Attention: This MediaPipe Solutions Preview is an early release. Learn more.

mp.tasks.vision.ObjectDetector

Stay organized with collections Save and categorize content based on your preferences.

Class that performs object detection on images.

The API expects a TFLite model with mandatory TFLite Model Metadata.

(kTfLiteUInt8/kTfLiteFloat32)

  • image input of size [batch x height x width x channels].
  • batch inference is not supported (batch is required to be 1).
  • only RGB inputs are supported (channels is required to be 3).
  • if type is kTfLiteFloat32, NormalizationOptions are required to be attached to the metadata for input normalization.

Output tensors must be the 4 outputs of a DetectionPostProcess op, i.e: (kTfLiteFloat32)

  • locations tensor of size [num_results x 4], the inner array representing bounding boxes in the form [top, left, right, bottom].
  • BoundingBoxProperties are required to be attached to the metadata and must specify type=BOUNDARIES and coordinate_type=RATIO. (kTfLiteFloat32)
  • classes tensor of size [num_results], each value representing the integer index of a class.
  • optional (but recommended) label map(s) can be attached as AssociatedFile-s with type TENSOR_VALUE_LABELS, containing one label per line. The first such AssociatedFile (if any) is used to fill the class_name field of the results. The display_name field is filled from the AssociatedFile (if any) whose locale matches the display_names_locale field of the ObjectDetectorOptions used at creation time ("en" by default, i.e. English). If none of these are available, only the index field of the results will be filled. (kTfLiteFloat32)
  • scores tensor of size [num_results], each value representing the score of the detected object.
  • optional score calibration can be attached using ScoreCalibrationOptions and an AssociatedFile with type TENSOR_AXIS_SCORE_CALIBRATION. See metadata_schema.fbs 1 for more details. (kTfLiteFloat32)
  • integer num_results as a tensor of size [1]

An example of such model can be found at: https://tfhub.dev/google/lite-model/object_detection/mobile_object_localizer_v1/1/metadata/1

graph_config The mediapipe vision task graph config proto.
running_mode The running mode of the mediapipe vision task.
packet_callback The optional packet callback for getting results asynchronously in the live stream mode.

ValueError The packet callback is not properly set based on the task's running mode.

Methods

close

View source

Shuts down the mediapipe vision task instance.

Raises
RuntimeError If the mediapipe vision task failed to close.

convert_to_normalized_rect

View source

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.
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 repesents the image processing options.

create_from_model_path

View source

Creates an ObjectDetector object from a TensorFlow Lite model and the default ObjectDetectorOptions.

Note that the created ObjectDetector instance is in image mode, for detecting objects on single image inputs.

Args
model_path Path to the model.

Returns
ObjectDetector object that's created from the model file and the default ObjectDetectorOptions.

Raises
ValueError If failed to create ObjectDetector object from the provided file such as invalid file path.
RuntimeError If other types of error occurred.

create_from_options

View source

Creates the ObjectDetector object from object detector options.

Args
options Options for the object detector task.

Returns
ObjectDetector object that's created from options.

Raises
ValueError If failed to create ObjectDetector object from ObjectDetectorOptions such as missing the model.
RuntimeError If other types of error occurred.

detect

View source

Performs object detection on the provided MediaPipe Image.

Only use this method when the ObjectDetector is created with the image running mode.

Args
image MediaPipe Image.

Returns
A detection result object that contains a list of detections, each detection has a bounding box that is expressed in the unrotated input frame of reference coordinates system, i.e. in [0,image_width) x [0, image_height), which are the dimensions of the underlying image data.

Raises
ValueError If any of the input arguments is invalid.
RuntimeError If object detection failed to run.

detect_async

View source

Sends live image data (an Image with a unique timestamp) to perform object detection.

Only use this method when the ObjectDetector 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 ObjectDetectorOptions. The detect_async method is designed to process live stream data such as camera input. To lower the overall latency, object detector 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 detection result object that contains a list of detections, each detection has a bounding box that is expressed in the unrotated input frame of reference coordinates system, i.e. in [0,image_width) x [0, image_height), which are the dimensions of the underlying image data.
  • The input image that the object detector runs on.
  • The input timestamp in milliseconds.

Args
image MediaPipe Image.
timestamp_ms The timestamp of the input image in milliseconds.

Raises
ValueError If the current input timestamp is smaller than what the object detector has already processed.

detect_for_video

View source

Performs object detection on the provided video frames.

Only use this method when the ObjectDetector 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.

Returns
A detection result object that contains a list of detections, each detection has a bounding box that is expressed in the unrotated input frame of reference coordinates system, i.e. in [0,image_width) x [0, image_height), which are the dimensions of the underlying image data.

Raises
ValueError If any of the input arguments is invalid.
RuntimeError If object detection failed to run.

__enter__

View source

Return self upon entering the runtime context.

__exit__

View source

Shuts down the mediapipe vision task instance on exit of the context manager.

Raises
RuntimeError If the mediapipe vision task failed to close.