read_image_batch

lib.image.read_image_batch(filenames: list[str], with_metadata: Literal[False] = False) → ndarray

lib.image.read_image_batch(filenames: list[str], with_metadata: Literal[True]) → tuple[ndarray, list[PNGHeader]]

Load a batch of images from the given file locations.

Leverages multi-threading to load multiple images from disk at the same time leading to vastly reduced image read times.

Parameters:

filenames (list[str]) – A of full paths to the images to be loaded.
with_metadata (bool) – Only returns a value if the images loaded are extracted Faceswap faces. If True then returns the Faceswap metadata stored within each Face’s .png exif header. Default: False

Returns:

batch – The batch of images in BGR channel order returned in the order of filenames
metadata – The faceswap metadata corresponding to each image in the batch. Only returned if with_metadata is True

Return type:

ndarray | tuple[ndarray, list[PNGHeader]]

Notes

As the images are compiled into a batch, they should be all of the same dimensions, otherwise a homogenous array will be returned

Example

>>> image_filenames = ["/path/to/image_1.png", "/path/to/image_2.png", "/path/to/image_3.png"]
>>> images = read_image_batch(image_filenames)
>>> print(images.shape)
... (3, 64, 64, 3)
>>> images, metadata = read_image_batch(image_filenames, with_metadata=True)
>>> print(images.shape)
... (3, 64, 64, 3)
>>> print(len(metadata))
... 3