tools.mask package

tools.mask.cli Module

Command Line Arguments for tools

class tools.mask.cli.MaskArgs(subparser: _SubParsersAction | None, command: str, description: str = 'default')

Class to parse the command line arguments for Mask tool

Parameters:

  • subparser (_SubParsersAction | None)

  • command (str)

  • description (str)

static get_argument_list()

Returns the argument list for the current command.

The argument list should be a list of dictionaries pertaining to each option for a command. This function should be overridden with the actual argument list for each command’s argument list.

See existing parsers for examples.

Returns:

The list of command line options for the given command

Return type:

list

static get_info()

Return command information

Classes

MaskArgs(subparser, command[, description])

Class to parse the command line arguments for Mask tool

Class Inheritance Diagram

Inheritance diagram of tools.mask.cli.MaskArgs

tools.mask.loader Module

Handles loading of faces/frames from source locations and pairing with alignments information

class tools.mask.loader.Loader(location: str, is_faces: bool)

Loader for reading source data from disk, and yielding the output paired with alignment information

Parameters:

  • location (str) – Full path to the source files location

  • is_faces (bool) – True if the source is a folder of faceswap extracted faces

add_alignments(alignments_object: Alignments | None) → None

Add the loaded alignments to _alignments for content matching

Parameters:

alignments_object (Alignments | None) – The alignments file object or None if not provided

Return type:

None

property file_list: list[str]

Full file list of source files to be loaded

property is_video: bool

True if the source is a video file otherwise False

load() → Generator[FrameFaces, None, None]

Load content from source and pair with corresponding alignment data

Yields:

The extract media object for the processed face

Return type:

Generator[FrameFaces, None, None]

property location: str

Full path to the source folder/video file location

property skip_count: int

The number of faces/frames that have been skipped due to no match in alignments file

Classes

Loader(location, is_faces)

Loader for reading source data from disk, and yielding the output paired with alignment information

tools.mask.mask Module

Tool to generate masks and previews of masks for existing alignments file

class tools.mask.mask.Mask(arguments: Namespace)

This tool is part of the Faceswap Tools suite and should be called from python tools.py mask command.

Faceswap Masks tool. Generate masks from existing alignments files, and output masks for preview.

Wrapper for the mask process to run in either batch mode or single use mode

Parameters:

arguments (Namespace) – The argparse arguments as passed in from tools.py

process() → None

The entry point for triggering the Extraction Process.

Should only be called from lib.cli.launcher.ScriptExecutor

Return type:

None

Classes

Mask(arguments)

This tool is part of the Faceswap Tools suite and should be called from python tools.py mask command.

tools.mask.mask_generate Module

Handles the generation of masks from faceswap for updating into an alignments file

class tools.mask.mask_generate.MaskGenerator(mask_type: str, update_all: bool, input_is_faces: bool, loader: loader.Loader, alignments: align.alignments.Alignments | None, input_location: str, config_file: str | None)

Uses faceswap’s extract pipeline to generate masks and update them into the alignments file and/or extracted face PNG Headers

Parameters:

  • mask_type (str) – The mask type to generate

  • update_all (bool) – True to update all faces, False to only update faces missing masks

  • input_is_faces (bool) – True if the input are faceswap extracted faces otherwise False

  • loader (loader.Loader) – The loader for loading source images/video from disk

  • config_file (str | None) – Full path to a custom config file to load. None for default config

  • alignments (align.alignments.Alignments | None)

  • input_location (str)

process() → T.Generator[FrameFaces, None, None]

Process the output from the extractor pipeline

Yields:

The FrameFaces object with updated masks

Return type:

T.Generator[FrameFaces, None, None]

Classes

MaskGenerator(mask_type, update_all, ...)

Uses faceswap's extract pipeline to generate masks and update them into the alignments file and/or extracted face PNG Headers

tools.mask.mask_import Module

Import mask processing for faceswap’s mask tool

class tools.mask.mask_import.Import(import_path: str, centering: CenteringType, storage_size: int, input_is_faces: bool, loader: loader.Loader, alignments: align.alignments.Alignments | None, input_location: str, mask_type: str)

Import masks from disk into an Alignments file

Parameters:

  • import_path (str) – The path to the input images

  • centering (CenteringType) – The centering to store the mask at

  • storage_size (int) – The size to store the mask at

  • input_is_faces (bool) – True if the input is aligned faces otherwise False

  • loader (loader.Loader) – The source file loader object

  • alignments (align.alignments.Alignments | None) – The alignments file object for the faces, if provided

  • mask_type (str) – The mask type to update to

  • input_location (str)

import_mask(media: FrameFaces) → None

Import the mask for the given Extract Media object

Parameters:

media (FrameFaces) – The extract media object containing the face(s) to import the mask for

Return type:

None

property skip_count: int

Number of masks that were skipped as they do not exist for given faces

property update_count: int

Number of masks that were skipped as they do not exist for given faces

Classes

Import(import_path, centering, storage_size, ...)

Import masks from disk into an Alignments file

tools.mask.mask_output Module

Output processing for faceswap’s mask tool

class tools.mask.mask_output.Output(arguments: Namespace, alignments: align.alignments.Alignments | None, file_list: list[str])

Handles outputting of masks for preview/editing to disk

Parameters:

  • arguments (Namespace) – The command line arguments that the mask tool was called with

  • alignments (align.alignments.Alignments | None) – The alignments file object (or None if not provided and input is faces)

  • file_list (list[str]) – Full file list for the loader. Used for extracting alignments from faces

close() → None

Shut down the image saver if it is open

Return type:

None

save(frame: str, idx: int, detected_face: DetectedFace, frame_dims: tuple[int, int] | None = None) → None

Build the mask preview image and save

Parameters:

  • frame (str) – The frame name in the alignments file

  • idx (int) – The index of the face for this frame in the alignments file

  • detected_face (DetectedFace) – A detected_face object for a face

  • frame_dims (tuple[int, int] | None) – The size of the original frame, if input is faces otherwise None. Default: None

Return type:

None

property should_save: bool

True if mask images should be output otherwise False

Classes

Output(arguments, alignments, file_list)

Handles outputting of masks for preview/editing to disk