drwnTextonFilterBank Class Reference

Implements a 17-dimensional filter bank. More...

Public Member Functions
	drwnTextonFilterBank (double k=1.0)
	construct a filterbank with bandwidth k
void	filter (const cv::Mat &img, std::vector< cv::Mat > &response) const
	Filtering function. The caller must provide a vector of CV32F destination matrices. The source image should be a 3-channel RGB color image (it is automatically converted to CIELab).

Static Public Attributes
static const int	NUM_FILTERS = 17
	number of filters in the filterbank

Protected Attributes
double	_kappa
	base filter bandwidth

Detailed Description

Implements a 17-dimensional filter bank.

The filter bank is defined in:

• J. Shotton, J. Winn, C. Rother, A. Criminisi. "TextonBoost for Image Understanding: Multi-Class Object Recognition and Segmentation by Jointly Modeling Texture, Layout, and Context," IJCV 2008.
• J. Winn, A. Criminisi, and T. Minka. "Categorization by learned universal visual dictionary," ICCV 2005.

The features produced by this filterbank are used in multi-class image segmentation (see Multi-class Image Segmentation).

See Also

drwnFilterBankResponse

---

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

• drwnTextonFilterBank.h
• drwnTextonFilterBank.cpp