View source on GitHub |
Class that performs face landmarks detection on images.
mp.tasks.vision.FaceLandmarker(
graph_config: mp.calculators.core.constant_side_packet_calculator_pb2.mediapipe_dot_framework_dot_calculator__pb2.CalculatorGraphConfig
,
running_mode: mp.tasks.vision.RunningMode
,
packet_callback: Optional[Callable[[Mapping[str, packet_module.Packet]], None]] = None
) -> None
Raises | |
---|---|
ValueError
|
The packet callback is not properly set based on the task's running mode. |
Methods
close
close() -> None
Shuts down the mediapipe vision task instance.
Raises | |
---|---|
RuntimeError
|
If the mediapipe vision task failed to close. |
convert_to_normalized_rect
convert_to_normalized_rect(
options: mp.tasks.vision.holistic_landmarker.image_processing_options_module.ImageProcessingOptions
,
image: mp.Image
,
roi_allowed: bool = True
) -> mp.tasks.components.containers.NormalizedRect
Converts from ImageProcessingOptions to NormalizedRect, performing sanity checks on-the-fly.
If the input ImageProcessingOptions is not present, returns a default NormalizedRect covering the whole image with rotation set to 0. If 'roi_allowed' is false, an error will be returned if the input ImageProcessingOptions has its 'region_of_interest' field set.
Args | |
---|---|
options
|
Options for image processing. |
image
|
The image to process. |
roi_allowed
|
Indicates if the region_of_interest field is allowed to be
set. By default, it's set to True.
|
Returns | |
---|---|
A normalized rect proto that represents the image processing options. |
create_from_model_path
@classmethod
create_from_model_path( model_path: str ) -> 'FaceLandmarker'
Creates an FaceLandmarker
object from a TensorFlow Lite model and the default FaceLandmarkerOptions
.
Note that the created FaceLandmarker
instance is in image mode, for
detecting face landmarks on single image inputs.
Args | |
---|---|
model_path
|
Path to the model. |
Returns | |
---|---|
FaceLandmarker object that's created from the model file and the
default FaceLandmarkerOptions .
|
Raises | |
---|---|
ValueError
|
If failed to create FaceLandmarker object from the
provided file such as invalid file path.
|
RuntimeError
|
If other types of error occurred. |
create_from_options
@classmethod
create_from_options( options:
mp.tasks.vision.FaceLandmarkerOptions
) -> 'FaceLandmarker'
Creates the FaceLandmarker
object from face landmarker options.
Args | |
---|---|
options
|
Options for the face landmarker task. |
Returns | |
---|---|
FaceLandmarker object that's created from options .
|
Raises | |
---|---|
ValueError
|
If failed to create FaceLandmarker object from
FaceLandmarkerOptions such as missing the model.
|
RuntimeError
|
If other types of error occurred. |
detect
detect(
image: mp.Image
,
image_processing_options: Optional[mp.tasks.vision.holistic_landmarker.image_processing_options_module.ImageProcessingOptions
] = None
) -> mp.tasks.vision.FaceLandmarkerResult
Performs face landmarks detection on the given image.
Only use this method when the FaceLandmarker is created with the image running mode.
The image can be of any size with format RGB or RGBA.
support is implemented.
Args | |
---|---|
image
|
MediaPipe Image. |
image_processing_options
|
Options for image processing. |
Returns | |
---|---|
The face landmarks detection results. |
Raises | |
---|---|
ValueError
|
If any of the input arguments is invalid. |
RuntimeError
|
If face landmarker detection failed to run. |
detect_async
detect_async(
image: mp.Image
,
timestamp_ms: int,
image_processing_options: Optional[mp.tasks.vision.holistic_landmarker.image_processing_options_module.ImageProcessingOptions
] = None
) -> None
Sends live image data to perform face landmarks detection.
The results will be available via the "result_callback" provided in the FaceLandmarkerOptions. Only use this method when the FaceLandmarker is created with the live stream running mode.
Only use this method when the FaceLandmarker is created with the live
stream running mode. The input timestamps should be monotonically increasing
for adjacent calls of this method. This method will return immediately after
the input image is accepted. The results will be available via the
result_callback
provided in the FaceLandmarkerOptions
. The
detect_async
method is designed to process live stream data such as
camera input. To lower the overall latency, face landmarker may drop the
input images if needed. In other words, it's not guaranteed to have output
per input image.
The result_callback
provides:
- The face landmarks detection results.
- The input image that the face landmarker runs on.
- The input timestamp in milliseconds.
Args | |
---|---|
image
|
MediaPipe Image. |
timestamp_ms
|
The timestamp of the input image in milliseconds. |
image_processing_options
|
Options for image processing. |
Raises | |
---|---|
ValueError
|
If the current input timestamp is smaller than what the face landmarker has already processed. |
detect_for_video
detect_for_video(
image: mp.Image
,
timestamp_ms: int,
image_processing_options: Optional[mp.tasks.vision.holistic_landmarker.image_processing_options_module.ImageProcessingOptions
] = None
) -> mp.tasks.vision.FaceLandmarkerResult
Performs face landmarks detection on the provided video frame.
Only use this method when the FaceLandmarker is created with the video running mode.
Only use this method when the FaceLandmarker is created with the video running mode. It's required to provide the video frame's timestamp (in milliseconds) along with the video frame. The input timestamps should be monotonically increasing for adjacent calls of this method.
Args | |
---|---|
image
|
MediaPipe Image. |
timestamp_ms
|
The timestamp of the input video frame in milliseconds. |
image_processing_options
|
Options for image processing. |
Returns | |
---|---|
The face landmarks detection results. |