Consistency checks for particle filters with application to image stabilization

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademic

    32 Downloads (Pure)


    An ‘inconsistent’ particle filter produces – in a statistical sense – larger estimation errors than predicted by the model on which the filter is based. Inconsistent behavior of a particle filter can be detected online by checking whether the predicted measurements (derived from the particles that represent the one-step-ahead prediction pdf) comply in a statistical sense with the observed measurements. This principle is demonstrated in an image stabilization application. We consider an image sequence of a scene consisting of a dynamic foreground and a static background. The motion of the camera (slow rotations and zooming) is modeled with an 8-dim state vector describing a projective geometrical transformation that, inversely applied to the current frame, compensates the camera motion. The dynamics of the state vector is modeled as a first order AR process. The measurements of the system are corner points (detected in the first frame) that are tracked. The particle filtering estimates the state vector using the measurements. However, the filter behaves inconsistently because a few corner points belong to the foreground. Using inconsistency checks these foreground points are detected and removed from the list of measurements.
    Original languageUndefined
    Title of host publicationProceedings of the First Annual IEEE BENELUX/DSP Valley Signal Processing Symposium, SPS-DARTS 2005
    Place of PublicationAntwerp, Belgium
    PublisherIEEE Benelux Chapter Signal Processing
    Number of pages4
    ISBN (Print)not assigned
    Publication statusPublished - 2005
    Event1st Annual IEEE BENELUX/DSP Valley Signal Processing Symposium, SPS-DARTS 2005 - Antwerp, Belgium
    Duration: 19 Apr 200520 Apr 2005
    Conference number: 1

    Publication series

    PublisherIEEE Benelux Chapter Signal Processing
    VolumePaper 178


    Conference1st Annual IEEE BENELUX/DSP Valley Signal Processing Symposium, SPS-DARTS 2005
    Abbreviated titleSPS-DARTS


    • METIS-227797
    • EWI-19134
    • IR-75561

    Cite this