Skip to content

PyFAME's Codebook

Here you can view PyFAME's complete codebook, containing detailed descriptions and metadata associated with all variables defined within the package.

Variable NameDescriptionSource File
FACE_OVAL_MASKA mask_type option for pyfame.mask_face_region(), specifying the standard full-face oval mask.predefined_constants.py
FACE_SKIN_MASKA mask_type option for pyfame.mask_face_region(), specifying the full-face oval minus the eyes and mouth.predefined_constants.py
EYES_MASKA mask_type option for pyfame.mask_face_region(), specifying a mask of both eyes and eyebrows.predefined_constants.py
IRISES_MASKA mask_type option for pyfame.mask_face_region(), specifying a mask of both eyes not including the eyebrows.predefined_constants.py
LIPS_MASKA mask_type option for pyfame.mask_face_region(), specifying a mask around the lips/mouth.predefined_constants.py
HEMI_FACE_LEFT_MASKA mask_type option for pyfame.mask_face_region(), specifying a left-half hemi-face mask.predefined_constants.py
HEMI_FACE_RIGHT_MASKA mask_type option for pyfame.mask_face_region(), specifying a right-half hemi-face mask.predefined_constants.py
HEMI_FACE_TOP_MASKA mask_type option for pyfame.mask_face_region(), specifying a top-half hemi-face mask.predefined_constants.py
HEMI_FACE_BOTTOM_MASKA mask_type option for pyfame.mask_face_region(), specifying a bottom-half hemi-face mask.predefined_constants.py
EYES_NOSE_MOUTH_MASKA mask_type option for pyfame.mask_face_region(), specifying a mask of the eyes, eyebrows, nose and mouth.predefined_constants.py
COLOR_REDA shift_color option for pyfame.face_color_shift(), specifying the color red.predefined_constants.py
COLOR_BLUEA shift_color option for pyfame.face_color_shift(), specifying the color blue.predefined_constants.py
COLOR_GREENA shift_color option for pyfame.face_color_shift(), specifying the color green.predefined_constants.py
COLOR_YELLOWA shift_color option for pyfame.face_color_shift(), specifying the color yellow.predefined_constants.py
COLOR_SPACE_BGRA color_space option for pyfame.extract_face_color_means, specifying the OpenCV default BGR color space.predefined_constants.py
COLOR_SPACE_HSVA color_space option for pyfame.extract_face_color_means, specifying the HSV color space.predefined_constants.py
COLOR_SPACE_GRAYSCALEA color_space option for pyfame.extract_face_color_means, specifying the grayscale color space.predefined_constants.py
OCCLUSION_FILL_BLACKAn occlusion_fill option for pyfame.occlude_face_region(), specifying for the occluded regions to be filled with black (0,0,0).predefined_constants.py
OCCLUSION_FILL_MEANAn occlusion_fill option for pyfame.occlude_face_region(), specifying for the occluded regions to be filled with the facial-skin's mean color.predefined_constants.py
OCCLUSION_FILL_BARAn occlusion_fill option for pyfame.occlude_face_region(), specifying for the occluded regions to be occluded by a horizontal bar.predefined_constants.py
BLUR_METHOD_AVERAGEA blur_method option for pyfame.blur_face_region(), specifying the average blurring method (cv2.blur()).predefined_constants.py
BLUR_METHOD_GAUSSIANA blur_method option for pyfame.blur_face_region(), specifying the gaussian blurring method (cv2.GaussianBlur()).predefined_constants.py
BLUR_METHOD_MEDIANA blur_method option for pyfame.blur_face_region(), specifying the median blurring method (cv2.medianBlur()).predefined_constants.py
NOISE_METHOD_PIXELATEA noise_method option for pyfame.apply_noise(), specifying for pixelation to be applied.predefined_constants.py
NOISE_METHOD_SALT_AND_PEPPERA noise_method option for pyfame.apply_noise(), specifying for salt and pepper noise to be applied.predefined_constants.py
NOISE_METHOD_GAUSSIANA noise_method option for pyfame.apply_noise(), specifying for Gaussian noise to be applied.predefined_constants.py
LOW_LEVEL_GRID_SCRAMBLEA scramble_method option for pyfame.facial_scramble(), specifying a grid-based scrambling of the face with semi-random horizontal shuffling of the grid squares. This scrambling method allows you to shuffle grid squares of the face while maintaining vertical positioning of facial landmarks.predefined_constants.py
HIGH_LEVEL_GRID_SCRAMBLEA scramble_method option for pyfame.facial_scramble(), specifying a grid-based scrambling of the face with fully-random shuffling of the grid squares.predefined_constants.py
LANDMARK_SCRAMBLEA scramble_method option for pyfame.facial_scramble(), specifying landmark positional scrambling on the face. The eyes and brows, nose, and mouth are cut out of the face and filled with the mean skin tone. Each landmark is then randomly reoriented and positioned over the underlying face.predefined_constants.py
SPARSE_OPTICAL_FLOWAn optical_flow_type option for pyfame.get_optical_flow(), specifying the use of the Lucas-Kanadae's sparse optical flow algorithm.predefined_constants.py
DENSE_OPTICAL_FLOWAn optical_flow_type option for pyfame.get_optical_flow(), specifying the use of Farneback's dense optical flow algorithm.predefined_constants.py
SHOW_HISTORY_ORIGINA history_mode option for pyfame.generate_point_light_display(), specifying that the history vectors should highlight each point's displacement from their origin.predefined_constants.py
SHOW_HISTORY_RELATIVEA history_mode option for pyfame.generate_point_light_display(), specifying that the history vectors should highlight each point's relative displacement history in the specified time window.predefined_constants.py
FRAME_SHUFFLE_RANDOMA shuffle_method option for pyfame.shuffle_frame_order() and pyfame.generate_shuffled_block_array(), specifying a fully-random shuffling of the input video's frame ordering.predefined_constants.py
FRAME_SHUFFLE_RANDOM_W_REPLACEMENTA shuffle_method option for pyfame.shuffle_frame_order() and pyfame.generate_shuffled_block_array(), specifying a random sampling with replacement of the input video's frame ordering.predefined_constants.py
FRAME_SHUFFLE_REVERSEA shuffle_method option for pyfame.shuffle_frame_order() and pyfame.generate_shuffled_block_array(), specifying a reversal of the input video's frame ordering.predefined_constants.py
FRAME_SHUFFLE_RIGHT_CYCLIC_SHIFTA shuffle_method option for pyfame.shuffle_frame_order() and pyfame.generate_shuffled_block_array(), specifying a right-cyclic rotation of the frame ordering within each block of frames.predefined_constants.py
FRAME_SHUFFLE_LEFT_CYCLIC_SHIFTA shuffle_method option for pyfame.shuffle_frame_order() and pyfame.generate_shuffled_block_array(), specifying a left-cyclic rotation of the frame ordering within each block of frames.predefined_constants.py
FRAME_SHUFFLE_PALINDROMEA shuffle_method option for pyfame.shuffle_frame_order() and pyfame.generate_shuffled_block_array(), specifying that for each block of frames, the block is reversed and then appended to the video adjacent to the original block. This results in each temporal segment of the video to play the same forwards and backwards, similar to how textual palindrome words are read.predefined_constants.py
FRAME_SHUFFLE_INTERLEAVEA shuffle_method option for pyfame.shuffle_frame_order() and pyfame.generate_shuffled_block_array(), specifying for each block of frames to be interleaved with one another.predefined_constants.py
NORMALIZE_IMAGES_PADA normalization_method option for pyfame.moviefy_images(), specifying for input image sizes to be normalized by padding each image to match the dimensions of the maximum image size provided.predefined_constants.py
NORMALIZE_IMAGES_CROPA normalization_method option for pyfame.moviefy_images(), specifying for input image sizes to be normalized by cropping each image to match the dimensions of the minimum image size provided.predefined_constants.py
FACE_OVAL_IDXA list containing the MediaPipe FaceMesh landmark indicies that encapsulate the face oval.landmarks.py
FACE_OVAL_TIGHT_IDXA list containing the MediaPipe FaceMesh landmark indicies that encapsulate a tighter face oval. This landmark set is provided as an alternative to FACE_OVAL_IDX for use with the pyfame.coloring family of functions. These landmarks are just inside the facial boundary, ensuring no background artifacts are included in the coloring manipulations.landmarks.py
LEFT_EYE_IDXA list containing the MediaPipe FaceMesh landmark indicies that encapsulate the left eye and brow.landmarks.py
RIGHT_EYE_IDXA list containing the MediaPipe FaceMesh landmark indicies that encapsulate the right eye and brow.landmarks.py
LEFT_IRIS_IDXA list containing the MediaPipe FaceMesh landmark indicies that encapsulate the left iris.landmarks.py
RIGHT_IRIS_IDXA list containing the MediaPipe FaceMesh landmark indicies that encapsulate the right iris.landmarks.py
NOSE_IDXA list containing the MediaPipe FaceMesh landmark indicies that encapsulate the nose.landmarks.py
MOUTH_IDXA list containing the MediaPipe FaceMesh landmark indicies that encapsulate the mouth region.landmarks.py
LIPS_IDXA list containing the MediaPipe FaceMesh landmark indicies that encapsulate specifically the upper and lower lips.landmarks.py
LEFT_CHEEK_IDXA list containing the MediaPipe FaceMesh landmark indicies that encapsulate the left cheek.landmarks.py
RIGHT_CHEEK_IDXA list containing the MediaPipe FaceMesh landmark indicies that encapsulate the right cheek.landmarks.py
CHIN_IDXA list containing the MediaPipe FaceMesh landmark indicies that encapsulate the chin.landmarks.py
HEMI_FACE_LEFT_IDXA list containing the MediaPipe FaceMesh landmark indicies that encapsulate the left half hemi-face.landmarks.py
HEMI_FACE_RIGHT_IDXA list containing the MediaPipe FaceMesh landmark indicies that encapsulate the right half hemi-face.landmarks.py
HEMI_FACE_TOP_IDXA list containing the MediaPipe FaceMesh landmark indicies that encapsulate the top half hemi-face.landmarks.py
HEMI_FACE_BOTTOM_IDXA list containing the MediaPipe FaceMesh landmark indicies that encapsulate the bottom half hemi-face.landmarks.py
FACE_OVAL_PATHA closed landmark path specifying the face oval as a region of interest. This path can be passed to almost all of the pyfame.core manipulations in order to define the region of application.landmarks.py
FACE_OVAL_TIGHT_PATHA closed landmark path specifying the tight face oval as a region of interest. This path can be passed to almost all of the pyfame.core manipulations in order to define the region of application.landmarks.py
LEFT_EYE_PATHA closed landmark path specifying the left eye as a region of interest. This path can be passed to almost all of the pyfame.core manipulations in order to define the region of application.landmarks.py
RIGHT_EYE_PATHA closed landmark path specifying the right eye as a region of interest. This path can be passed to almost all of the pyfame.core manipulations in order to define the region of application.landmarks.py
LEFT_IRIS_PATHA closed landmark path specifying the left iris as a region of interest. This path can be passed to almost all of the pyfame.core manipulations in order to define the region of application.landmarks.py
RIGHT_IRIS_PATHA closed landmark path specifying the right iris as a region of interest. This path can be passed to almost all of the pyfame.core manipulations in order to define the region of application.landmarks.py
NOSE_PATHA closed landmark path specifying the nose as a region of interest. This path can be passed to almost all of the pyfame.core manipulations in order to define the region of application.landmarks.py
MOUTH_PATHA closed landmark path specifying the mouth as a region of interest. This path can be passed to almost all of the pyfame.core manipulations in order to define the region of application.landmarks.py
LIPS_PATHA closed landmark path specifying the lips as a region of interest. This path can be passed to almost all of the pyfame.core manipulations in order to define the region of application.landmarks.py
HEMI_FACE_LEFT_PATHA closed landmark path specifying the left half hemi-face as a region of interest. This path can be passed to almost all of the pyfame.core manipulations in order to define the region of application.landmarks.py
HEMI_FACE_RIGHT_PATHA closed landmark path specifying the right half hemi-face as a region of interest. This path can be passed to almost all of the pyfame.core manipulations in order to define the region of application.landmarks.py
HEMI_FACE_TOP_PATHA closed landmark path specifying the top half hemi-face as a region of interest. This path can be passed to almost all of the pyfame.core manipulations in order to define the region of application.landmarks.py
HEMI_FACE_BOTTOM_PATHA closed landmark path specifying the bottom half hemi-face as a region of interest. This path can be passed to almost all of the pyfame.core manipulations in order to define the region of application.landmarks.py
CHEEKS_PATHA concave polygon path, required to be computed in place. Each concave path is initialized to a unique placeholder, which is used to identify the specific path and help compute its indicies internally. This path specifies both cheeks as a region of interest. This path can be passed to almost all of the pyfame.core manipulations in order to define the region of application.landmarks.py
LEFT_CHEEK_PATHA concave polygon path, required to be computed in place. Each concave path is initialized to a unique placeholder, which is used to identify the specific path and help compute its indicies internally. This path specifies the left cheek as a region of interest. This path can be passed to almost all of the pyfame.core manipulations in order to define the region of application.landmarks.py
RIGHT_CHEEK_PATHA concave polygon path, required to be computed in place. Each concave path is initialized to a unique placeholder, which is used to identify the specific path and help compute its indicies internally. This path specifies the right cheek as a region of interest. This path can be passed to almost all of the pyfame.core manipulations in order to define the region of application.landmarks.py
CHEEKS_NOSE_PATHA concave polygon path, required to be computed in place. Each concave path is initialized to a unique placeholder, which is used to identify the specific path and help compute its indicies internally. This path specifies both cheeks and the nose as a region of interest. This path can be passed to almost all of the pyfame.core manipulations in order to define the region of application.landmarks.py
BOTH_EYES_PATHA concave polygon path, required to be computed in place. Each concave path is initialized to a unique placeholder, which is used to identify the specific path and help compute its indicies internally. This path specifies both eyes as a region of interest. This path can be passed to almost all of the pyfame.core manipulations in order to define the region of application.landmarks.py
FACE_SKIN_PATHA concave polygon path, required to be computed in place. Each concave path is initialized to a unique placeholder, which is used to identify the specific path and help compute its indicies internally. This path specifies the facial skin (the face oval minus the eyes, brows and lips) as a region of interest. This path can be passed to almost all of the pyfame.core manipulations in order to define the region of application.landmarks.py
CHIN_PATHA concave polygon path, required to be computed in place. Each concave path is initialized to a unique placeholder, which is used to identify the specific path and help compute its indicies internally. This path specifies the chin as a region of interest. This path can be passed to almost all of the pyfame.core manipulations in order to define the region of application.landmarks.py