Camera geometry blocks

These blocks model camera projection functions.

class Camera(*args: Any, **kwargs: Any)

Bases: FunctionBlock

CAMERA

Camera projection block.

Parameters:

• camera (Camera subclass, optional) – Camera model, defaults to None

• blockargs (dict) – Additional options for the block, passed as a dictionary

Returns:

a CAMERA block

Return type:

Camera instance

Camera projection model.

Block ports

input pose:

Camera pose as an SE3 object.

input P:

world points as ndarray(3,N)

output p:

image plane points as ndarray(2,N)

__init__(camera=None, args={}, **blockargs)

Parameters:

• camera (Camera subclass, optional) – Camera model, defaults to None

• blockargs (dict) – Additional options for the block, passed as a dictionary

Returns:

a CAMERA block

Return type:

Camera instance

Camera projection model.

Block ports

input pose:

Camera pose as an SE3 object.

input P:

world points as ndarray(3,N)

output p:

image plane points as ndarray(2,N)

class Visjac_p(*args: Any, **kwargs: Any)

Bases: FunctionBlock

VISJAC_P

Interaction-matrix block for image points.

Parameters:

• camera (Camera subclass, optional) – Camera model, defaults to None

• depth (float or ndarray) – Point depth

• depthest (bool, optional) – Use depth estimation, defaults to True

• blockargs (dict) – Additional options for the block, passed as a dictionary

Returns:

a VISJAC_P block

Return type:

Visjac_p instance

If the Jacobian

__init__(camera, depth=1, depthest=False, **blockargs)

Parameters:

• camera (Camera subclass, optional) – Camera model, defaults to None

• depth (float or ndarray) – Point depth

• depthest (bool, optional) – Use depth estimation, defaults to True

• blockargs (dict) – Additional options for the block, passed as a dictionary

Returns:

a VISJAC_P block

Return type:

Visjac_p instance

If the Jacobian

class EstPose_p(*args: Any, **kwargs: Any)

Bases: FunctionBlock

ESTPOSE_P

Pose-estimation block for image points.

Parameters:

• camera (Camera subclass, optional) – Camera model, defaults to None

• P (ndarray(2,N)) – World point coordinates

• frame (str, optional) – return pose of points with respect to reference frame which is one of: ‘world’ [default] or ‘camera’

• method (str, optional) – pose estimation algorithm one of: ‘iterative’ [default], ‘epnp’, ‘p3p’, ‘ap3p’, ‘ippe’, ‘ippe-square’

• blockargs (dict) – Additional options for the block, passed as a dictionary

Returns:

a ESTPOSE_P block

Return type:

EstPose_p instance

__init__(camera, P, frame='world', method='iterative', **blockargs)

Parameters:

• camera (Camera subclass, optional) – Camera model, defaults to None

• P (ndarray(2,N)) – World point coordinates

• frame (str, optional) – return pose of points with respect to reference frame which is one of: ‘world’ [default] or ‘camera’

• method (str, optional) – pose estimation algorithm one of: ‘iterative’ [default], ‘epnp’, ‘p3p’, ‘ap3p’, ‘ippe’, ‘ippe-square’

• blockargs (dict) – Additional options for the block, passed as a dictionary

Returns:

a ESTPOSE_P block

Return type:

EstPose_p instance

class ImagePlane(*args: Any, **kwargs: Any)

Bases: GraphicsBlock

IMAGEPLANE

Graphics block that displays points on the image plane.

Create a block that plots image plane coordinates.

Parameters:

• camera (Camera instance) – a camera model

• style (str or dict, list of strings or dicts; one per line, optional) – styles for each point to be plotted

• grid (bool or sequence) – draw a grid, defaults to True. Can be boolean or a tuple of options for grid()

• retain (bool, optional) – keep previous image plane points, defaults to False

• watch (bool, optional) – add these signals to the watchlist, defaults to False

• init (callable, optional) – function to initialize the graphics, defaults to None

• blockargs (dict) – Additional options for the block, passed as a dictionary

Returns:

An IMAGEPLANE block

Return type:

ImagePlane instance

Create a block that plots points on a camera object’s virtual image plane.

Examples:

SCOPE()
SCOPE(nin=2)
SCOPE(nin=2, scale=[-1,2])
SCOPE(styles='k--')
SCOPE(styles=[{'color': 'blue'}, {'color': 'red', 'linestyle': '--'}])
SCOPE(styles=['k', 'r--'])

__init__(camera, style=None, labels=None, grid=True, retain=False, watch=False, init=None, **blockargs)

Create a block that plots image plane coordinates.

Parameters:

• camera (Camera instance) – a camera model

• style (str or dict, list of strings or dicts; one per line, optional) – styles for each point to be plotted

• grid (bool or sequence) – draw a grid, defaults to True. Can be boolean or a tuple of options for grid()

• retain (bool, optional) – keep previous image plane points, defaults to False

• watch (bool, optional) – add these signals to the watchlist, defaults to False

• init (callable, optional) – function to initialize the graphics, defaults to None

• blockargs (dict) – Additional options for the block, passed as a dictionary

Returns:

An IMAGEPLANE block

Return type:

ImagePlane instance

Create a block that plots points on a camera object’s virtual image plane.

Examples:

SCOPE()
SCOPE(nin=2)
SCOPE(nin=2, scale=[-1,2])
SCOPE(styles='k--')
SCOPE(styles=[{'color': 'blue'}, {'color': 'red', 'linestyle': '--'}])
SCOPE(styles=['k', 'r--'])