All Classes

Class	Description
AspectRatioHandling	How We should handle the situation where the input image and output image/NDArray have different aspect ratios? Use in ImageToNDArrayConfig and ImageResizeStep See ImageToNDArrayConfig for more details CENTER_CROP: Crop the larger dimension down to the correct aspect ratio (and then resize if necessary). PAD: Zero pad the smaller dimension to make the aspect ratio match the output (and then resize if necessary) STRETCH: Simply resize the image to the required aspect ratio, distorting the image if necessary
CameraFrameCaptureStep
ColorConstants
CropFixedGridStep	As per CropGridStep but the x/y location values are hardcoded into the configuration, instead of coming dynamically from the input Data instance
CropGridStep	Crop sub images out of a larger image, based on a grid.
DrawBoundingBoxStep	Scale: NONE: No scaling AT_LEAST: Scale up if necessary, so H >= scaleH and W >= scaleW AT_MOST: Scale down if necessary, so H <= scaleH and W <= scaleW
DrawBoundingBoxStep.Scale
DrawFaceKeyPointsStep	A step for drawing face keypoints.
DrawFaceKeyPointsStep.Scale
DrawFixedGridStep	As per DrawGridStep but the x/y location values are hardcoded into the configuration, instead of coming dynamically from the input Data instance
DrawGridStep	Draw a grid on the specified image, based on the x/y coordinates of the corners, and the number of segments within the grid in both directions. The 4 corner coordinates are defined as points, and come from Data.getListPoint(name), in the following order: topLeft, topRight, bottomLeft, bottomRight gridX is the number of grid segments between (topLeft and topRight) and (bottomLeft and bottomRight). gridY is the number of grid segments between (topLeft and bottomLeft) and (topRight and bottomRight) The colors and line thicknesses can be configured.
DrawHeatmapStep	Draw a heatmap using 2D points. Heat will accumulate over the lifecycle of this step. Configuration: points: Names of the points to be used for the heatmap.
DrawPointsStep	Draw 2D points Configuration: classColors: Optional: A list of colors to use for each class.
DrawSegmentationStep	Draw segmentation mask, optionally on an image. Configuration: classColors: Optional: A list of colors to use for each class.
ExtractBoundingBoxStep	ExtractBoundingBoxStep: Given one or more bounding boxes, and an input image, extract from the input region an image that corresponds to the bounding box region. i.e.: output as images the input image region covered by the bounding boxes Note: supports both BoundingBox and List<BoundingBox> fields.
GrayScaleStep	Transforms an image to be gray scale.
ImageConfigModuleInfo
ImageCropStep	Crop an image to the specified rectangular region.
ImageNormalization	NONE: No image normalization will be applied SCALE: Divide images by maxValue/2, or divide by 255/2 if maxValue is not specified, then subtract 1.0, to give output in range [-1,1].
ImageNormalization.Type
ImageResizeStep	Resize an image - scaling up or down as needed to comply with the specified output height/width.
ImageToNDArrayConfig	Configuration for converting Images to NDArrays.
ImageToNDArrayConfig.ListHandling	See ImageToNDArrayConfig - listHandling field
ImageToNDArrayStep	A PipelineStep for converting Images to NDArrays.
KonduitServingImageConfigJsonMapping
NDChannelLayout	Represents the type (and order) of the channels for an image after it has been converted to an NDArray RGB: 3 channels, ordered according to: red, green, blue - most common for TensorFlow, Keras, and some other libraries BGR: 3 channels, ordered according to: blue, green, red - the default for OpenCV, JavaCV, DL4J RGBA: 4 channels, ordered according to: red, green, blue, alpha BGRA: 4 channels, ordered according to: blue, green, red, alpha GRAYSCALE: 1 channel - grayscale
NDFormat
PerspectiveTransformStep	PerspectiveTransformStep: Applies a perspective transformation to Images, Points and Bounding Boxes. The transformation is defined either statically with the Step definition or dynamically with a list of 4 points in an input.
RelativeToAbsoluteStep	For a given set of Point, List<Point>, BoundingBox or List<BoundingBox> that are defined in relative terms (i.e., all values 0.0 to 1.0 in terms of "fraction of image height/width"), convert these to absolute values (i.e., pixels) using: (a) An input image, as specified via the imageName configuration, OR (b) An image height, as specified by imageH and imageW configuration Note that an ImageToNDArrayConfig can be provided, to account for the fact that the original image may have been cropped before being passed into a network that produced the Points or BoundingBoxes. If the imageToNDArrayConfig field is null, it is assumed no cropping has occurred.
ShowImageStep
VideoFrameCaptureStep	VideoFrameCaptureStep extracts a single frame from a video each time inference is called.