aliceVision::feature::AKAZEOptions Struct Reference

#include <AKAZE.hpp>

Public Attributes
int	nbOctaves = 4
	octave to process
int	nbSlicePerOctave = 4
	levels per octave
float	sigma0 = 1.6f
	initial sigma offset (used to suppress low level noise)
float	threshold = 0.0008f
	hessian determinant threshold
float	descFactor = 1.0f
	magnifier used to describe an interest point
std::size_t	gridSize = 4
	grid size for filtering
std::size_t	maxTotalKeypoints = 1000
	maximum number of keypoints

Detailed Description

Bibliography: [1] "Fast Explicit Diffusion for Accelerated Features in Nonlinear Scale Spaces." Conference : BMVC.

Authors

Pablo F. Alcantarilla, Jesús Nuevo and Adrien Bartoli

Date

September 2013

Note

This implementation is done for AliceVision:

• code look the same but less memory are allocated
• only Perona and Malik G2 diffusion (PM_G2) is implemented.

See also

Reference implementation (OpenCV based) by Pablo F. Alcantarilla (1), Jesus Nuevo (2)

• https://github.com/pablofdezalc/akaze Toshiba Research Europe Ltd (1) TrueVision Solutions (2)

---

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

• /home/docs/checkouts/readthedocs.org/user_builds/alicevision/checkouts/latest/src/aliceVision/feature/akaze/AKAZE.hpp