# Copyright 2022 The MediaPipe Authors. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. """MediaPipe image segmenter task.""" import ctypes import dataclasses from typing import Callable, Optional from mediapipe.tasks.python.core import async_result_dispatcher from mediapipe.tasks.python.core import base_options as base_options_module from mediapipe.tasks.python.core import base_options_c from mediapipe.tasks.python.core import mediapipe_c_bindings from mediapipe.tasks.python.core import mediapipe_c_utils from mediapipe.tasks.python.core import serial_dispatcher from mediapipe.tasks.python.core.optional_dependencies import doc_controls from mediapipe.tasks.python.vision.core import image as image_module from mediapipe.tasks.python.vision.core import image_processing_options as image_processing_options_module from mediapipe.tasks.python.vision.core import image_processing_options_c from mediapipe.tasks.python.vision.core import vision_task_running_mode as running_mode_module _BaseOptions = base_options_module.BaseOptions _ImageProcessingOptions = image_processing_options_module.ImageProcessingOptions _RunningMode = running_mode_module.VisionTaskRunningMode _AsyncResultDispatcher = async_result_dispatcher.AsyncResultDispatcher class MpStringListC(ctypes.Structure): _fields_ = [ ('strings', ctypes.POINTER(ctypes.c_char_p)), ('num_strings', ctypes.c_int), ] class ImageSegmenterResultC(ctypes.Structure): _fields_ = [ ('confidence_masks', ctypes.POINTER(ctypes.c_void_p)), ('confidence_masks_count', ctypes.c_uint32), ('category_mask', ctypes.c_void_p), ('has_category_mask', ctypes.c_uint32), ('quality_scores', ctypes.POINTER(ctypes.c_float)), ('quality_scores_count', ctypes.c_uint32), ] _C_TYPES_RESULT_CALLBACK = ctypes.CFUNCTYPE( None, ctypes.c_int32, # MpStatus ctypes.POINTER(ImageSegmenterResultC), ctypes.c_void_p, # MpImage ctypes.c_int64, # timestamp_ms ) class ImageSegmenterOptionsC(ctypes.Structure): """C types for ImageSegmenterOptions.""" _fields_ = [ ('base_options', base_options_c.BaseOptionsC), ('running_mode', ctypes.c_int), ('display_names_locale', ctypes.c_char_p), ('output_confidence_masks', ctypes.c_bool), ('output_category_mask', ctypes.c_bool), ('result_callback', _C_TYPES_RESULT_CALLBACK), ] @classmethod @doc_controls.do_not_generate_docs def from_c_options( cls, base_options: base_options_c.BaseOptionsC, running_mode: _RunningMode, output_confidence_masks: bool, output_category_mask: bool, result_callback: _C_TYPES_RESULT_CALLBACK, ) -> 'ImageSegmenterOptionsC': """Creates an ImageSegmenterOptionsC object from the given options.""" empty_string = ctypes.c_char_p(b'') return cls( base_options=base_options, running_mode=running_mode.ctype, display_names_locale=empty_string, output_confidence_masks=output_confidence_masks, output_category_mask=output_category_mask, result_callback=result_callback, ) _CTYPES_SIGNATURES = ( mediapipe_c_utils.CStatusFunction( 'MpImageSegmenterCreate', ( ctypes.POINTER(ImageSegmenterOptionsC), ctypes.POINTER(ctypes.c_void_p), # Output param for segmenter ), ), mediapipe_c_utils.CStatusFunction( 'MpImageSegmenterSegmentImage', ( ctypes.c_void_p, ctypes.c_void_p, ctypes.POINTER(image_processing_options_c.ImageProcessingOptionsC), ctypes.POINTER(ImageSegmenterResultC), ), ), mediapipe_c_utils.CStatusFunction( 'MpImageSegmenterSegmentForVideo', ( ctypes.c_void_p, ctypes.c_void_p, ctypes.POINTER(image_processing_options_c.ImageProcessingOptionsC), ctypes.c_int64, ctypes.POINTER(ImageSegmenterResultC), ), ), mediapipe_c_utils.CStatusFunction( 'MpImageSegmenterSegmentAsync', ( ctypes.c_void_p, ctypes.c_void_p, ctypes.POINTER(image_processing_options_c.ImageProcessingOptionsC), ctypes.c_int64, ), ), mediapipe_c_utils.CStatusFunction( 'MpImageSegmenterGetLabels', ( ctypes.c_void_p, ctypes.POINTER(MpStringListC), ), ), mediapipe_c_utils.CFunction( 'MpImageSegmenterCloseResult', [ctypes.POINTER(ImageSegmenterResultC)], None, ), mediapipe_c_utils.CStatusFunction( 'MpImageSegmenterClose', (ctypes.c_void_p,), ), mediapipe_c_utils.CFunction( 'MpStringListFree', [ ctypes.c_void_p, ], None, ), ) @dataclasses.dataclass class ImageSegmenterResult: """Output result of ImageSegmenter. confidence_masks: multiple masks of float image where, for each mask, each pixel represents the prediction confidence, usually in the [0, 1] range. category_mask: a category mask of uint8 image where each pixel represents the class which the pixel in the original image was predicted to belong to. """ confidence_masks: Optional[list[image_module.Image]] = None category_mask: Optional[image_module.Image] = None @classmethod @doc_controls.do_not_generate_docs def from_ctypes( cls, c_result: ImageSegmenterResultC ) -> 'ImageSegmenterResult': """Creates an `ImageSegmenterResult` object from the given ctypes struct.""" confidence_masks = None category_mask = None if (c_result.confidence_masks_count > 0): confidence_masks = [ image_module.Image.create_from_ctypes(c_result.confidence_masks[i]) for i in range(c_result.confidence_masks_count) ] if c_result.has_category_mask > 0: category_mask = image_module.Image.create_from_ctypes( c_result.category_mask ) return cls( confidence_masks=confidence_masks, category_mask=category_mask, ) @dataclasses.dataclass class ImageSegmenterOptions: """Options for the image segmenter task. Attributes: base_options: Base options for the image segmenter task. running_mode: The running mode of the task. Default to the image mode. Image segmenter task has three running modes: 1) The image mode for segmenting objects on single image inputs. 2) The video mode for segmenting objects on the decoded frames of a video. 3) The live stream mode for segmenting objects on a live stream of input data, such as from camera. output_confidence_masks: Whether to output confidence masks. output_category_mask: Whether to output category mask. result_callback: The user-defined result callback for processing live stream data. The result callback should only be specified when the running mode is set to the live stream mode. """ base_options: _BaseOptions running_mode: _RunningMode = _RunningMode.IMAGE output_confidence_masks: bool = True output_category_mask: bool = False result_callback: Optional[ Callable[[ImageSegmenterResult, image_module.Image, int], None] ] = None class ImageSegmenter: """Class that performs image segmentation on images. The API expects a TFLite model with mandatory TFLite Model Metadata. Input tensor: (kTfLiteUInt8/kTfLiteFloat32) - image input of size `[batch x height x width x channels]`. - batch inference is not supported (`batch` is required to be 1). - RGB and greyscale inputs are supported (`channels` is required to be 1 or 3). - if type is kTfLiteFloat32, NormalizationOptions are required to be attached to the metadata for input normalization. Output tensors: (kTfLiteUInt8/kTfLiteFloat32) - list of segmented masks. - if `output_category_mask` is True, uint8 Image, Image vector of size 1. - if `output_confidence_masks` is True, float32 Image list of size `channels`. - batch is always 1 An example of such model can be found at: https://tfhub.dev/tensorflow/lite-model/deeplabv3/1/metadata/2 """ _lib: serial_dispatcher.SerialDispatcher _handle: ctypes.c_void_p _dispatcher: _AsyncResultDispatcher _async_callback: _C_TYPES_RESULT_CALLBACK def __init__( self, lib: serial_dispatcher.SerialDispatcher, handle: ctypes.c_void_p, dispatcher: _AsyncResultDispatcher, async_callback: _C_TYPES_RESULT_CALLBACK, ) -> None: """Initializes the `ImageSegmenter` object. Args: lib: The dispatch library to use for the image segmenter. handle: The C pointer to the image segmenter. dispatcher: The async result handler for the image segmenter. async_callback: The c callback for the image segmenter. """ self._lib = lib self._handle = handle self._dispatcher = dispatcher self._async_callback = async_callback self._labels = None @classmethod def create_from_model_path(cls, model_path: str) -> 'ImageSegmenter': """Creates an `ImageSegmenter` object from a TensorFlow Lite model and the default `ImageSegmenterOptions`. Note that the created `ImageSegmenter` instance is in image mode, for performing image segmentation on single image inputs. Args: model_path: Path to the model. Returns: `ImageSegmenter` object that's created from the model file and the default `ImageSegmenterOptions`. Raises: ValueError: If failed to create `ImageSegmenter` object from the provided file such as invalid file path. RuntimeError: If other types of error occurred. """ base_options = _BaseOptions(model_asset_path=model_path) options = ImageSegmenterOptions( base_options=base_options, running_mode=_RunningMode.IMAGE ) return cls.create_from_options(options) @classmethod def create_from_options( cls, options: ImageSegmenterOptions ) -> 'ImageSegmenter': """Creates the `ImageSegmenter` object from image segmenter options. Args: options: Options for the image segmenter task. Returns: `ImageSegmenter` object that's created from `options`. Raises: ValueError: If failed to create `ImageSegmenter` object from `ImageSegmenterOptions` such as missing the model. RuntimeError: If other types of error occurred. """ running_mode_module.validate_running_mode( options.running_mode, options.result_callback ) lib = mediapipe_c_bindings.load_shared_library(_CTYPES_SIGNATURES) def convert_result( c_result_ptr: ctypes.POINTER(ImageSegmenterResultC), image_ptr: ctypes.c_void_p, timestamp_ms: int, ) -> tuple[ImageSegmenterResult, image_module.Image, int]: """Converts the C data types to the desired Python data types.""" c_result = c_result_ptr[0] py_result = ImageSegmenterResult.from_ctypes(c_result) py_image = image_module.Image.create_from_ctypes(image_ptr) return (py_result, py_image, timestamp_ms) dispatcher = _AsyncResultDispatcher(converter=convert_result) c_callback = dispatcher.wrap_callback( options.result_callback, _C_TYPES_RESULT_CALLBACK ) c_options = ImageSegmenterOptionsC.from_c_options( base_options=options.base_options.to_ctypes(), running_mode=options.running_mode, output_confidence_masks=options.output_confidence_masks, output_category_mask=options.output_category_mask, result_callback=c_callback, ) segmenter_handle = ctypes.c_void_p() lib.MpImageSegmenterCreate( ctypes.byref(c_options), ctypes.byref(segmenter_handle) ) return cls( lib=lib, handle=segmenter_handle, dispatcher=dispatcher, async_callback=c_callback, ) def segment( self, image: image_module.Image, image_processing_options: Optional[_ImageProcessingOptions] = None, ) -> ImageSegmenterResult: """Performs the actual segmentation task on the provided MediaPipe Image. Args: image: MediaPipe Image. image_processing_options: Options for image processing. Returns: If the output_type is CATEGORY_MASK, the returned vector of images is per-category segmented image mask. If the output_type is CONFIDENCE_MASK, the returned vector of images contains only one confidence image mask. A segmentation result object that contains a list of segmentation masks as images. Raises: ValueError: If any of the input arguments is invalid. RuntimeError: If image segmentation failed to run. """ c_image = image._image_ptr # pylint: disable=protected-access c_result = ImageSegmenterResultC() options_c = ( ctypes.byref(image_processing_options.to_ctypes()) if image_processing_options else None ) self._lib.MpImageSegmenterSegmentImage( self._handle, c_image, options_c, ctypes.byref(c_result), ) result = ImageSegmenterResult.from_ctypes(c_result) self._lib.MpImageSegmenterCloseResult(ctypes.byref(c_result)) return result def segment_for_video( self, image: image_module.Image, timestamp_ms: int, image_processing_options: Optional[_ImageProcessingOptions] = None, ) -> ImageSegmenterResult: """Performs segmentation on the provided video frames. Only use this method when the ImageSegmenter 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: If the output_type is CATEGORY_MASK, the returned vector of images is per-category segmented image mask. If the output_type is CONFIDENCE_MASK, the returned vector of images contains only one confidence image mask. A segmentation result object that contains a list of segmentation masks as images. Raises: ValueError: If any of the input arguments is invalid. RuntimeError: If image segmentation failed to run. """ c_image = image._image_ptr # pylint: disable=protected-access c_result = ImageSegmenterResultC() options_c = ( ctypes.byref(image_processing_options.to_ctypes()) if image_processing_options else None ) self._lib.MpImageSegmenterSegmentForVideo( self._handle, c_image, options_c, timestamp_ms, ctypes.byref(c_result), ) result = ImageSegmenterResult.from_ctypes(c_result) self._lib.MpImageSegmenterCloseResult(ctypes.byref(c_result)) return result def segment_async( self, image: image_module.Image, timestamp_ms: int, image_processing_options: Optional[_ImageProcessingOptions] = None, ) -> None: """Sends live image data (an Image with a unique timestamp) to perform image segmentation. Only use this method when the ImageSegmenter 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 `ImageSegmenterOptions`. The `segment_async` method is designed to process live stream data such as camera input. To lower the overall latency, image segmenter may drop the input images if needed. In other words, it's not guaranteed to have output per input image. The `result_callback` prvoides: - A segmentation result object that contains a list of segmentation masks as images. - The input image that the image segmenter 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 image segmenter has already processed. """ c_image = image._image_ptr # pylint: disable=protected-access options_c = ( ctypes.byref(image_processing_options.to_ctypes()) if image_processing_options else None ) self._lib.MpImageSegmenterSegmentAsync( self._handle, c_image, options_c, timestamp_ms, ) @property def labels(self) -> list[str]: """Get the category label list the ImageSegmenter can recognize. For CATEGORY_MASK type, the index in the category mask corresponds to the category in the label list. For CONFIDENCE_MASK type, the output mask list at index corresponds to the category in the label list. If there is no label map provided in the model file, empty label list is returned. """ if not self._labels: c_labels = MpStringListC() self._lib.MpImageSegmenterGetLabels( self._handle, ctypes.byref(c_labels), ) self._labels = [] for i in range(c_labels.num_strings): c_label = c_labels.strings[i] label = ctypes.string_at(c_label).decode('utf-8') self._labels.append(label) self._lib.MpStringListFree(ctypes.byref(c_labels)) return self._labels def close(self): """Closes ImageSegmenter.""" if not self._handle: return self._lib.MpImageSegmenterClose(self._handle) self._handle = None self._dispatcher.close() self._lib.close() def __enter__(self): """Returns `self` upon entering the runtime context.""" return self def __exit__(self, exc_type, exc_value, traceback): """Closes ImageSegmenters and exits the context manager. Args: exc_type: The exception type that caused the context manager to exit. exc_value: The exception value that caused the context manager to exit. traceback: The exception traceback that caused the context manager to exit. Raises: RuntimeError: If the MediaPipe Gesture Recognizer task failed to close. """ self.close() def __del__(self): self.close()