SSDLikeModel
class SSDLikeModel(internalModel: OnnxInferenceModel, metadata: SSDLikeModelMetadata, modelKindDescription: String?) : SSDLikeModelBase<Bitmap> , CameraXCompatibleModel, InferenceModel
Content copied to clipboard
Special model class for detection objects on images with built-in preprocessing and post-processing. Suitable for models with SSD like output decoding.
It internally uses ONNXModels.ObjectDetection.EfficientDetLite0 or other SSDLike models trained on the COCO dataset.
Since
0.5
Parameters
internalModel
model used to make predictions
Constructors
SSDLikeModel
Link copied to clipboard
fun SSDLikeModel(internalModel: OnnxInferenceModel, metadata: SSDLikeModelMetadata, modelKindDescription: String? = null)
Content copied to clipboard
Functions
convert
Link copied to clipboard
open override fun convert(output: OrtSession.Result): List<DetectedObject>
Content copied to clipboard
Converts raw model output to the result.
detectObjects
Link copied to clipboard
Returns the detected object for the given image sorted by the score.
initializeWith
Link copied to clipboard
open override fun initializeWith(vararg executionProviders: ExecutionProvider)
Content copied to clipboard
Initialize the model with the specified executions providers.
predictSoftly
Link copied to clipboard
open override fun predictSoftly(inputData: FloatArray, predictionTensorName: String): FloatArray
Content copied to clipboard
summary
Link copied to clipboard
Properties
inputDimensions
Link copied to clipboard
internalModel
Link copied to clipboard
modelKindDescription
Link copied to clipboard
preprocessing
Link copied to clipboard
open override val preprocessing: Operation<Bitmap, Pair<FloatArray, TensorShape>>
Content copied to clipboard
targetRotation
Link copied to clipboard