cv::Volume Class Reference

#include <opencv2/3d/volume.hpp>

Collaboration diagram for cv::Volume:

Public Types
enum	BoundingBoxPrecision {   VOLUME_UNIT = 0 ,   VOXEL = 1 }

Public Member Functions
	Volume (VolumeType vtype=VolumeType::TSDF, const VolumeSettings &settings=VolumeSettings(VolumeType::TSDF))
	Constructor of custom volume.
	~Volume ()
void	fetchNormals (InputArray points, OutputArray normals) const
	Extract the all data from volume.
void	fetchPointsNormals (OutputArray points, OutputArray normals) const
	Extract the all data from volume.
void	fetchPointsNormalsColors (OutputArray points, OutputArray normals, OutputArray colors) const
	Extract the all data from volume.
void	getBoundingBox (OutputArray bb, int precision) const
	Gets bounding box in volume coordinates with given precision: VOLUME_UNIT - up to volume unit VOXEL - up to voxel (currently not supported)
bool	getEnableGrowth () const
	Returns if new volume units are allocated during integration or not. Makes sense for HashTSDF only.
size_t	getTotalVolumeUnits () const
	return number of volume units in volume.
int	getVisibleBlocks () const
	return visible blocks in volume.
void	integrate (const OdometryFrame &frame, InputArray pose)
	Integrates the input data to the volume.
void	integrate (InputArray depth, InputArray image, InputArray pose)
	Integrates the input data to the volume.
void	integrate (InputArray depth, InputArray pose)
	Integrates the input data to the volume.
void	raycast (InputArray cameraPose, int height, int width, InputArray K, OutputArray points, OutputArray normals) const
	Renders the volume contents into an image. The resulting points and normals are in camera's coordinate system.
void	raycast (InputArray cameraPose, int height, int width, InputArray K, OutputArray points, OutputArray normals, OutputArray colors) const
	Renders the volume contents into an image. The resulting points and normals are in camera's coordinate system.
void	raycast (InputArray cameraPose, OutputArray points, OutputArray normals) const
	Renders the volume contents into an image. The resulting points and normals are in camera's coordinate system.
void	raycast (InputArray cameraPose, OutputArray points, OutputArray normals, OutputArray colors) const
	Renders the volume contents into an image. The resulting points and normals are in camera's coordinate system.
void	reset ()
	clear all data in volume.
void	setEnableGrowth (bool v)
	Enables or disables new volume unit allocation during integration. Makes sense for HashTSDF only.

Member Enumeration Documentation

BoundingBoxPrecision

enum cv::Volume::BoundingBoxPrecision

Enumerator
VOLUME_UNIT
VOXEL

Constructor & Destructor Documentation

Volume()

cv::Volume::Volume ( VolumeType vtype = VolumeType::TSDF, const VolumeSettings & settings = VolumeSettings(VolumeType::TSDF) )

explicit

Python:
	cv.Volume(	[, vtype[, settings]]	) ->	<Volume object>

Constructor of custom volume.

Parameters

vtype	the volume type [TSDF, HashTSDF, ColorTSDF].
settings	the custom settings for volume.

~Volume()

cv::Volume::~Volume

(

)

Member Function Documentation

fetchNormals()

void cv::Volume::fetchNormals	(	InputArray	points,
		OutputArray	normals ) const

Python:
	cv.Volume.fetchNormals(	points[, normals]	) ->	normals

Extract the all data from volume.

Parameters

points	the input exist point.
normals	the storage of normals (corresponding to input points) in the image.

fetchPointsNormals()

void cv::Volume::fetchPointsNormals	(	OutputArray	points,
		OutputArray	normals ) const

Python:
	cv.Volume.fetchPointsNormals(	[, points[, normals]]	) ->	points, normals

Extract the all data from volume.

Parameters

points	the storage of all points.
normals	the storage of all normals, corresponding to points.

fetchPointsNormalsColors()

void cv::Volume::fetchPointsNormalsColors	(	OutputArray	points,
		OutputArray	normals,
		OutputArray	colors ) const

Python:
	cv.Volume.fetchPointsNormalsColors(	[, points[, normals[, colors]]]	) ->	points, normals, colors

Extract the all data from volume.

Parameters

points	the storage of all points.
normals	the storage of all normals, corresponding to points.
colors	the storage of all colors, corresponding to points (only for ColorTSDF).

getBoundingBox()

void cv::Volume::getBoundingBox	(	OutputArray	bb,
		int	precision ) const

Python:
	cv.Volume.getBoundingBox(	precision[, bb]	) ->	bb

Gets bounding box in volume coordinates with given precision: VOLUME_UNIT - up to volume unit VOXEL - up to voxel (currently not supported)

Parameters

bb	6-float 1d array containing (min_x, min_y, min_z, max_x, max_y, max_z) in volume coordinates
precision	bounding box calculation precision

getEnableGrowth()

bool cv::Volume::getEnableGrowth

(

)

const

Python:
	cv.Volume.getEnableGrowth(		) ->	retval

Returns if new volume units are allocated during integration or not. Makes sense for HashTSDF only.

getTotalVolumeUnits()

size_t cv::Volume::getTotalVolumeUnits

(

)

const

Python:
	cv.Volume.getTotalVolumeUnits(		) ->	retval

return number of volume units in volume.

getVisibleBlocks()

int cv::Volume::getVisibleBlocks

(

)

const

Python:
	cv.Volume.getVisibleBlocks(		) ->	retval

return visible blocks in volume.

integrate() [1/3]

void cv::Volume::integrate	(	const OdometryFrame &	frame,
		InputArray	pose )

Python:
	cv.Volume.integrate(	depth, pose	) ->	None
	cv.Volume.integrateColor(	depth, image, pose	) ->	None
	cv.Volume.integrateFrame(	frame, pose	) ->	None

Integrates the input data to the volume.

Camera intrinsics are taken from volume settings structure.

Parameters

frame	the object from which to take depth and image data. For color TSDF a depth data should be registered with color data, i.e. have the same intrinsics & camera pose. This can be done using function registerDepth() from 3d module.
pose	the pose of camera in global coordinates.

integrate() [2/3]

void cv::Volume::integrate	(	InputArray	depth,
		InputArray	image,
		InputArray	pose )

Python:
	cv.Volume.integrate(	depth, pose	) ->	None
	cv.Volume.integrateColor(	depth, image, pose	) ->	None
	cv.Volume.integrateFrame(	frame, pose	) ->	None

Integrates the input data to the volume.

Camera intrinsics are taken from volume settings structure.

Parameters

depth	the depth image.
image	the color image (only for ColorTSDF). For color TSDF a depth data should be registered with color data, i.e. have the same intrinsics & camera pose. This can be done using function registerDepth() from 3d module.
pose	the pose of camera in global coordinates.

integrate() [3/3]

void cv::Volume::integrate	(	InputArray	depth,
		InputArray	pose )

Python:
	cv.Volume.integrate(	depth, pose	) ->	None
	cv.Volume.integrateColor(	depth, image, pose	) ->	None
	cv.Volume.integrateFrame(	frame, pose	) ->	None

Integrates the input data to the volume.

Camera intrinsics are taken from volume settings structure.

Parameters

depth	the depth image.
pose	the pose of camera in global coordinates.

raycast() [1/4]

void cv::Volume::raycast	(	InputArray	cameraPose,
		int	height,
		int	width,
		InputArray	K,
		OutputArray	points,
		OutputArray	normals ) const

Python:
	cv.Volume.raycast(	cameraPose[, points[, normals]]	) ->	points, normals
	cv.Volume.raycastColor(	cameraPose[, points[, normals[, colors]]]	) ->	points, normals, colors
	cv.Volume.raycastEx(	cameraPose, height, width, K[, points[, normals]]	) ->	points, normals
	cv.Volume.raycastExColor(	cameraPose, height, width, K[, points[, normals[, colors]]]	) ->	points, normals, colors

Renders the volume contents into an image. The resulting points and normals are in camera's coordinate system.

Rendered image size and camera intrinsics are taken from volume settings structure.

Parameters

cameraPose	the pose of camera in global coordinates.
height	the height of result image
width	the width of result image
K	camera raycast intrinsics
points	image to store rendered points.
normals	image to store rendered normals corresponding to points.

raycast() [2/4]

void cv::Volume::raycast	(	InputArray	cameraPose,
		int	height,
		int	width,
		InputArray	K,
		OutputArray	points,
		OutputArray	normals,
		OutputArray	colors ) const

Python:
	cv.Volume.raycast(	cameraPose[, points[, normals]]	) ->	points, normals
	cv.Volume.raycastColor(	cameraPose[, points[, normals[, colors]]]	) ->	points, normals, colors
	cv.Volume.raycastEx(	cameraPose, height, width, K[, points[, normals]]	) ->	points, normals
	cv.Volume.raycastExColor(	cameraPose, height, width, K[, points[, normals[, colors]]]	) ->	points, normals, colors

Renders the volume contents into an image. The resulting points and normals are in camera's coordinate system.

Rendered image size and camera intrinsics are taken from volume settings structure.

Parameters

cameraPose	the pose of camera in global coordinates.
height	the height of result image
width	the width of result image
K	camera raycast intrinsics
points	image to store rendered points.
normals	image to store rendered normals corresponding to points.
colors	image to store rendered colors corresponding to points (only for ColorTSDF).

raycast() [3/4]

void cv::Volume::raycast	(	InputArray	cameraPose,
		OutputArray	points,
		OutputArray	normals ) const

Python:
	cv.Volume.raycast(	cameraPose[, points[, normals]]	) ->	points, normals
	cv.Volume.raycastColor(	cameraPose[, points[, normals[, colors]]]	) ->	points, normals, colors
	cv.Volume.raycastEx(	cameraPose, height, width, K[, points[, normals]]	) ->	points, normals
	cv.Volume.raycastExColor(	cameraPose, height, width, K[, points[, normals[, colors]]]	) ->	points, normals, colors

Renders the volume contents into an image. The resulting points and normals are in camera's coordinate system.

Rendered image size and camera intrinsics are taken from volume settings structure.

Parameters

cameraPose	the pose of camera in global coordinates.
points	image to store rendered points.
normals	image to store rendered normals corresponding to points.

raycast() [4/4]

void cv::Volume::raycast	(	InputArray	cameraPose,
		OutputArray	points,
		OutputArray	normals,
		OutputArray	colors ) const

Python:
	cv.Volume.raycast(	cameraPose[, points[, normals]]	) ->	points, normals
	cv.Volume.raycastColor(	cameraPose[, points[, normals[, colors]]]	) ->	points, normals, colors
	cv.Volume.raycastEx(	cameraPose, height, width, K[, points[, normals]]	) ->	points, normals
	cv.Volume.raycastExColor(	cameraPose, height, width, K[, points[, normals[, colors]]]	) ->	points, normals, colors

Renders the volume contents into an image. The resulting points and normals are in camera's coordinate system.

Rendered image size and camera intrinsics are taken from volume settings structure.

Parameters

cameraPose	the pose of camera in global coordinates.
points	image to store rendered points.
normals	image to store rendered normals corresponding to points.
colors	image to store rendered colors corresponding to points (only for ColorTSDF).

reset()

void cv::Volume::reset

(

)

Python:
	cv.Volume.reset(		) ->	None

clear all data in volume.

setEnableGrowth()

void cv::Volume::setEnableGrowth

(

bool

v

)

Python:
	cv.Volume.setEnableGrowth(	v	) ->	None

Enables or disables new volume unit allocation during integration. Makes sense for HashTSDF only.

---

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

• opencv2/3d/volume.hpp