Face Engine SDK 5.25.0
A face detection, recognition and tracking engine.
Loading...
Searching...
No Matches
fsdk::IQualityEstimator Struct Referenceabstract

Image quality estimator interface. More...

#include <IQualityEstimator.h>

Inheritance diagram for fsdk::IQualityEstimator:
fsdk::IRefCounted

Public Member Functions

virtual Result< FSDKErrorestimate (const Image &warp, Quality &quality) const noexcept=0
 Estimate the quality.
 
virtual Result< FSDKErrorestimate (Span< const Image > warps, Span< Quality > quality) const noexcept=0
 Estimate the quality of multiple warped images in a single estimate function call.
 
virtual Result< FSDKErrorestimate (const Image &warp, SubjectiveQuality &quality) const noexcept=0
 Estimate the quality.
 
virtual Result< FSDKErrorestimate (Span< const Image > warps, Span< SubjectiveQuality > quality) const noexcept=0
 Estimate the subjective quality of multiple warped images in a single estimate function call.
 
virtual Result< FSDKErrorvalidate (Span< const Image > warps, Span< Result< FSDKError > > errors) const noexcept=0
 Validate input of multiple frames in a single function call.
 
- Public Member Functions inherited from fsdk::IRefCounted
virtual int32_t retain () noexcept=0
 Increase strong reference count.
 
virtual int32_t retainLocked () noexcept=0
 Increase strong reference count thread safely.
 
virtual int32_t release () noexcept=0
 Decrease strong reference count.
 
virtual int32_t getRefCount () const noexcept=0
 Get actual strong reference count.
 
virtual int32_t retainWeak () noexcept=0
 Increase weak reference count.
 
virtual int32_t releaseWeak () noexcept=0
 Decrease weak reference count.
 
virtual int32_t getWeakRefCount () const noexcept=0
 Get actual weak reference count.
 

Detailed Description

Image quality estimator interface.

Note
This estimator is designed to work with a person face image; you should pass a warped face detection image.
See also
IWarper for details. Quality estimator detects the same attributes as all the other estimators:
  • over/under exposure;
  • blurriness;
  • natural/unnatural colors; It is different in the sense that it computes all the estimations at once and returns the results merged somehow into a single value instead of several separate values. This way one can obtain a single scalar quality metric of a person face image. The estimated value is a probability that the image is good for both recognition and viewing purposes.

Member Function Documentation

◆ estimate() [1/4]

virtual Result< FSDKError > fsdk::IQualityEstimator::estimate ( const Image & warp,
Quality & quality ) const
pure virtualnoexcept

Estimate the quality.

Parameters
[in]warpimage with warped face.
[out]qualityoutput structure with quality params. Complex quality estimation available by method getQuality.
Returns
Result with error code.
See also
Quality, Image, Result and FSDKError for details.
Note
warp format must be R8G8B8,
See also
Format.

◆ estimate() [2/4]

virtual Result< FSDKError > fsdk::IQualityEstimator::estimate ( const Image & warp,
SubjectiveQuality & quality ) const
pure virtualnoexcept

Estimate the quality.

Parameters
[in]warpimage with warped face.
[out]qualityoutput structure with subjective quality params.
Returns
Result with error code.
See also
SubjectiveQuality, Image, Result and FSDKError for details.
Note
warp format must be R8G8B8,
See also
Format.

◆ estimate() [3/4]

virtual Result< FSDKError > fsdk::IQualityEstimator::estimate ( Span< const Image > warps,
Span< Quality > quality ) const
pure virtualnoexcept

Estimate the quality of multiple warped images in a single estimate function call.

Parameters
[in]warpsspan of images with warped faces.
[out]qualityoutput quality span.
Returns
Result with error code.
See also
Span, Quality, Image, Result and FSDKError for details.
Note
warps format must be R8G8B8,
See also
Format.
Note
all spans should be based on user owned continuous collections.
all spans should be equal size.

◆ estimate() [4/4]

virtual Result< FSDKError > fsdk::IQualityEstimator::estimate ( Span< const Image > warps,
Span< SubjectiveQuality > quality ) const
pure virtualnoexcept

Estimate the subjective quality of multiple warped images in a single estimate function call.

Parameters
[in]warpsspan of images with warped faces.
[out]qualityoutput SubjectiveQuality span.
Returns
Result with error code.
See also
Span, SubjectiveQuality, Image, Result and FSDKError for details.
Note
warps format must be R8G8B8,
See also
Format.
Note
all spans should be based on user owned continuous collections.
all spans should be equal size.

◆ validate()

virtual Result< FSDKError > fsdk::IQualityEstimator::validate ( Span< const Image > warps,
Span< Result< FSDKError > > errors ) const
pure virtualnoexcept

Validate input of multiple frames in a single function call.

Parameters
[in]warpsspan of images with warped faces.
[out]errorsoutput span of errors for each image.
Returns
Result with error code.
See also
Span, Image, Result and FSDKError for details.
Note
warps format must be R8G8B8,
See also
Format.
Note
all spans should be based on user owned continuous collections.
all spans should be equal size.

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