Skip to contents

Fill gaps in motion capture data

Source: R/mocap-gaps.R
fillGaps.Rd

Fills NA gaps in one or more position assays of a PhysioExperiment object using interpolation. Gaps larger than max_gap are left as NA.

Usage

fillGaps(pe, method = c("spline", "linear"), max_gap = 50, assay_names = NULL)

Arguments

pe

A PhysioExperiment object

method

Interpolation method: "linear" uses approx, "spline" uses smooth.spline

max_gap

Maximum gap size (in samples) to fill. Gaps larger than this are left as NA. Set to Inf to fill all gaps.

assay_names

Character vector of assay names to fill. If NULL (default), fills all position assays that are present (position_x, position_y, position_z, keypoint_x, keypoint_y).

Value

PhysioExperiment with filled assays (modified in place)

References

Federolf PA (2013). "A novel approach to solve the 'missing marker problem' in marker-based motion analysis that does not require additional assumptions about the biodynamic model." Journal of Biomechanics, 46(13), 2173-2178.

See also

detectGaps(), reportGaps(), fillGapsLinear(), fillGapsSpline()

Examples

pe <- PhysioCore::PhysioExperiment(
  assays = S4Vectors::SimpleList(
    position_x = matrix(c(1, NA, NA, 4, 5), ncol = 1),
    position_y = matrix(c(10, NA, NA, 40, 50), ncol = 1)
  ),
  colData = S4Vectors::DataFrame(label = "M1", type = "marker"),
  samplingRate = 120
)
pe_filled <- fillGaps(pe, method = "linear", max_gap = 50)