sketchkit.cleanup package¶

Subpackages¶

sketchkit.cleanup.MasterSketch package

Submodules¶

sketchkit.cleanup.cleaner module¶

class sketchkit.cleanup.cleaner.Cleaner(method: str = 'MasterSketch', device: str = 'cuda')[source]¶

Bases: object

A class for rough sketch cleanup.

This class provides functionality to convert a raster rough sketch to a clean sketch.

method¶

The vectorization method that is used.

Type:: str

device¶

The device that is used.

Type:: str

run(input: ndarray, output_folder: str = 'cleanup') → ndarray[source]¶

Simplify the input raster image.

Parameters:: input (numpy.ndarray) – input raster image in shape (H, W), with values in [0, 255].
Returns:: an image in H*W in [0, 255].
Return type:: sketch (numpy.ndarray)

Module contents¶

class sketchkit.cleanup.Cleaner(method: str = 'MasterSketch', device: str = 'cuda')[source]¶

Bases: object

A class for rough sketch cleanup.

This class provides functionality to convert a raster rough sketch to a clean sketch.

method¶

The vectorization method that is used.

Type:: str

device¶

The device that is used.

Type:: str

run(input: ndarray, output_folder: str = 'cleanup') → ndarray[source]¶

Simplify the input raster image.

Parameters:: input (numpy.ndarray) – input raster image in shape (H, W), with values in [0, 255].
Returns:: an image in H*W in [0, 255].
Return type:: sketch (numpy.ndarray)

class sketchkit.cleanup.MasterSketch(device: str = 'cuda')[source]¶

Bases: object

A class for rough sketch cleanup using the method of paper “Mastering Sketching: Adversarial Augmentation for Structured Prediction” in SIGGRAPH 2018.

device¶

The device that is used.

Type:: str

_download_models()[source]¶: Download the pretrained models

run(image_array: ndarray, image_folder: str = './', out_folder: str = 'cleanup') → ndarray[source]¶

Sketch simplification for the input raster image.

Parameters:

image_array (numpy.ndarray) – input raster image in shape (H, W), with values in [0, 255].
image_folder (str, optional) – the folder of the input raster image.
out_folder (str, optional) – an output folder to store the output file.

Returns:

a sketch image in H*W in [0, 255].

Return type:

pred_out (numpy.ndimage)