uki(ldZddlZddlZddlmZmZddlmZddlmZddl m Z ddl m Z ddl mZdd lmZdd lmZdd lmZdd lmZdd lmZddlmZddlmZddlmZej@Z!ejDZ#ejHZ%e jLZ'ejPZ)ejTZ+ejXZ-dZ.dZ/dZ0dZ1dZ2dZ3dZ4dZ5dZ6dZ7dZ8dZ9ejtGddZ;ejtGdd ZGd&d'ej~Z@y)(z%MediaPipe interactive segmenter task.N)ListOptional)packet_creator) packet_getter)image)!image_segmenter_graph_options_pb2)segmenter_options_pb2)keypoint base_options) task_info) doc_controls)base_vision_task_api)image_processing_options)vision_task_running_mode)render_data_pb2confidence_masksCONFIDENCE_MASKS category_mask CATEGORY_MASKimage_in image_outroi_inROI norm_rect_in NORM_RECTIMAGEzFmediapipe.tasks.vision.interactive_segmenter.InteractiveSegmenterGraphcheZdZUdZdZeeeje d<dZ eeje d<y)InteractiveSegmenterResulta]Output result of InteractiveSegmenter. 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. Nrr) __name__ __module__ __qualname____doc__rrr image_moduleImage__annotations__rn/home/cdr/jupyterlab/.venv/lib/python3.12/site-packages/mediapipe/tasks/python/vision/interactive_segmenter.pyrr<s9:>HT,"4"456=04-,,,-4r(rcbeZdZUdZeed<dZeed<dZeed<e jde fdZ y ) InteractiveSegmenterOptionszOptions for the interactive segmenter task. Attributes: base_options: Base options for the interactive segmenter task. output_confidence_masks: Whether to output confidence masks. output_category_mask: Whether to output category mask. r Toutput_confidence_masksFoutput_category_maskreturncr|jj}d|_t}t ||S)z8Generates an ImageSegmenterGraphOptions protobuf object.F)r segmenter_options)r to_pb2use_stream_mode_SegmenterOptionsProto _ImageSegmenterGraphOptionsProto)selfbase_options_protosegmenter_options_protos r)r1z"InteractiveSegmenterOptions.to_pb2Ys>**113).&46 +'1 r(N) r r!r"r# _BaseOptionsr&r,boolr-rdo_not_generate_docsr4r1r'r(r)r+r+KsE"&4&$$$$6%r(r+cneZdZUdZGddej Zeed<dZe e jed<y)RegionOfInterestz.The Region-Of-Interest (ROI) to interact with.ceZdZdZdZy)RegionOfInterest.FormatrN)r r!r" UNSPECIFIEDKEYPOINTr'r(r)Formatr>is KHr(rBformatNr ) r r!r"r#enumEnumrBr&r rkeypoint_moduleNormalizedKeypointr'r(r)r<r<es16tyy .;?(H_77 8?r(r<roir.ct}||jtjjk(r t d|jtjj k(r|j|jj}d|j_ |j}d|_ |jj|_|jj|_|St dt d)z1Converts region of interest to render data proto.z&RegionOfInterest format not specified.Tz7Please specify the Region-of-interest for segmentation.zUnrecognized format.)_RenderDataProtorCr<rBr@ ValueErrorrAr render_annotationsaddcolorrpoint normalizedxy)rHresult annotationrQs r)_convert_roi_to_render_datarWqs  &_ zz%,,888 ? @@ '..77 7  !..224    ,,..,,.. N OO)**r(c zeZdZdZededdfdZededdfdZ d de jd e d e e defd Zy) InteractiveSegmenteraKClass that performs interactive segmentation on images. Users can represent user interaction through `RegionOfInterest`, which gives a hint to InteractiveSegmenter to perform segmentation focusing on the given region of interest. 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_type` is CATEGORY_MASK, uint8 Image, Image vector of size 1. - if `output_type` is CONFIDENCE_MASK, 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 model_pathr.cTt|}t|}|j|S)aCreates an `InteractiveSegmenter` object from a TensorFlow Lite model and the default `InteractiveSegmenterOptions`. Note that the created `InteractiveSegmenter` instance is in image mode, for performing image segmentation on single image inputs. Args: model_path: Path to the model. Returns: `InteractiveSegmenter` object that's created from the model file and the default `InteractiveSegmenterOptions`. Raises: ValueError: If failed to create `InteractiveSegmenter` object from the provided file such as invalid file path. RuntimeError: If other types of error occurred. )model_asset_pathr )r8r+create_from_options)clsrZr optionss r)create_from_model_pathz+InteractiveSegmenter.create_from_model_paths)&  & I ''U'C  r(NrrHrc |j||d}t|}|jtt j |t t j|tt j|ji}t}t|vr!tj|t|_t|vr!tj |t|_|S)aPerforms the actual segmentation task on the provided MediaPipe Image. The image can be of any size with format RGB. Args: image: MediaPipe Image. roi: Optional user-specified region of interest for segmentation. 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. F) roi_allowed)convert_to_normalized_rectrW_process_image_datarrr create_imagert create_protorvr1rrmrget_image_listrro get_imager)r5rrHrnormalized_rectrender_data_protooutput_packetssegmentation_results r)segmentzInteractiveSegmenter.segments455 %U6O4C8--~::5A.556GH ; ;  " " $! /N56$6-:-I-I 6 7.*"^3*7*A*A 3 4+' r()N)r r!r"r# classmethodstrr`r+r]r$r%r<r_ImageProcessingOptionsrrr'r(r)rYrYs8,c,6L,,,-/---fEI 1   1 1!))@ A 1 " 1r(rY)Ar# dataclassesrDtypingrrmediapipe.pythonrr$mediapipe.python._framework_bindingsrr$/mediapipe.tasks.cc.vision.image_segmenter.protorr ,mediapipe.tasks.python.components.containersr rFmediapipe.tasks.python.corer base_options_moduler task_info_module1mediapipe.tasks.python.core.optional_dependenciesr"mediapipe.tasks.python.vision.corerrimage_processing_options_modulermediapipe.utilr BaseOptionsr8 RenderDatarKSegmenterOptionsr3ImageSegmenterGraphOptionsr4VisionTaskRunningModerxImageProcessingOptionsrTaskInforprmrlrornrrrjrtrsrvrurirq dataclassrr+r<rWBaseVisionTaskApirYr'r(r)rs`, !+*F]QTKEJCjG*".. "--.??%@@!(== 9PP  % % 2*,$"$ '  L   5 5 5 2 @@@+%5+:J+.U/AAUr(