Line Feature based Multiple Video Synchronization

• Author(s) / Creator(s)

  • Lei, Cheng
  • Yang, Herb

• Technical report TR05-20. In this report, we present a novel method for synchronizing multiple (more than 2) un-calibrated video sequences recording the same event by free-moving full-perspective cameras. Unlike previous point feature based synchronization methods, our method takes advantage of line features for their better performance in measuring geometric alignment when video frames are synchronized. In particular, the tri-ocular geometric constraint of line features, which is evaluated by tri-focal transfer, is enforced when building the timeline maps for sequences to be synchronized. A hierarchical approach is used to reduce the computational complexity. To achieve sub-frame synchronization accuracy, a Levenberg-Marquardt method based optimization is performed to refine the synchronization. The experimental results on several synthetic and real video datasets demonstrate the effectiveness and robustness of our method over previous methods. | TRID-ID TR05-20

• Date created

  2005

• Subjects / Keywords

  • Trifocal tensor
  • Sub-frame synchronization
  • Video synchronization
  • Geometric alignment

• Type of Item

  Report

• DOI

  https://doi.org/10.7939/R3F47H09G

• License

  Attribution 3.0 International