cv::kinfu::KinFu Class Reference

KinectFusion implementation. More...

#include <opencv2/rgbd/kinfu.hpp>

Collaboration diagram for cv::kinfu::KinFu:

Public Member Functions
virtual	~KinFu ()
virtual void	getCloud (OutputArray points, OutputArray normals) const =0
	Gets points and normals of current 3d mesh.
virtual void	getNormals (InputArray points, OutputArray normals) const =0
	Calculates normals for given points.
virtual const Params &	getParams () const =0
	Get current parameters.
virtual void	getPoints (OutputArray points) const =0
	Gets points of current 3d mesh.
virtual Affine3f	getPose () const =0
	Get current pose in voxel space.
virtual void	render (OutputArray image) const =0
	Renders a volume into an image.
virtual void	render (OutputArray image, const Matx44f &cameraPose) const =0
	Renders a volume into an image.
virtual void	reset ()=0
	Resets the algorithm.
virtual bool	update (InputArray depth)=0
	Process next depth frame.

Static Public Member Functions
static Ptr< KinFu >	create (const Ptr< Params > &_params)

Detailed Description

KinectFusion implementation.

This class implements a 3d reconstruction algorithm described in [134] paper.

It takes a sequence of depth images taken from depth sensor (or any depth images source such as stereo camera matching algorithm or even raymarching renderer). The output can be obtained as a vector of points and their normals or can be Phong-rendered from given camera pose.

An internal representation of a model is a voxel cuboid that keeps TSDF values which are a sort of distances to the surface (for details read the [134] article about TSDF). There is no interface to that representation yet.

KinFu uses OpenCL acceleration automatically if available. To enable or disable it explicitly use cv::setUseOptimized() or cv::ocl::setUseOpenCL().

This implementation is based on kinfu-remake.

Note that the KinectFusion algorithm was patented and its use may be restricted by the list of patents mentioned in README.md file in this module directory.

That's why you need to set the OPENCV_ENABLE_NONFREE option in CMake to use KinectFusion.

Constructor & Destructor Documentation

~KinFu()

virtual cv::kinfu::KinFu::~KinFu ( )

virtual

Member Function Documentation

create()

static Ptr< KinFu > cv::kinfu::KinFu::create ( const Ptr< Params > & _params )

static

Python:
	cv.kinfu.KinFu.create(	_params	) ->	retval
	cv.kinfu.KinFu_create(	_params	) ->	retval

getCloud()

virtual void cv::kinfu::KinFu::getCloud ( OutputArray points, OutputArray normals ) const

pure virtual

Python:
	cv.kinfu.KinFu.getCloud(	[, points[, normals]]	) ->	points, normals

Gets points and normals of current 3d mesh.

The order of normals corresponds to order of points. The order of points is undefined.

Parameters

points	vector of points which are 4-float vectors
normals	vector of normals which are 4-float vectors

getNormals()

virtual void cv::kinfu::KinFu::getNormals ( InputArray points, OutputArray normals ) const

pure virtual

Python:
	cv.kinfu.KinFu.getNormals(	points[, normals]	) ->	normals

Calculates normals for given points.

Parameters

points	input vector of points which are 4-float vectors
normals	output vector of corresponding normals which are 4-float vectors

getParams()

virtual const Params & cv::kinfu::KinFu::getParams ( ) const

pure virtual

Get current parameters.

getPoints()

virtual void cv::kinfu::KinFu::getPoints ( OutputArray points ) const

pure virtual

Python:
	cv.kinfu.KinFu.getPoints(	[, points]	) ->	points

Gets points of current 3d mesh.

The order of points is undefined.

Parameters

points

vector of points which are 4-float vectors

getPose()

virtual Affine3f cv::kinfu::KinFu::getPose ( ) const

pure virtual

Get current pose in voxel space.

render() [1/2]

virtual void cv::kinfu::KinFu::render ( OutputArray image ) const

pure virtual

Python:
	cv.kinfu.KinFu.render(	[, image]	) ->	image
	cv.kinfu.KinFu.render(	cameraPose[, image]	) ->	image

Renders a volume into an image.

Renders a 0-surface of TSDF using Phong shading into a CV_8UC4 Mat. Light pose is fixed in KinFu params.

Parameters

image

resulting image

render() [2/2]

virtual void cv::kinfu::KinFu::render ( OutputArray image, const Matx44f & cameraPose ) const

pure virtual

Python:
	cv.kinfu.KinFu.render(	[, image]	) ->	image
	cv.kinfu.KinFu.render(	cameraPose[, image]	) ->	image

Renders a volume into an image.

Renders a 0-surface of TSDF using Phong shading into a CV_8UC4 Mat. Light pose is fixed in KinFu params.

Parameters

image	resulting image
cameraPose	pose of camera to render from. If empty then render from current pose which is a last frame camera pose.

reset()

virtual void cv::kinfu::KinFu::reset ( )

pure virtual

Python:
	cv.kinfu.KinFu.reset(		) ->	None

Resets the algorithm.

Clears current model and resets a pose.

update()

virtual bool cv::kinfu::KinFu::update ( InputArray depth )

pure virtual

Python:
	cv.kinfu.KinFu.update(	depth	) ->	retval

Process next depth frame.

Integrates depth into voxel space with respect to its ICP-calculated pose. Input image is converted to CV_32F internally if has another type.

Parameters

depth

one-channel image which size and depth scale is described in algorithm's parameters

Returns

true if succeeded to align new frame with current scene, false if opposite

---

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

• opencv2/rgbd/kinfu.hpp