viam.services.vision.vision

Classes

`CaptureAllResult`	CaptureAllResult represents the collection of things that you have requested from the
`Vision`	Vision represents a Vision service.

Module Contents

class viam.services.vision.vision.CaptureAllResult(image: viam.media.video.ViamImage | None = None, classifications: List[viam.proto.service.vision.Classification] | None = None, detections: List[viam.proto.service.vision.Detection] | None = None, objects: List[viam.proto.common.PointCloudObject] | None = None, extra: Mapping[str, viam.utils.ValueTypes] | None = None)[source]: CaptureAllResult represents the collection of things that you have requested from the CaptureAllFromCamera method. This is used most often for visualization purposes, since normally, returning the image on every call to a classifier/detector/etc would be costly and unnecessary. The default result for each field is None rather than the empty list to distinguish between “there was no request for the classifier/detector to return a result” vs. “the classifier/detector was requested, but there were no results”.

class viam.services.vision.vision.Vision(name: str)[source]

Bases: viam.services.service_base.ServiceBase

Vision represents a Vision service.

This acts as an abstract base class for any drivers representing specific vision implementations. This cannot be used on its own. If the __init__() function is overridden, it must call the super().__init__() function.

For more information, see Computer Vision service.

SUBTYPE: Final

Properties: TypeAlias = GetPropertiesResponse: Properties is a class that states what features are supported on the associated vision service. Currently, these are the following properties: - classifications_supported (bool): GetClassifications and GetClassificationsFromCamera are implemented. - detections_supported (bool): GetDetections and GetDetectionsFromCamera are implemented. - object_point_clouds_supported (bool): GetObjectPointClouds is implemented.

abstract capture_all_from_camera(camera_name: str, return_image: bool = False, return_classifications: bool = False, return_detections: bool = False, return_object_point_clouds: bool = False, *, extra: Mapping[str, viam.utils.ValueTypes] | None = None, timeout: float | None = None) → CaptureAllResult[source]

Async:

Get the next image, detections, classifications, and objects all together, given a camera name. Used for visualization.

camera_name = "cam1"

# Grab the detector you configured on your machine
my_detector = VisionClient.from_robot(robot, "my_detector")

# capture all from the next image from the camera
result = await my_detector.capture_all_from_camera(
    camera_name,
    return_image=True,
    return_detections=True,
)

Parameters:

camera_name (str) – The name of the camera to use for detection
return_image (bool) – Ask the vision service to return the camera’s latest image
return_classifications (bool) – Ask the vision service to return its latest classifications
return_detections (bool) – Ask the vision service to return its latest detections
return_object_point_clouds (bool) – Ask the vision service to return its latest 3D segmentations

Returns:

A class that stores all potential returns from the vision service. It can return the image from the camera along with its associated detections, classifications, and objects, as well as any extra info the model may provide.

Return type:

vision.CaptureAllResult

For more information, see Computer Vision service.

abstract get_detections_from_camera(camera_name: str, *, extra: Mapping[str, viam.utils.ValueTypes] | None = None, timeout: float | None = None) → List[viam.proto.service.vision.Detection][source]

Async:

Get a list of detections in the next image given a camera and a detector

camera_name = "cam1"

# Grab the detector you configured on your machine
my_detector = VisionClient.from_robot(robot, "my_detector")

# Get detections from the next image from the camera
detections = await my_detector.get_detections_from_camera(camera_name)

Parameters:: camera_name (str) – The name of the camera to use for detection
Raises:: ViamError – Raised if given an image without a specified width and height
Returns:: A list of 2D bounding boxes, their labels, and the confidence score of the labels, around the found objects in the next 2D image from the given camera, with the given detector applied to it.
Return type:: List[viam.proto.service.vision.Detection]

For more information, see Computer Vision service.

abstract get_detections(image: viam.media.video.ViamImage, *, extra: Mapping[str, viam.utils.ValueTypes] | None = None, timeout: float | None = None) → List[viam.proto.service.vision.Detection][source]

Async:

Get a list of detections in the given image using the specified detector

# Grab camera from the machine
cam1 = Camera.from_robot(robot, "cam1")

# Get the detector you configured on your machine
my_detector = VisionClient.from_robot(robot, "my_detector")

# Get an image from the camera
img = await cam1.get_image()

# Get detections from that image
detections = await my_detector.get_detections(img)

Parameters:: image (Image | RawImage) – The image to get detections from
Raises:: ViamError – Raised if given an image without a specified width and height
Returns:: A list of 2D bounding boxes, their labels, and the confidence score of the labels, around the found objects in the next 2D image from the given camera, with the given detector applied to it.
Return type:: List[viam.proto.service.vision.Detection]

For more information, see Computer Vision service.

abstract get_classifications_from_camera(camera_name: str, count: int, *, extra: Mapping[str, viam.utils.ValueTypes] | None = None, timeout: float | None = None) → List[viam.proto.service.vision.Classification][source]

Async:

Get a list of classifications in the next image given a camera and a classifier

camera_name = "cam1"

# Grab the classifier you configured on your machine
my_classifier = VisionClient.from_robot(robot, "my_classifier")

# Get the 2 classifications with the highest confidence scores from the next image from the camera
classifications = await my_classifier.get_classifications_from_camera(
    camera_name, 2)

Parameters:

camera_name (str) – The name of the camera to use for detection
count (int) – The number of classifications desired

Returns:

The list of Classifications

Return type:

List[viam.proto.service.vision.Classification]

For more information, see Computer Vision service.

abstract get_classifications(image: viam.media.video.ViamImage, count: int, *, extra: Mapping[str, viam.utils.ValueTypes] | None = None, timeout: float | None = None) → List[viam.proto.service.vision.Classification][source]

Async:

Get a list of classifications in the given image using the specified classifier

# Grab camera from the machine
cam1 = Camera.from_robot(robot, "cam1")

# Get the classifier you configured on your machine
my_classifier = VisionClient.from_robot(robot, "my_classifier")

# Get an image from the camera
img = await cam1.get_image()

# Get the 2 classifications with the highest confidence scores
classifications = await my_classifier.get_classifications(img, 2)

Parameters:

image (Image | RawImage) – The image to get detections from
count (int) – The number of classifications desired

Returns:

The list of Classifications

Return type:

List[viam.proto.service.vision.Classification]

For more information, see Computer Vision service.

abstract get_object_point_clouds(camera_name: str, *, extra: Mapping[str, viam.utils.ValueTypes] | None = None, timeout: float | None = None) → List[viam.proto.common.PointCloudObject][source]

Async:

Returns a list of the 3D point cloud objects and associated metadata in the latest picture obtained from the specified 3D camera (using the specified segmenter).

To deserialize the returned information into a numpy array, use the Open3D library.

import numpy as np
import open3d as o3d

# Grab the 3D camera from the machine
cam1 = Camera.from_robot(robot, "cam1")
# Grab the object segmenter you configured on your machine
my_segmenter = VisionClient.from_robot(robot, "my_segmenter")
# Get the objects from the camera output
objects = await my_segmenter.get_object_point_clouds(cam1)
# write the first object point cloud into a temporary file
with open("/tmp/pointcloud_data.pcd", "wb") as f:
    f.write(objects[0].point_cloud)
pcd = o3d.io.read_point_cloud("/tmp/pointcloud_data.pcd")
points = np.asarray(pcd.points)

Parameters:: camera_name (str) – The name of the camera
Returns:: The pointcloud objects with metadata
Return type:: List[viam.proto.common.PointCloudObject]

For more information, see Computer Vision service.

abstract get_properties(*, extra: Mapping[str, viam.utils.ValueTypes] | None = None, timeout: float | None = None) → Properties[source]

Async:

Get info about what vision methods the vision service provides. Currently returns boolean values that state whether the service implements the classification, detection, and/or 3D object segmentation methods.

# Grab the detector you configured on your machine
my_detector = VisionClient.from_robot(robot, "my_detector")
properties = await my_detector.get_properties()
properties.detections_supported      # returns True
properties.classifications_supported # returns False

Returns:: The properties of the vision service
Return type:: Properties

For more information, see Computer Vision service.

classmethod from_robot(robot: viam.robot.client.RobotClient, name: str) → typing_extensions.Self

Get the service named name from the provided robot.

async def connect() -> RobotClient:
    # Replace "<API-KEY>" (including brackets) with your API key and "<API-KEY-ID>" with your API key ID
    options = RobotClient.Options.with_api_key("<API-KEY>", "<API-KEY-ID>")
    # Replace "<MACHINE-URL>" (included brackets) with your machine's connection URL or FQDN
    return await RobotClient.at_address("<MACHINE-URL>", options)

async def main():
    robot = await connect()

    # Can be used with any resource, using the motion service as an example
    motion = MotionClient.from_robot(robot=robot, name="builtin")

    robot.close()

Parameters:

robot (RobotClient) – The robot
name (str) – The name of the service

Returns:

The service, if it exists on the robot

Return type:

Self

abstract do_command(command: Mapping[str, viam.utils.ValueTypes], *, timeout: float | None = None, **kwargs) → Mapping[str, viam.utils.ValueTypes]

Async:

Send/receive arbitrary commands.

service = SERVICE.from_robot(robot, "builtin")  # replace SERVICE with the appropriate class

my_command = {
  "cmnd": "dosomething",
  "someparameter": 52
}

# Can be used with any resource, using the motion service as an example
await service.do_command(command=my_command)

Parameters:: command (Dict[str, ValueTypes]) – The command to execute
Returns:: Result of the executed command
Return type:: Dict[str, ValueTypes]

classmethod get_resource_name(name: str) → viam.proto.common.ResourceName

Get the ResourceName for this Resource with the given name

# Can be used with any resource, using an arm as an example
my_arm_name = my_arm.get_resource_name("my_arm")

Parameters:: name (str) – The name of the Resource
Returns:: The ResourceName of this Resource
Return type:: ResourceName

get_operation(kwargs: Mapping[str, Any]) → viam.operations.Operation

Get the Operation associated with the currently running function.

When writing custom resources, you should get the Operation by calling this function and check to see if it’s cancelled. If the Operation is cancelled, then you can perform any necessary (terminating long running tasks, cleaning up connections, etc. ).

Parameters:: kwargs (Mapping[str, Any]) – The kwargs object containing the operation
Returns:: The operation associated with this function
Return type:: viam.operations.Operation

async close()

Safely shut down the resource and prevent further use.

Close must be idempotent. Later configuration may allow a resource to be “open” again. If a resource does not want or need a close function, it is assumed that the resource does not need to return errors when future non-Close methods are called.

await component.close()