Perceiver Class Reference

Basic implementation of a perceiver class. More...

Inheritance diagram for Perceiver:

Collaboration diagram for Perceiver:

Public Member Functions
def	__init__ (self, theParams, theDetector, theTracker, trackFilter)
	Constructor for the Perceiver class. More...
def	adapt (self)
	Adapt parts of the process based on measurements and corrections. More...
def	correct (self)
	Correct the estimated state based on measured and predicted. More...
def	displayDebug (self, fh, dbState)
	Display the debug state of the Perceiver. More...
def	displayState (self, dState=None)
	Display the current state of the Perceiver. More...
def	emptyState (self)
	Return state structure with no information. More...
def	free (self)
def	get (self, fname)
	Get the state or parameters of the tracker. More...
def	getState (self)
	Returns the current state structure. More...
def	info (self)
	Return the information structure used for saving or otherwise determining the tracker setup for reproducibility. More...
def	measure (self, I)
	Recover track point or track frame based on detector + trackPointer output. More...
def	predict (self)
	Predict next measurement, if applicable. More...
def	process (self, I)
	Run the tracking pipeline for one step/image measurement. More...
def	set (self, fname, fval)
	Set the state or parameters of the rigid body tracker. More...
def	setState (self, nstate)
	Sets the state of the tracker. More...

Static Public Member Functions
def	displayFull (cstate, dispArgs)
	Fill rigid body display routine. More...
def	displaySimple (cstate, dispArgs)
	Basic rigid body display routine. More...

Public Attributes
	detector
	The detector instance to use. More...
	filter
	The track filter to use. More...
	haveObs
	Was an observation measured? - e.g. More...
	haveRun
	Has not been run before. More...
	haveState
	Do we have a state estimate? - e.g. More...
	I
	Image passed for processing. More...
	params
	Perceiver runtime parameters. More...
	tMeas
	The last measured track state of the target. More...
	tPts
	Track points. More...
	tracker
	The track pointer to use. More...

Detailed Description

Basic implementation of a perceiver class.

A Perceiver is, at minimum, the combination of a detector and a tracker or track pointer. It indicates where a "target" is in the image both as a region and as a single point or rigid body frame. If there is a way to meaningfully filter the point/frame, then adding a trackFilter will incorporate that operation.

Constructor & Destructor Documentation

__init__()

def __init__	(		self,
			theParams,
			theDetector,
			theTracker,
			trackFilter
	)

Constructor for the Perceiver class.

Parameters

[in]	theParams	Option set of paramters.
[in]	theDetector	The binary segmentation method.
[in]	theTracker	The binary image trackpoint method.
[in]	trackFilter	The track point filtering / data association approach.

Member Function Documentation

adapt()

def adapt

(

self

)

Adapt parts of the process based on measurements and corrections.

correct()

def correct

(

self

)

Correct the estimated state based on measured and predicted.

displayDebug()

def displayDebug	(		self,
			fh,
			dbState
	)

Display the debug state of the Perceiver.

displayFull()

def displayFull (   cstate,   dispArgs  )

static

Fill rigid body display routine.

Plots SE(2) frame and marker positions.

displaySimple()

def displaySimple (   cstate,   dispArgs  )

static

Basic rigid body display routine.

Plots SE(2) frame.

displayState()

def displayState	(		self,
			dState = None
	)

Display the current state of the Perceiver.

emptyState()

def emptyState

(

self

)

Return state structure with no information.

Parameters

[out]

estate

The state structure with no content.

free()

def free

(

self

)

get()

def get	(		self,
			fname
	)

Get the state or parameters of the tracker.

Parameters

[in]	fname	Name of the field to set.
[out]	fval	Value returned.

getState()

def getState

(

self

)

Returns the current state structure.

Returns

The current state structure.

Note

May need some corrections due to uncertainty about tMeas.

info()

def info

(

self

)

Return the information structure used for saving or otherwise determining the tracker setup for reproducibility.

Parameters

[out]

tinfo

The tracking configuration information structure.

measure()

def measure	(		self,
			I
	)

Recover track point or track frame based on detector + trackPointer output.

Parameters

[in]

I

Image for generating perceived measurement.

predict()

def predict

(

self

)

Predict next measurement, if applicable.

process()

def process	(		self,
			I
	)

Run the tracking pipeline for one step/image measurement.

Parameters

[in]

I

The image to process.

set()

def set	(		self,
			fname,
			fval
	)

Set the state or parameters of the rigid body tracker.

Parameters

[in]	fname	Name of the field to set.
[in]	fval	Value to set.

setState()

def setState	(		self,
			nstate
	)

Sets the state of the tracker.

Parameters

[in]

nstate

The new state structure.

Member Data Documentation

detector

detector

The detector instance to use.

filter

filter

The track filter to use.

haveObs

haveObs

Was an observation measured? - e.g.

detect for tracker

haveRun

haveRun

Has not been run before.

haveState

haveState

Do we have a state estimate? - e.g.

tracker activity.

I

I

Image passed for processing.

params

params

Perceiver runtime parameters.

tMeas

tMeas

The last measured track state of the target.

tPts

tPts

Track points.

[from track filter?]

tracker

tracker

The track pointer to use.

---

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

• /home/pvela/python/perceiver/perceiver/perceiver.py