Extract landmark coordinates — get

Extract the coordinates at landmark positions from shapes.

Usage

get_ldk(x, ldk = NULL, ..., .cols = NULL, .ldk_col = NULL, .name = NULL)

Arguments

x: A matrix (nx2), list of matrices, or tibble with coo columns.
ldk: Integer vector. Landmark indices (for single matrix or list).
...: Additional arguments (reserved for future use).
.cols: Column name(s) to process when x is a tibble. If NULL, automatically detects columns containing coo objects.
.ldk_col: Character. Name of landmark column. If NULL, uses colname_ldk.
.name: Character. Name for the output column when x is a tibble. If NULL, uses "colname_ldk_coords" (e.g., "coo_ldk_coords").

Value

If x is a single matrix: returns a matrix with landmark coordinates
If x is a list: returns a list of matrices with landmark coordinates
If x is a tibble: returns the tibble with new landmark coordinates column added

Details

For tibbles, this function automatically finds the landmark column using the naming convention colname_ldk (e.g., coo_ldk for a coo column).

Unlike other get_* functions, get_ldk() DOES add a column to tibbles for convenience, since landmark coordinates are commonly needed alongside the original coordinates.

Examples

# Single matrix with landmark indices
get_ldk(shapes$cat, ldk = c(1, 10, 50, 100))
#>      [,1] [,2]
#> [1,]  200   62
#> [2,]  182   42
#> [3,]  136   69
#> [4,]  231  126

# List with single landmark vector (applied to all)
get_ldk(shapes, ldk = c(1, 25, 50))
#> $cat
#>      [,1] [,2]
#> [1,]  200   62
#> [2,]  180   19
#> [3,]  136   69
#> 
#> $dog
#>      [,1] [,2]
#> [1,]  200  106
#> [2,]  112   81
#> [3,]  169  166
#> 
#> $heart
#>      [,1] [,2]
#> [1,]  200   37
#> [2,]  103  127
#> [3,]  160  212
#> 
#> $leaf2
#>      [,1] [,2]
#> [1,]  200   62
#> [2,]  133   72
#> [3,]  152  163
#> 

# Tibble - auto-detect landmark column and add coordinates
bot %>% get_ldk()
#> Warning: Landmark column 'coo_ldk' not found. Skipping.
#> # A tibble: 40 × 3
#>    coo       type   dummy
#>    <out>     <fct>  <fct>
#>  1 (138 x 2) whisky a    
#>  2 (168 x 2) whisky a    
#>  3 (189 x 2) whisky a    
#>  4 (129 x 2) whisky a    
#>  5 (152 x 2) whisky a    
#>  6 (161 x 2) whisky a    
#>  7 (124 x 2) whisky a    
#>  8 (126 x 2) whisky a    
#>  9 (183 x 2) whisky a    
#> 10 (193 x 2) whisky a    
#> # ℹ 30 more rows

# Tibble with custom output name
bot %>% get_ldk(.name = "landmarks")
#> Warning: Landmark column 'coo_ldk' not found. Skipping.
#> # A tibble: 40 × 3
#>    coo       type   dummy
#>    <out>     <fct>  <fct>
#>  1 (138 x 2) whisky a    
#>  2 (168 x 2) whisky a    
#>  3 (189 x 2) whisky a    
#>  4 (129 x 2) whisky a    
#>  5 (152 x 2) whisky a    
#>  6 (161 x 2) whisky a    
#>  7 (124 x 2) whisky a    
#>  8 (126 x 2) whisky a    
#>  9 (183 x 2) whisky a    
#> 10 (193 x 2) whisky a    
#> # ℹ 30 more rows