Class PredictPoses

Namespace

Bonsai.Sleap

Assembly

Bonsai.Sleap.dll

Represents an operator that performs markerless multi-pose estimation for each image in the sequence using a SLEAP model.

PredictPoses implements the [top-down-model network]. The usual input of this operation will be a sequence of full frames where multiple instances are expected to be found. This operator will output a PoseCollection with N number of instances found in the image. Indexing a PoseCollection will return a Pose where we can access the Centroid of each detected instance along with the Pose containing information on all trained body parts.

To access the data of a specific body part we use the GetBodyPart. We set Name to match the part name defined in the training_config.json file. From that moment, the operator will always emit the selected BodyPart object and its inferred position (BodyPart.Position).

public class PredictPoses : Transform<IplImage, PoseCollection>

Inheritance

object

Combinator<IplImage, PoseCollection>

Transform<IplImage, PoseCollection>

PredictPoses

Inherited Members

Combinator<IplImage, PoseCollection>.Process(IObservable<IplImage>)

object.ToString()

object.Equals(object)

object.Equals(object, object)

object.ReferenceEquals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

Properties

CentroidMinConfidence

Gets or sets a value specifying the confidence threshold used to discard centroid predictions. If no value is specified, all estimated centroid positions are returned.

ColorConversion

Gets or sets a value specifying the optional color conversion used to prepare RGB video frames for inference. If no value is specified, no color conversion is performed.

ModelFileName

Gets or sets a value specifying the path to the exported Protocol Buffer file containing the pretrained SLEAP model.

PartMinConfidence

Gets or sets a value specifying the confidence threshold used to discard predicted body part positions. If no value is specified, all estimated positions are returned.

ScaleFactor

Gets or sets a value specifying the scale factor used to resize video frames for inference. If no value is specified, no resizing is performed.

TrainingConfig

Gets or sets a value specifying the path to the configuration JSON file containing training metadata.

Methods

Process(IObservable<IplImage>)

Performs markerless multi-pose estimation for each image in an observable sequence using a SLEAP model.

See Also

PredictCentroids

PredictPoseIdentities

PredictSinglePose

GetBodyPart