eduzhai > Applied Sciences > Engineering >

Automatic Quality Assessment for Audio-Visual Verification Systems The LOVe submission to NIST SRE Challenge 2019

  • king
  • (0) Download
  • 20210506
  • Save

... pages left unread,continue reading

Document pages: 5 pages

Abstract: Fusion of scores is a cornerstone of multimodal biometric systems composed ofindependent unimodal parts. In this work, we focus on quality-dependent fusionfor speaker-face verification. To this end, we propose a universal model whichcan be trained for automatic quality assessment of both face and speakermodalities. This model estimates the quality of representations produced byunimodal systems which are then used to enhance the score-level fusion ofspeaker and face verification modules. We demonstrate the improvements broughtby this quality-dependent fusion on the recent NIST SRE19 Audio-VisualChallenge dataset.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...