Struct ScaleAwareDetectorModel 

pub struct ScaleAwareDetectorModel { /* private fields */ }

Generic scale-aware object detection model.

This unified implementation supports detection models that require scale information as auxiliary inputs (PicoDet, PP-DocLayout, etc.), eliminating code duplication by parameterizing the inference mode.

Implementations

impl ScaleAwareDetectorModel

pub fn new( inference: OrtInfer, preprocess_config: ScaleAwareDetectorPreprocessConfig, inference_mode: ScaleAwareDetectorInferenceMode, ) -> Result<ScaleAwareDetectorModel, OCRError>

Creates a new scale-aware detector model.

pub fn preprocess( &self, images: Vec<ImageBuffer<Rgb<u8>, Vec<u8>>>, ) -> Result<(ArrayBase<OwnedRepr<f32>, Dim<[usize; 4]>>, Vec<ImageScaleInfo>, Vec<[f32; 2]>, Vec<[f32; 2]>), OCRError>

Preprocesses images for object detection.

Returns:

• Batch tensor ready for inference
• Image shapes after resizing [h, w, ratio_h, ratio_w]
• Original shapes [h, w]
• Resized shapes [h, w]

pub fn infer( &self, batch_tensor: &ArrayBase<OwnedRepr<f32>, Dim<[usize; 4]>>, orig_shapes: &[[f32; 2]], resized_shapes: &[[f32; 2]], ) -> Result<ArrayBase<OwnedRepr<f32>, Dim<[usize; 4]>>, OCRError>

Runs inference on the preprocessed batch tensor.

Automatically handles different inference modes:

• ScaleFactorOnly: Only passes scale_factor
• ScaleFactorAndImageShape: Passes both scale_factor and im_shape

pub fn postprocess( &self, predictions: ArrayBase<OwnedRepr<f32>, Dim<[usize; 4]>>, _config: &ScaleAwareDetectorPostprocessConfig, ) -> Result<ScaleAwareDetectorModelOutput, OCRError>

Postprocesses model predictions.

Returns raw predictions for adapter layer to process.

pub fn forward( &self, images: Vec<ImageBuffer<Rgb<u8>, Vec<u8>>>, config: &ScaleAwareDetectorPostprocessConfig, ) -> Result<(ScaleAwareDetectorModelOutput, Vec<ImageScaleInfo>), OCRError>

Runs the complete forward pass: preprocess -> infer -> postprocess.

Trait Implementations

impl Debug for ScaleAwareDetectorModel

fn fmt(&self, f: &mut Formatter<'_>) -> Result<(), Error>

Formats the value using the given formatter.