package google.events.cloud.visionai.v1

Get desktop application:
View/edit binary Protocol Buffers messages

Message describing AI-enabled Devices Input Config.

Used in: ProcessorConfig

(message has no fields)

Represents a hardware accelerator type.

Used in: MachineSpec

• ACCELERATOR_TYPE_UNSPECIFIED = 0

  Unspecified accelerator type, which means no accelerator.

• NVIDIA_TESLA_K80 = 1

  Nvidia Tesla K80 GPU.

• NVIDIA_TESLA_P100 = 2

  Nvidia Tesla P100 GPU.

• NVIDIA_TESLA_V100 = 3

  Nvidia Tesla V100 GPU.

• NVIDIA_TESLA_P4 = 4

  Nvidia Tesla P4 GPU.

• NVIDIA_TESLA_T4 = 5

  Nvidia Tesla T4 GPU.

• NVIDIA_TESLA_A100 = 8

  Nvidia Tesla A100 GPU.

• TPU_V2 = 6

  TPU v2.

• TPU_V3 = 7

  TPU v3.

Message describing the Analysis object.

Used in: AnalysisEventData

• string name = 1

  The name of resource.

• optional protobuf.Timestamp create_time = 2

  Output only. The create timestamp.

• optional protobuf.Timestamp update_time = 3

  Output only. The update timestamp.

• map<string, string> labels = 4

  Labels as key value pairs.

• optional AnalysisDefinition analysis_definition = 5

  The definition of the analysis.

• map<string, string> input_streams_mapping = 6

  Map from the input parameter in the definition to the real stream. E.g., suppose you had a stream source operator named "input-0" and you try to receive from the real stream "stream-0". You can add the following mapping: [input-0: stream-0].

• map<string, string> output_streams_mapping = 7

  Map from the output parameter in the definition to the real stream. E.g., suppose you had a stream sink operator named "output-0" and you try to send to the real stream "stream-0". You can add the following mapping: [output-0: stream-0].

• bool disable_event_watch = 8

  Boolean flag to indicate whether you would like to disable the ability to automatically start a Process when new event happening in the input Stream. If you would like to start a Process manually, the field needs to be set to true.

The CloudEvent raised when an Analysis is created.

• optional AnalysisEventData data = 1

  The data associated with the event.

Defines a full analysis. This is a description of the overall live analytics pipeline. You may think of this as an edge list representation of a multigraph. This may be directly authored by a human in protobuf textformat, or it may be generated by a programming API (perhaps Python or JavaScript depending on context).

Used in: Analysis

• repeated AnalyzerDefinition analyzers = 1

  Analyzer definitions.

The CloudEvent raised when an Analysis is deleted.

• optional AnalysisEventData data = 1

  The data associated with the event.

The data within all Analysis events.

Used in: AnalysisCreatedEvent, AnalysisDeletedEvent, AnalysisUpdatedEvent

• optional Analysis payload = 1

  Optional. The Analysis event payload. Unset for deletion events.

The CloudEvent raised when an Analysis is updated.

• optional AnalysisEventData data = 1

  The data associated with the event.

Defines an Analyzer. An analyzer processes data from its input streams using the logic defined in the Operator that it represents. Of course, it produces data for the output streams declared in the Operator.

Used in: AnalysisDefinition

• string analyzer = 1

  The name of this analyzer. Tentatively [a-z][a-z0-9]*(_[a-z0-9]+)*.

• string operator = 2

  The name of the operator that this analyzer runs. Must match the name of a supported operator.

• repeated AnalyzerDefinition.StreamInput inputs = 3

  Input streams.

• map<string, AttributeValue> attrs = 4

  The attribute values that this analyzer applies to the operator. Supply a mapping between the attribute names and the actual value you wish to apply. If an attribute name is omitted, then it will take a preconfigured default value.

• optional AnalyzerDefinition.DebugOptions debug_options = 5

  Debug options.

Options available for debugging purposes only.

Used in: AnalyzerDefinition

• map<string, string> environment_variables = 1

  Environment variables.

The inputs to this analyzer. We accept input name references of the following form: <analyzer-name>:<output-argument-name> Example: Suppose you had an operator named "SomeOp" that has 2 output arguments, the first of which is named "foo" and the second of which is named "bar", and an operator named "MyOp" that accepts 2 inputs. Also suppose that there is an analyzer named "some-analyzer" that is running "SomeOp" and another analyzer named "my-analyzer" running "MyOp". To indicate that "my-analyzer" is to consume "some-analyzer"'s "foo" output as its first input and "some-analyzer"'s "bar" output as its second input, you can set this field to the following: input = ["some-analyzer:foo", "some-analyzer:bar"]

Used in: AnalyzerDefinition

• string input = 1

  The name of the stream input (as discussed above).

Message describing Application object

Used in: ApplicationEventData

• string name = 1

  name of resource

• optional protobuf.Timestamp create_time = 2

  Output only. [Output only] Create timestamp

• optional protobuf.Timestamp update_time = 3

  Output only. [Output only] Update timestamp

• map<string, string> labels = 4

  Labels as key value pairs

• string display_name = 5

  Required. A user friendly display name for the solution.

• string description = 6

  A description for this application.

• optional ApplicationConfigs application_configs = 7

  Application graph configuration.

• optional Application.ApplicationRuntimeInfo runtime_info = 8

  Output only. Application graph runtime info. Only exists when application state equals to DEPLOYED.

• Application.State state = 9

  Output only. State of the application.

• Application.BillingMode billing_mode = 12

  Billing mode of the application.

Message storing the runtime information of the application.

Used in: Application

• optional protobuf.Timestamp deploy_time = 1

  Timestamp when the engine be deployed

• repeated ApplicationRuntimeInfo.GlobalOutputResource global_output_resources = 3

  Globally created resources like warehouse dataschemas.

• optional ApplicationRuntimeInfo.MonitoringConfig monitoring_config = 4

  Monitoring-related configuration for this application.

Message about output resources from application.

Used in: ApplicationRuntimeInfo

• string output_resource = 1

  The full resource name of the outputted resources.

• string producer_node = 2

  The name of graph node who produces the output resource name. For example: output_resource: /projects/123/locations/us-central1/corpora/my-corpus/dataSchemas/my-schema producer_node: occupancy-count

• string key = 3

  The key of the output resource, it has to be unique within the same producer node. One producer node can output several output resources, the key can be used to match corresponding output resources.

Monitoring-related configuration for an application.

Used in: ApplicationRuntimeInfo

• bool enabled = 1

  Whether this application has monitoring enabled.

Billing mode of the Application

Used in: Application

• BILLING_MODE_UNSPECIFIED = 0

  The default value.

• PAYG = 1

  Pay as you go billing mode.

• MONTHLY = 2

  Monthly billing mode.

State of the Application

Used in: Application

• STATE_UNSPECIFIED = 0

  The default value. This value is used if the state is omitted.

• CREATED = 1

  State CREATED.

• DEPLOYING = 2

  State DEPLOYING.

• DEPLOYED = 3

  State DEPLOYED.

• UNDEPLOYING = 4

  State UNDEPLOYING.

• DELETED = 5

  State DELETED.

• ERROR = 6

  State ERROR.

• CREATING = 7

  State CREATING.

• UPDATING = 8

  State Updating.

• DELETING = 9

  State Deleting.

• FIXING = 10

  State Fixing.

Message storing the graph of the application.

Used in: Application, Draft

• repeated Node nodes = 1

  A list of nodes in the application graph.

The CloudEvent raised when an Application is created.

• optional ApplicationEventData data = 1

  The data associated with the event.

The CloudEvent raised when an Application is deleted.

• optional ApplicationEventData data = 1

  The data associated with the event.

The data within all Application events.

Used in: ApplicationCreatedEvent, ApplicationDeletedEvent, ApplicationUpdatedEvent

• optional Application payload = 1

  Optional. The Application event payload. Unset for deletion events.

The CloudEvent raised when an Application is updated.

• optional ApplicationEventData data = 1

  The data associated with the event.

Represents an actual value of an operator attribute.

Used in: AnalyzerDefinition

• oneof value

  Attribute value.

  • int64 i = 1

    int.

  • float f = 2

    float.

  • bool b = 3

    bool.

  • bytes s = 4

    string.

The metric specification that defines the target resource utilization (CPU utilization, accelerator's duty cycle, and so on) for calculating the desired replica count.

Used in: DedicatedResources

• string metric_name = 1

  Required. The resource metric name. Supported metrics: * For Online Prediction: * `aiplatform.googleapis.com/prediction/online/accelerator/duty_cycle` * `aiplatform.googleapis.com/prediction/online/cpu/utilization`

• int32 target = 2

  The target resource utilization in percentage (1% - 100%) for the given metric; once the real usage deviates from the target by a certain percentage, the machine replicas change. The default value is 60 (representing 60%) if not provided.

Message of configurations for BigQuery processor.

Used in: ProcessorConfig

• string table = 1

  BigQuery table resource for Vision AI Platform to ingest annotations to.

• map<string, string> cloud_function_mapping = 2

  Data Schema By default, Vision AI Application will try to write annotations to the target BigQuery table using the following schema: ingestion_time: TIMESTAMP, the ingestion time of the original data. application: STRING, name of the application which produces the annotation. instance: STRING, Id of the instance which produces the annotation. node: STRING, name of the application graph node which produces the annotation. annotation: STRING or JSON, the actual annotation protobuf will be converted to json string with bytes field as 64 encoded string. It can be written to both String or Json type column. To forward annotation data to an existing BigQuery table, customer needs to make sure the compatibility of the schema. The map maps application node name to its corresponding cloud function endpoint to transform the annotations directly to the google.cloud.bigquery.storage.v1.AppendRowsRequest (only avro_rows or proto_rows should be set). If configured, annotations produced by corresponding application node will sent to the Cloud Function at first before be forwarded to BigQuery. If the default table schema doesn't fit, customer is able to transform the annotation output from Vision AI Application to arbitrary BigQuery table schema with CloudFunction. * The cloud function will receive AppPlatformCloudFunctionRequest where the annotations field will be the json format of Vision AI annotation. * The cloud function should return AppPlatformCloudFunctionResponse with AppendRowsRequest stored in the annotations field. * To drop the annotation, simply clear the annotations field in the returned AppPlatformCloudFunctionResponse.

• bool create_default_table_if_not_exists = 3

  If true, App Platform will create the BigQuery DataSet and the BigQuery Table with default schema if the specified table doesn't exist. This doesn't work if any cloud function customized schema is specified since the system doesn't know your desired schema. JSON column will be used in the default table created by App Platform.

Message describing the Cluster object.

Used in: ClusterEventData

• string name = 1

  Output only. Name of the resource.

• optional protobuf.Timestamp create_time = 2

  Output only. The create timestamp.

• optional protobuf.Timestamp update_time = 3

  Output only. The update timestamp.

• map<string, string> labels = 4

  Labels as key value pairs

• map<string, string> annotations = 5

  Annotations to allow clients to store small amounts of arbitrary data.

• string dataplane_service_endpoint = 6

  Output only. The DNS name of the data plane service

• Cluster.State state = 7

  Output only. The current state of the cluster.

• string psc_target = 8

  Output only. The private service connection service target name.

The current state of the cluster.

Used in: Cluster

• STATE_UNSPECIFIED = 0

  Not set.

• PROVISIONING = 1

  The PROVISIONING state indicates the cluster is being created.

• RUNNING = 2

  The RUNNING state indicates the cluster has been created and is fully usable.

• STOPPING = 3

  The STOPPING state indicates the cluster is being deleted.

• ERROR = 4

  The ERROR state indicates the cluster is unusable. It will be automatically deleted.

The CloudEvent raised when a Cluster is created.

• optional ClusterEventData data = 1

  The data associated with the event.

The CloudEvent raised when a Cluster is deleted.

• optional ClusterEventData data = 1

  The data associated with the event.

The data within all Cluster events.

Used in: ClusterCreatedEvent, ClusterDeletedEvent, ClusterUpdatedEvent

• optional Cluster payload = 1

  Optional. The Cluster event payload. Unset for deletion events.

The CloudEvent raised when a Cluster is updated.

• optional ClusterEventData data = 1

  The data associated with the event.

Describes the source info for a custom processor.

Used in: Processor

• oneof artifact_path

  The path where App Platform loads the artifacts for the custom processor.

  • string vertex_model = 2

    The resource name original model hosted in the vertex AI platform.

• CustomProcessorSourceInfo.SourceType source_type = 1

  The original product which holds the custom processor's functionality.

• map<string, string> additional_info = 4

  Output only. Additional info related to the imported custom processor. Data is filled in by app platform during the processor creation.

• optional CustomProcessorSourceInfo.ModelSchema model_schema = 5

  Model schema files which specifies the signature of the model. For VERTEX_CUSTOM models, instances schema is required. If instances schema is not specified during the processor creation, VisionAI Platform will try to get it from Vertex, if it doesn't exist, the creation will fail.

The schema is defined as an OpenAPI 3.0.2 [Schema Object](https://github.com/OAI/OpenAPI-Specification/blob/main/versions/3.0.2.md#schemaObject).

Used in: CustomProcessorSourceInfo

• optional GcsSource instances_schema = 1

  Cloud Storage location to a YAML file that defines the format of a single instance used in prediction and explanation requests.

• optional GcsSource parameters_schema = 2

  Cloud Storage location to a YAML file that defines the prediction and explanation parameters.

• optional GcsSource predictions_schema = 3

  Cloud Storage location to a YAML file that defines the format of a single prediction or explanation.

Source type of the imported custom processor.

Used in: CustomProcessorSourceInfo

• SOURCE_TYPE_UNSPECIFIED = 0

  Source type unspecified.

• VERTEX_AUTOML = 1

  Custom processors coming from Vertex AutoML product.

• VERTEX_CUSTOM = 2

  Custom processors coming from general custom models from Vertex.

• PRODUCT_RECOGNIZER = 3

  Source for Product Recognizer.

All supported data types.

Used in: ProcessorIOSpec.GraphInputChannelSpec, ProcessorIOSpec.GraphOutputChannelSpec

• DATA_TYPE_UNSPECIFIED = 0

  The default value of DataType.

• VIDEO = 1

  Video data type like H264.

• IMAGE = 3

  Image data type.

• PROTO = 2

  Protobuf data type, usually used for general data blob.

A description of resources that are dedicated to a DeployedModel, and that need a higher degree of manual configuration.

Used in: VertexCustomConfig

• optional MachineSpec machine_spec = 1

  Required. Immutable. The specification of a single machine used by the prediction.

• int32 min_replica_count = 2

  Required. Immutable. The minimum number of machine replicas this DeployedModel will be always deployed on. This value must be greater than or equal to 1. If traffic against the DeployedModel increases, it may dynamically be deployed onto more replicas, and as traffic decreases, some of these extra replicas may be freed.

• int32 max_replica_count = 3

  Immutable. The maximum number of replicas this DeployedModel may be deployed on when the traffic against it increases. If the requested value is too large, the deployment will error, but if deployment succeeds then the ability to scale the model to that many replicas is guaranteed (barring service outages). If traffic against the DeployedModel increases beyond what its replicas at maximum may handle, a portion of the traffic will be dropped. If this value is not provided, will use [min_replica_count][google.cloud.visionai.v1.DedicatedResources.min_replica_count] as the default value. The value of this field impacts the charge against Vertex CPU and GPU quotas. Specifically, you will be charged for max_replica_count * number of cores in the selected machine type) and (max_replica_count * number of GPUs per replica in the selected machine type).

• repeated AutoscalingMetricSpec autoscaling_metric_specs = 4

  Immutable. The metric specifications that overrides a resource utilization metric (CPU utilization, accelerator's duty cycle, and so on) target value (default to 60 if not set). At most one entry is allowed per metric. If [machine_spec.accelerator_count][google.cloud.visionai.v1.MachineSpec.accelerator_count] is above 0, the autoscaling will be based on both CPU utilization and accelerator's duty cycle metrics and scale up when either metrics exceeds its target value while scale down if both metrics are under their target value. The default target value is 60 for both metrics. If [machine_spec.accelerator_count][google.cloud.visionai.v1.MachineSpec.accelerator_count] is 0, the autoscaling will be based on CPU utilization metric only with default target value 60 if not explicitly set. For example, in the case of Online Prediction, if you want to override target CPU utilization to 80, you should set [autoscaling_metric_specs.metric_name][google.cloud.visionai.v1.AutoscalingMetricSpec.metric_name] to `aiplatform.googleapis.com/prediction/online/cpu/utilization` and [autoscaling_metric_specs.target][google.cloud.visionai.v1.AutoscalingMetricSpec.target] to `80`.

Message describing Draft object

Used in: DraftEventData

• string name = 1

  name of resource

• optional protobuf.Timestamp create_time = 2

  Output only. [Output only] Create timestamp

• optional protobuf.Timestamp update_time = 7

  Output only. [Output only] Create timestamp

• map<string, string> labels = 3

  Labels as key value pairs

• string display_name = 4

  Required. A user friendly display name for the solution.

• string description = 5

  A description for this application.

• optional ApplicationConfigs draft_application_configs = 6

  The draft application configs which haven't been updated to an application.

The CloudEvent raised when a Draft is created.

• optional DraftEventData data = 1

  The data associated with the event.

The CloudEvent raised when a Draft is deleted.

• optional DraftEventData data = 1

  The data associated with the event.

The data within all Draft events.

Used in: DraftCreatedEvent, DraftDeletedEvent, DraftUpdatedEvent

• optional Draft payload = 1

  Optional. The Draft event payload. Unset for deletion events.

The CloudEvent raised when a Draft is updated.

• optional DraftEventData data = 1

  The data associated with the event.

Message describing the Event object.

Used in: EventEventData

• string name = 1

  Name of the resource.

• optional protobuf.Timestamp create_time = 2

  Output only. The create timestamp.

• optional protobuf.Timestamp update_time = 3

  Output only. The update timestamp.

• map<string, string> labels = 4

  Labels as key value pairs.

• map<string, string> annotations = 5

  Annotations to allow clients to store small amounts of arbitrary data.

• Event.Clock alignment_clock = 6

  The clock used for joining streams.

• optional protobuf.Duration grace_period = 7

  Grace period for cleaning up the event. This is the time the controller waits for before deleting the event. During this period, if there is any active channel on the event. The deletion of the event after grace_period will be ignored.

Clock that will be used for joining streams.

Used in: Event

• CLOCK_UNSPECIFIED = 0

  Clock is not specified.

• CAPTURE = 1

  Use the timestamp when the data is captured. Clients need to sync the clock.

• INGEST = 2

  Use the timestamp when the data is received.

The CloudEvent raised when an Event is created.

• optional EventEventData data = 1

  The data associated with the event.

The CloudEvent raised when an Event is deleted.

• optional EventEventData data = 1

  The data associated with the event.

The data within all Event events.

Used in: EventCreatedEvent, EventDeletedEvent, EventUpdatedEvent

• optional Event payload = 1

  Optional. The Event event payload. Unset for deletion events.

The CloudEvent raised when an Event is updated.

• optional EventEventData data = 1

  The data associated with the event.

The Google Cloud Storage location for the input content.

Used in: CustomProcessorSourceInfo.ModelSchema

• repeated string uris = 1

  Required. References to a Google Cloud Storage paths.

Message of configurations for General Object Detection processor.

Used in: ProcessorConfig

(message has no fields)

Specification of a single machine.

Used in: DedicatedResources

• string machine_type = 1

  Immutable. The type of the machine. See the [list of machine types supported for prediction](https://cloud.google.com/vertex-ai/docs/predictions/configure-compute#machine-types) See the [list of machine types supported for custom training](https://cloud.google.com/vertex-ai/docs/training/configure-compute#machine-types). For [DeployedModel][] this field is optional, and the default value is `n1-standard-2`. For [BatchPredictionJob][] or as part of [WorkerPoolSpec][] this field is required.

• AcceleratorType accelerator_type = 2

  Immutable. The type of accelerator(s) that may be attached to the machine as per [accelerator_count][google.cloud.visionai.v1.MachineSpec.accelerator_count].

• int32 accelerator_count = 3

  The number of accelerators to attach to the machine.

Message describing MediaWarehouseConfig.

Used in: ProcessorConfig

• string corpus = 1

  Resource name of the Media Warehouse corpus. Format: projects/${project_id}/locations/${location_id}/corpora/${corpus_id}

• string region = 2

  Deprecated.

• optional protobuf.Duration ttl = 3

  The duration for which all media assets, associated metadata, and search documents can exist.

All the supported model types in Vision AI App Platform.

Used in: Processor

• MODEL_TYPE_UNSPECIFIED = 0

  Processor Type UNSPECIFIED.

• IMAGE_CLASSIFICATION = 1

  Model Type Image Classification.

• OBJECT_DETECTION = 2

  Model Type Object Detection.

• VIDEO_CLASSIFICATION = 3

  Model Type Video Classification.

• VIDEO_OBJECT_TRACKING = 4

  Model Type Object Tracking.

• VIDEO_ACTION_RECOGNITION = 5

  Model Type Action Recognition.

• OCCUPANCY_COUNTING = 6

  Model Type Occupancy Counting.

• PERSON_BLUR = 7

  Model Type Person Blur.

• VERTEX_CUSTOM = 8

  Model Type Vertex Custom.

Message describing node object.

Used in: ApplicationConfigs

• oneof stream_output_config

  • bool output_all_output_channels_to_stream = 6

    By default, the output of the node will only be available to downstream nodes. To consume the direct output from the application node, the output must be sent to Vision AI Streams at first. By setting output_all_output_channels_to_stream to true, App Platform will automatically send all the outputs of the current node to Vision AI Stream resources (one stream per output channel). The output stream resource will be created by App Platform automatically during deployment and deleted after application un-deployment. Note that this config applies to all the Application Instances. The output stream can be override at instance level by configuring the `output_resources` section of Instance resource. `producer_node` should be current node, `output_resource_binding` should be the output channel name (or leave it blank if there is only 1 output channel of the processor) and `output_resource` should be the target output stream.

• string name = 1

  Required. A unique name for the node.

• string display_name = 2

  A user friendly display name for the node.

• optional ProcessorConfig node_config = 3

  Node config.

• string processor = 4

  Processor name refer to the chosen processor resource.

• repeated Node.InputEdge parents = 5

  Parent node. Input node should not have parent node. For V1 Alpha1/Beta only media warehouse node can have multiple parents, other types of nodes will only have one parent.

Message describing one edge pointing into a node.

Used in: Node

• string parent_node = 1

  The name of the parent node.

• string parent_output_channel = 2

  The connected output artifact of the parent node. It can be omitted if target processor only has 1 output artifact.

• string connected_input_channel = 3

  The connected input channel of the current node's processor. It can be omitted if target processor only has 1 input channel.

Normalized Polygon.

Used in: StreamAnnotation

• repeated NormalizedVertex normalized_vertices = 1

  The bounding polygon normalized vertices. Top left corner of the image will be [0, 0].

Normalized Pplyline, which represents a curve consisting of connected straight-line segments.

Used in: StreamAnnotation

• repeated NormalizedVertex normalized_vertices = 1

  A sequence of vertices connected by straight lines.

A vertex represents a 2D point in the image. NOTE: the normalized vertex coordinates are relative to the original image and range from 0 to 1.

Used in: NormalizedPolygon, NormalizedPolyline

• float x = 1

  X coordinate.

• float y = 2

  Y coordinate.

Message describing OccupancyCountConfig.

Used in: ProcessorConfig

• bool enable_people_counting = 1

  Whether to count the appearances of people, output counts have 'people' as the key.

• bool enable_vehicle_counting = 2

  Whether to count the appearances of vehicles, output counts will have 'vehicle' as the key.

• bool enable_dwelling_time_tracking = 3

  Whether to track each invidual object's loitering time inside the scene or specific zone.

Message describing FaceBlurConfig.

Used in: ProcessorConfig

• PersonBlurConfig.PersonBlurType person_blur_type = 1

  Person blur type.

• bool faces_only = 2

  Whether only blur faces other than the whole object in the processor.

Type of Person Blur

Used in: PersonBlurConfig

• PERSON_BLUR_TYPE_UNSPECIFIED = 0

  PersonBlur Type UNSPECIFIED.

• FULL_OCCULUSION = 1

  FaceBlur Type full occlusion.

• BLUR_FILTER = 2

  FaceBlur Type blur filter.

Message describing PersonVehicleDetectionConfig.

Used in: ProcessorConfig

• bool enable_people_counting = 1

  At least one of enable_people_counting and enable_vehicle_counting fields must be set to true. Whether to count the appearances of people, output counts have 'people' as the key.

• bool enable_vehicle_counting = 2

  Whether to count the appearances of vehicles, output counts will have 'vehicle' as the key.

Message describing PersonalProtectiveEquipmentDetectionConfig.

Used in: ProcessorConfig

• bool enable_face_coverage_detection = 1

  Whether to enable face coverage detection.

• bool enable_head_coverage_detection = 2

  Whether to enable head coverage detection.

• bool enable_hands_coverage_detection = 3

  Whether to enable hands coverage detection.

Message describing the Process object.

Used in: ProcessEventData

• string name = 1

  The name of resource.

• optional protobuf.Timestamp create_time = 2

  Output only. The create timestamp.

• optional protobuf.Timestamp update_time = 3

  Output only. The update timestamp.

• string analysis = 4

  Required. Reference to an existing Analysis resource.

• repeated string attribute_overrides = 5

  Optional. Attribute overrides of the Analyzers. Format for each single override item: "{analyzer_name}:{attribute_key}={value}"

• optional RunStatus run_status = 6

  Optional. Status of the Process.

• RunMode run_mode = 7

  Optional. Run mode of the Process.

• string event_id = 8

  Optional. Event ID of the input/output streams. This is useful when you have a StreamSource/StreamSink operator in the Analysis, and you want to manually specify the Event to read from/write to.

• string batch_id = 9

  Optional. Optional: Batch ID of the Process.

• int32 retry_count = 10

  Optional. Optional: The number of retries for a process in submission mode the system should try before declaring failure. By default, no retry will be performed.

The CloudEvent raised when a Process is created.

• optional ProcessEventData data = 1

  The data associated with the event.

The CloudEvent raised when a Process is deleted.

• optional ProcessEventData data = 1

  The data associated with the event.

The data within all Process events.

Used in: ProcessCreatedEvent, ProcessDeletedEvent, ProcessUpdatedEvent

• optional Process payload = 1

  Optional. The Process event payload. Unset for deletion events.

The CloudEvent raised when a Process is updated.

• optional ProcessEventData data = 1

  The data associated with the event.

Message describing Processor object. Next ID: 19

Used in: ProcessorEventData

• string name = 1

  name of resource.

• optional protobuf.Timestamp create_time = 2

  Output only. [Output only] Create timestamp.

• optional protobuf.Timestamp update_time = 3

  Output only. [Output only] Update timestamp.

• map<string, string> labels = 4

  Labels as key value pairs.

• string display_name = 5

  Required. A user friendly display name for the processor.

• string description = 10

  Illustrative sentences for describing the functionality of the processor.

• Processor.ProcessorType processor_type = 6

  Output only. Processor Type.

• ModelType model_type = 13

  Model Type.

• optional CustomProcessorSourceInfo custom_processor_source_info = 7

  Source info for customer created processor.

• Processor.ProcessorState state = 8

  Output only. State of the Processor.

• optional ProcessorIOSpec processor_io_spec = 11

  Output only. [Output only] The input / output specifications of a processor, each type of processor has fixed input / output specs which cannot be altered by customer.

• string configuration_typeurl = 14

  Output only. The corresponding configuration can be used in the Application to customize the behavior of the processor.

• repeated StreamAnnotationType supported_annotation_types = 15

• bool supports_post_processing = 17

  Indicates if the processor supports post processing.

Used in: Processor

• PROCESSOR_STATE_UNSPECIFIED = 0

  Unspecified Processor state.

• CREATING = 1

  Processor is being created (not ready for use).

• ACTIVE = 2

  Processor is and ready for use.

• DELETING = 3

  Processor is being deleted (not ready for use).

• FAILED = 4

  Processor deleted or creation failed .

Type

Used in: Processor

• PROCESSOR_TYPE_UNSPECIFIED = 0

  Processor Type UNSPECIFIED.

• PRETRAINED = 1

  Processor Type PRETRAINED. Pretrained processor is developed by Vision AI App Platform with state-of-the-art vision data processing functionality, like occupancy counting or person blur. Pretrained processor is usually publicly available.

• CUSTOM = 2

  Processor Type CUSTOM. Custom processors are specialized processors which are either uploaded by customers or imported from other GCP platform (for example Vertex AI). Custom processor is only visible to the creator.

• CONNECTOR = 3

  Processor Type CONNECTOR. Connector processors are special processors which perform I/O for the application, they do not processing the data but either deliver the data to other processors or receive data from other processors.

Next ID: 29

Used in: Node

• oneof processor_config

  • VideoStreamInputConfig video_stream_input_config = 9

    Configs of stream input processor.

  • AIEnabledDevicesInputConfig ai_enabled_devices_input_config = 20

    Config of AI-enabled input devices.

  • MediaWarehouseConfig media_warehouse_config = 10

    Configs of media warehouse processor.

  • PersonBlurConfig person_blur_config = 11

    Configs of person blur processor.

  • OccupancyCountConfig occupancy_count_config = 12

    Configs of occupancy count processor.

  • PersonVehicleDetectionConfig person_vehicle_detection_config = 15

    Configs of Person Vehicle Detection processor.

  • VertexAutoMLVisionConfig vertex_automl_vision_config = 13

    Configs of Vertex AutoML vision processor.

  • VertexAutoMLVideoConfig vertex_automl_video_config = 14

    Configs of Vertex AutoML video processor.

  • VertexCustomConfig vertex_custom_config = 17

    Configs of Vertex Custom processor.

  • GeneralObjectDetectionConfig general_object_detection_config = 18

    Configs of General Object Detection processor.

  • BigQueryConfig big_query_config = 19

    Configs of BigQuery processor.

  • PersonalProtectiveEquipmentDetectionConfig personal_protective_equipment_detection_config = 22

    Configs of personal_protective_equipment_detection_config

The CloudEvent raised when a Processor is created.

• optional ProcessorEventData data = 1

  The data associated with the event.

The CloudEvent raised when a Processor is deleted.

• optional ProcessorEventData data = 1

  The data associated with the event.

The data within all Processor events.

Used in: ProcessorCreatedEvent, ProcessorDeletedEvent, ProcessorUpdatedEvent

• optional Processor payload = 1

  Optional. The Processor event payload. Unset for deletion events.

Message describing the input / output specifications of a processor.

Used in: Processor

• repeated ProcessorIOSpec.GraphInputChannelSpec graph_input_channel_specs = 3

  For processors with input_channel_specs, the processor must be explicitly connected to another processor.

• repeated ProcessorIOSpec.GraphOutputChannelSpec graph_output_channel_specs = 4

  The output artifact specifications for the current processor.

• repeated ProcessorIOSpec.InstanceResourceInputBindingSpec instance_resource_input_binding_specs = 5

  The input resource that needs to be fed from the application instance.

• repeated ProcessorIOSpec.InstanceResourceOutputBindingSpec instance_resource_output_binding_specs = 6

  The output resource that the processor will generate per instance. Other than the explicitly listed output bindings here, all the processors' GraphOutputChannels can be binded to stream resource. The bind name then is the same as the GraphOutputChannel's name.

Message for input channel specification.

Used in: ProcessorIOSpec

• string name = 1

  The name of the current input channel.

• DataType data_type = 2

  The data types of the current input channel. When this field has more than 1 value, it means this input channel can be connected to either of these different data types.

• repeated string accepted_data_type_uris = 5

  If specified, only those detailed data types can be connected to the processor. For example, jpeg stream for MEDIA, or PredictionResult proto for PROTO type. If unspecified, then any proto is accepted.

• bool required = 3

  Whether the current input channel is required by the processor. For example, for a processor with required video input and optional audio input, if video input is missing, the application will be rejected while the audio input can be missing as long as the video input exists.

• int64 max_connection_allowed = 4

  How many input edges can be connected to this input channel. 0 means unlimited.

Message for output channel specification.

Used in: ProcessorIOSpec

• string name = 1

  The name of the current output channel.

• DataType data_type = 2

  The data type of the current output channel.

• string data_type_uri = 3

Message for instance resource channel specification. External resources are virtual nodes which are not expressed in the application graph. Each processor expresses its out-graph spec, so customer is able to override the external source or destinations to the

Used in: ProcessorIOSpec

• oneof resource_type

  • string config_type_uri = 2

    The configuration proto that includes the Googleapis resources. I.e. type.googleapis.com/google.cloud.vision.v1.StreamWithAnnotation

  • string resource_type_uri = 3

    The direct type url of Googleapis resource. i.e. type.googleapis.com/google.cloud.vision.v1.Asset

• string name = 1

  Name of the input binding, unique within the processor.

Used in: ProcessorIOSpec

• string name = 1

  Name of the output binding, unique within the processor.

• string resource_type_uri = 2

  The resource type uri of the acceptable output resource.

• bool explicit = 3

  Whether the output resource needs to be explicitly set in the instance. If it is false, the processor will automatically generate it if required.

The CloudEvent raised when a Processor is updated.

• optional ProcessorEventData data = 1

  The data associated with the event.

RunMode represents the mode to launch the Process on.

Used in: Process

• RUN_MODE_UNSPECIFIED = 0

  Mode is unspecified.

• LIVE = 1

  Live mode. Meaning the Process is launched to handle live video source, and possible packet drops are expected.

• SUBMISSION = 2

  Submission mode. Meaning the Process is launched to handle bounded video files, with no packet drop. Completion status is tracked.

Message describing the status of the Process.

Used in: Process

• RunStatus.State state = 1

  The state of the Process.

• string reason = 2

  The reason of becoming the state.

State represents the running status of the Process.

Used in: RunStatus

• STATE_UNSPECIFIED = 0

  State is unspecified.

• INITIALIZING = 1

  INITIALIZING means the Process is scheduled but yet ready to handle real traffic.

• RUNNING = 2

  RUNNING means the Process is up running and handling traffic.

• COMPLETED = 3

  COMPLETED means the Process has completed the processing, especially for non-streaming use case.

• FAILED = 4

  FAILED means the Process failed to complete the processing.

• PENDING = 5

  PENDING means the Process is created but yet to be scheduled.

Message describing the Series object.

Used in: SeriesEventData

• string name = 1

  Name of the resource.

• optional protobuf.Timestamp create_time = 2

  Output only. The create timestamp.

• optional protobuf.Timestamp update_time = 3

  Output only. The update timestamp.

• map<string, string> labels = 4

  Labels as key value pairs.

• map<string, string> annotations = 5

  Annotations to allow clients to store small amounts of arbitrary data.

• string stream = 6

  Required. Stream that is associated with this series.

• string event = 7

  Required. Event that is associated with this series.

The CloudEvent raised when a Series is created.

• optional SeriesEventData data = 1

  The data associated with the event.

The CloudEvent raised when a Series is deleted.

• optional SeriesEventData data = 1

  The data associated with the event.

The data within all Series events.

Used in: SeriesCreatedEvent, SeriesDeletedEvent, SeriesUpdatedEvent

• optional Series payload = 1

  Optional. The Series event payload. Unset for deletion events.

The CloudEvent raised when a Series is updated.

• optional SeriesEventData data = 1

  The data associated with the event.

Message describing the Stream object. The Stream and the Event resources are many to many; i.e., each Stream resource can associate to many Event resources and each Event resource can associate to many Stream resources.

Used in: StreamEventData

• string name = 1

  Name of the resource.

• optional protobuf.Timestamp create_time = 2

  Output only. The create timestamp.

• optional protobuf.Timestamp update_time = 3

  Output only. The update timestamp.

• map<string, string> labels = 4

  Labels as key value pairs.

• map<string, string> annotations = 5

  Annotations to allow clients to store small amounts of arbitrary data.

• string display_name = 6

  The display name for the stream resource.

• bool enable_hls_playback = 7

  Whether to enable the HLS playback service on this stream.

• string media_warehouse_asset = 8

  The name of the media warehouse asset for long term storage of stream data. Format: projects/${p_id}/locations/${l_id}/corpora/${c_id}/assets/${a_id} Remain empty if the media warehouse storage is not needed for the stream.

message about annotations about Vision AI stream resource.

Used in: StreamWithAnnotation, StreamWithAnnotation.NodeAnnotation

• oneof annotation_payload

  • NormalizedPolygon active_zone = 5

    Annotation for type ACTIVE_ZONE

  • NormalizedPolyline crossing_line = 6

    Annotation for type CROSSING_LINE

• string id = 1

  ID of the annotation. It must be unique when used in the certain context. For example, all the annotations to one input streams of a Vision AI application.

• string display_name = 2

  User-friendly name for the annotation.

• string source_stream = 3

  The Vision AI stream resource name.

• StreamAnnotationType type = 4

  The actual type of Annotation.

Enum describing all possible types of a stream annotation.

Used in: Processor, StreamAnnotation

• STREAM_ANNOTATION_TYPE_UNSPECIFIED = 0

  Type UNSPECIFIED.

• STREAM_ANNOTATION_TYPE_ACTIVE_ZONE = 1

  active_zone annotation defines a polygon on top of the content from an image/video based stream, following processing will only focus on the content inside the active zone.

• STREAM_ANNOTATION_TYPE_CROSSING_LINE = 2

  crossing_line annotation defines a polyline on top of the content from an image/video based Vision AI stream, events happening across the line will be captured. For example, the counts of people who goes acroos the line in Occupancy Analytic Processor.

The CloudEvent raised when a Stream is created.

• optional StreamEventData data = 1

  The data associated with the event.

The CloudEvent raised when a Stream is deleted.

• optional StreamEventData data = 1

  The data associated with the event.

The data within all Stream events.

Used in: StreamCreatedEvent, StreamDeletedEvent, StreamUpdatedEvent

• optional Stream payload = 1

  Optional. The Stream event payload. Unset for deletion events.

The CloudEvent raised when a Stream is updated.

• optional StreamEventData data = 1

  The data associated with the event.

Message describing Vision AI stream with application specific annotations. All the StreamAnnotation object inside this message MUST have unique id.

Used in: VideoStreamInputConfig

• string stream = 1

  Vision AI Stream resource name.

• repeated StreamAnnotation application_annotations = 2

  Annotations that will be applied to the whole application.

• repeated StreamWithAnnotation.NodeAnnotation node_annotations = 3

  Annotations that will be applied to the specific node of the application. If the same type of the annotations is applied to both application and node, the node annotation will be added in addition to the global application one. For example, if there is one active zone annotation for the whole application and one active zone annotation for the Occupancy Analytic processor, then the Occupancy Analytic processor will have two active zones defined.

Message describing annotations specific to application node.

Used in: StreamWithAnnotation

• string node = 1

  The node name of the application graph.

• repeated StreamAnnotation annotations = 2

  The node specific stream annotations.

Message describing VertexAutoMLVideoConfig.

Used in: ProcessorConfig

• float confidence_threshold = 1

  Only entities with higher score than the threshold will be returned. Value 0.0 means returns all the detected entities.

• repeated string blocked_labels = 2

  Labels specified in this field won't be returned.

• int32 max_predictions = 3

  At most this many predictions will be returned per output frame. Value 0 means to return all the detected entities.

• float bounding_box_size_limit = 4

  Only Bounding Box whose size is larger than this limit will be returned. Object Tracking only. Value 0.0 means to return all the detected entities.

Message of configurations of Vertex AutoML Vision Processors.

Used in: ProcessorConfig

• float confidence_threshold = 1

  Only entities with higher score than the threshold will be returned. Value 0.0 means to return all the detected entities.

• int32 max_predictions = 2

  At most this many predictions will be returned per output frame. Value 0 means to return all the detected entities.

Message describing VertexCustomConfig.

Used in: ProcessorConfig

• int32 max_prediction_fps = 1

  The max prediction frame per second. This attribute sets how fast the operator sends prediction requests to Vertex AI endpoint. Default value is 0, which means there is no max prediction fps limit. The operator sends prediction requests at input fps.

• optional DedicatedResources dedicated_resources = 2

  A description of resources that are dedicated to the DeployedModel, and that need a higher degree of manual configuration.

• string post_processing_cloud_function = 3

  If not empty, the prediction result will be sent to the specified cloud function for post processing. * The cloud function will receive AppPlatformCloudFunctionRequest where the annotations field will be the json format of proto PredictResponse. * The cloud function should return AppPlatformCloudFunctionResponse with PredictResponse stored in the annotations field. * To drop the prediction output, simply clear the payload field in the returned AppPlatformCloudFunctionResponse.

• bool attach_application_metadata = 4

  If true, the prediction request received by custom model will also contain metadata with the following schema: 'appPlatformMetadata': { 'ingestionTime': DOUBLE; (UNIX timestamp) 'application': STRING; 'instanceId': STRING; 'node': STRING; 'processor': STRING; }

Message describing Video Stream Input Config. This message should only be used as a placeholder for builtin:stream-input processor, actual stream binding should be specified using corresponding API.

Used in: ProcessorConfig

• repeated string streams = 1

• repeated StreamWithAnnotation streams_with_annotation = 2