EfficientDetD2

object EfficientDetD2 : ONNXModels.ObjectDetection<EfficientDetObjectDetectionModel>
Content copied to clipboard

This model is a real-time neural network for object detection that detects 90 different classes (labels are available via org.jetbrains.kotlinx.dl.impl.dataset.Coco.V2017.labels method).

Internally it uses the EfficientNets as backbone networks.

The model have an input with the shape is (1x768x768x3) by default. H and W could be changed by user to any values.

The model has 1 output:

  • detections:0 with 7 numbers as [unknown number, ymin, _xmin_, ymax, xmax, score, coco label].

NOTE: The detections are limited to 100.

See also

Functions

model
Link copied to clipboard
open fun model(modelHub: ModelHub): OnnxInferenceModel
Content copied to clipboard
pretrainedModel
Link copied to clipboard
open override fun pretrainedModel(modelHub: ModelHub): EfficientDetObjectDetectionModel
Content copied to clipboard

Properties

inputShape
Link copied to clipboard
open override val inputShape: LongArray
Content copied to clipboard

Shape of the input accepted by this model, without batch size.

modelRelativePath
Link copied to clipboard
open override val modelRelativePath: String
Content copied to clipboard
preprocessor
Link copied to clipboard
open val preprocessor: Operation<Pair<FloatArray, TensorShape>, Pair<FloatArray, TensorShape>>
Content copied to clipboard