op_interpolate: Interpolate missing or low-confidence values in a dataset

In duet: Analysing Non-Verbal Communication in Dyadic Interactions from Video Data

Interpolate missing or low-confidence values in a dataset

Description

This function performs interpolation for x and y coordinate columns in a dataset based on confidence thresholds. It groups the data by person and region and uses spline interpolation to estimate missing or low-confidence values.

Usage

op_interpolate(
  data,
  confidence_threshold,
  missing = TRUE,
  treat_na_conf_as_low = FALSE
)

Arguments

data	A data frame containing x, y, confidence columns, and grouping columns (person, region).
confidence_threshold	A numeric value specifying the confidence threshold below which values will be interpolated.
missing	Logical. If TRUE, interpolate missing values (NA) in addition to low-confidence values.
treat_na_conf_as_low	Logical. If TRUE, treat NA in the confidence column as low confidence.

Value

A modified data frame with interpolated x and y values for low-confidence or missing rows.

Examples

# Load example data from the package
data_path <- system.file("extdata/csv_data/A-B_body_dyad.csv", package = "duet")
data <- read.csv(data_path)

# Interpolate missing or low-confidence values
result <- op_interpolate(
  data = data,
  confidence_threshold = 0.5,
  missing = TRUE,
  treat_na_conf_as_low = TRUE
)

print(result)

duet documentation built on April 3, 2025, 5:52 p.m.