Struct UVDocModel 

pub struct UVDocModel { /* private fields */ }

Pure UVDoc model implementation.

This model performs document rectification (unwarping) on distorted document images.

Implementations

impl UVDocModel

pub fn new( inference: OrtInfer, normalizer: NormalizeImage, postprocessor: UVDocPostProcess, rec_image_shape: [usize; 3], ) -> UVDocModel

Creates a new UVDoc model.

pub fn preprocess( &self, images: Vec<ImageBuffer<Rgb<u8>, Vec<u8>>>, ) -> Result<(ArrayBase<OwnedRepr<f32>, Dim<[usize; 4]>>, Vec<(u32, u32)>), OCRError>

Preprocesses images for rectification.

Arguments

• images - Input images to preprocess

Returns

A tuple of (batch_tensor, original_sizes)

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

Runs inference on the preprocessed batch.

Arguments

• batch_tensor - Preprocessed batch tensor

Returns

Model predictions as a 4D tensor

pub fn postprocess( &self, predictions: &ArrayBase<OwnedRepr<f32>, Dim<[usize; 4]>>, original_sizes: &[(u32, u32)], ) -> Result<Vec<ImageBuffer<Rgb<u8>, Vec<u8>>>, OCRError>

Postprocesses model predictions to rectified images.

Arguments

• predictions - Model predictions
• original_sizes - Original image sizes (width, height)

Returns

Rectified images resized to original dimensions

pub fn forward( &self, images: Vec<ImageBuffer<Rgb<u8>, Vec<u8>>>, ) -> Result<UVDocModelOutput, OCRError>

Performs complete forward pass: preprocess -> infer -> postprocess.

Arguments

• images - Input images to rectify

Returns

UVDocModelOutput containing rectified images

Trait Implementations

impl Debug for UVDocModel

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

Formats the value using the given formatter.