Randomly apply a perspective transform to a dataset item

Applies a random perspective transformation to the image inside a dataset item with a given probability. The degree of distortion is controlled by distortion_scale. When a transform occurs, the same perspective mapping is applied to bounding boxes and masks so that the targets stay aligned with the image.

item_transform_random_perspective(
  x,
  distortion_scale = 0.5,
  p = 0.5,
  interpolation = 2,
  fill = 0
)

Arguments

x

A dataset item, typically an image_with_bounding_box or image_with_segmentation_mask object containing an image tensor and associated target data.

distortion_scale

(numeric): Argument to control the degree of distortion and ranges from 0 to 1. Default is 0.5.

p

(numeric): Probability of the image being transformed. Default is 0.5.

interpolation

(integer, optional): Desired interpolation. An integer 0 = nearest, 2 = bilinear, and 3 = bicubic or a name from magick::filter_types().

fill

(numeric vector or numeric): Pixel fill value for constant fill. Default is 0. If a numeric vector of length 3, it is used to fill R, G, B channels respectively. This value is only used when the padding_mode is constant. Only integer values are supported for Tensors.

Value

A dataset item of the same class. With probability p, the image and targets are perspective transformed; otherwise they are returned unchanged.

Details

The transformation is drawn again for every item, so that a dataset wrapped with this transform yields a different perspective on each access.

See also

Other item_random_transforms: item_transform_random_affine(), item_transform_random_crop(), item_transform_random_erasing(), item_transform_random_horizontal_flip(), item_transform_random_resize_crop(), item_transform_random_rotation(), item_transform_random_vertical_flip()

Examples

if (FALSE) { # \dontrun{
url <- "https://upload.wikimedia.org/wikipedia/commons/b/b6/Felis_catus-cat_on_snow.jpg"
img <- base_loader(url) |> transform_to_tensor()

boxes <- torch_tensor(matrix(c(600, 200, 2880, 1860), ncol = 4), dtype = torch_float32())

before <- list(x = img, y = list(boxes = boxes, labels = "cat"))
class(before) <- c("image_with_bounding_box", "list")

after <- item_transform_random_perspective(before)

before_plot <- draw_bounding_boxes(before, colors = "blue", width = 10)
after_plot <- draw_bounding_boxes(after, colors = "red", width = 10)
tensor_image_browse(before_plot)
tensor_image_browse(after_plot)
} # }