VART-X  0.3.0
vart::InferResultImplBase Class Referenceabstract

Abstract base class for inference result implementation used by InferResult. More...

#include <vart_inferresult_impl_base.hpp>

Public Member Functions

 InferResultImplBase ()=default
 Default construction is allowed for derived implementations. More...
 
virtual ~InferResultImplBase ()
 Virtual destructor to ensure proper cleanup of derived implementations. More...
 
virtual void transform (InferResScaleInfo &info)=0
 transform() - Transforms the infer results to the input frame resolution. More...
 
virtual InferResultDataget_infer_result ()=0
 get_infer_result() - Get inference results for a frame. More...
 

Detailed Description

Abstract base class for inference result implementation used by InferResult.

Note
Applications should use vart::InferResult rather than this type directly, unless extending or implementing a new result type.

Constructor & Destructor Documentation

◆ InferResultImplBase()

vart::InferResultImplBase::InferResultImplBase ( )
default

Default construction is allowed for derived implementations.

◆ ~InferResultImplBase()

virtual vart::InferResultImplBase::~InferResultImplBase ( )
inlinevirtual

Virtual destructor to ensure proper cleanup of derived implementations.

Member Function Documentation

◆ get_infer_result()

virtual InferResultData* vart::InferResultImplBase::get_infer_result ( )
pure virtual

get_infer_result() - Get inference results for a frame.

Returns
Pointer to inference-result data owned by the implementation. The returned pointer's lifetime is tied to the implementation instance; callers must not free it or use it past the lifetime of the owning InferResult.

◆ transform()

virtual void vart::InferResultImplBase::transform ( InferResScaleInfo info)
pure virtual

transform() - Transforms the infer results to the input frame resolution.

Parameters
[in]infoWidth and height info containing model input dimensions and input frame dimensions used to transform the infer results.

If we have inference results, like detection, on a 200x200 model input and want to draw bounding boxes on the original frame, which may be 1920x1080, this function transforms the bounding box dimensions in the infer results to the resolution of the input frame.


The documentation for this class was generated from the following file: