superviseddescent Namespace Reference

Classes
class	ColPivHouseholderQRSolver
class	LinearRegressor
class	NoNormalisation
class	PartialPivLUSolver
class	Regressor
class	Regulariser
class	SupervisedDescentOptimiser

Functions
void	no_eval (const cv::Mat &current_predictions)

Detailed Description

The main namespace of the supervised descent library.

superviseddescent is a C++11 implementation of the supervised descent method, which is a generic algorithm to perform optimisation of arbitrary functions. The basic idea is to learn the gradient direction of a function from data using a series of regressors. As the gradient direction is learned directly from data, the function does not have to be differentiable.

The theory is based on the idea of [1] Supervised Descent Method and Its Applications to Face Alignment, from X. Xiong & F. De la Torre, CVPR 2013.

Function Documentation

void superviseddescent::no_eval ( const cv::Mat &  current_predictions )

inline

A callback function that gets called after each training of a regressor. This is the default function that performs nothing.

Parameters

[in]

current_predictions

Predictions of the current regressor.