop
CitySophia Antipolis


Dive into the research topics of 'Transductive parameter transfer, bags of dense trajectories and MILES for no-audio multimodal speech detection'. Together they form a unique fingerprint.

Cite this