Synchronizing multimodal data streams acquired using commodity hardware

  • Authors:

  • Martial Michel;Vincent Stanford

  • Affiliations:

  • Systems Plus, Inc., Rockville, MD and National Institute of Standards and Technology, Gaithersburg, MD;National Institute of Standards and Technology, Gaithersburg, MD

  • Venue:
  • Proceedings of the 4th ACM international workshop on Video surveillance and sensor networks
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

We have developed tools and techniques that allow video frame level synchronization of multiple free-running commodity video cameras,microphones,and computer nodes using non-realtime operating systems.The techniques rely on physical audiovisual synchronization pulses,statistical procedures to correlate and interpolate the multiple timestamp streams,and software tools for review to produce smoothed and drift-corrected timestamp streams in our multimodal corpora. In this article we present those techniques and tools. Our project is open source and we are seeking collaborative developers for future work.