Monitor Class Reference

A simple interface class for monitoring the outcomes of a perceived scene. More...

Inheritance diagram for Monitor:

Collaboration diagram for Monitor:

Public Member Functions
def	__init__ (self, theParams, thePerceiver, theActivity, theReporter=None)
	Constructor for the perceiver.monitor 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, dbState=None)
	Display the debug state. More...
def	displayState (self, dState=None)
	Display the perceiver state and activity state per configuration specification. More...
def	emptyState (self)
	Returns an empty activity state structure. More...
def	free (self)
	Destructor. More...
def	getState (self)
	Returns the current activity 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)
	Run activity detection process to generate activity state measurement. More...
def	predict (self)
	Predict next measurement, if applicable. More...
def	process (self, I)
	Run perceive + activity recognize pipeline for one step/image measurement. More...
def	setState (self, nstate)
	Sets the state of the tracker. More...

Public Attributes
	activity
	Activity detection/recognition instance. More...
	params
	perceiver
	Perceiver instance. More...
	reporter
	Takes activity outcomes and creates report out. More...

Detailed Description

A simple interface class for monitoring the outcomes of a perceived scene.

A Monitor tacks on some form of action recognition to make sense of the perceiver information. If the intent is to compare against some target state or establish completion, then take a look at the Progress subclass. It tacks on a target/goal state and includes a progress estimator.

The choice was made to have the Monitor class independent of the Perceiver class (i.e., not a derived class of it). Rather it will receive a Perceiver instance and snag the signals from it as needed. Doing so permits a little more flexibility regarding how the actual implementation operates, which may be a future need. It is still put in the perceiver package based on the implicit messaging of the "perceiver" name.

Constructor & Destructor Documentation

__init__()

def __init__	(		self,
			theParams,
			thePerceiver,
			theActivity,
			theReporter = None
	)

Constructor for the perceiver.monitor class.

Parameters

[in]	theParams	Option set of paramters.
[in]	thePerceiver	Perceiver instance (or possibly not).
[in]	theActivity	Activity detector/recognizer.
[in]	theReporter	Reporting mechanism for activity outputs.

Member Function Documentation

adapt()

def adapt

(

self

)

Adapt parts of the process based on measurements and corrections.

Currently the perceiver and activty detector have already run their version of adapt by this point. Overload this class should that not be the case.

correct()

def correct

(

self

)

Correct the estimated state based on measured and predicted.

Currently the perceiver and activty detector have already run their version of correct by this point. Overload this class should that not be the case.

displayDebug()

def displayDebug	(		self,
			dbState = None
	)

Display the debug state.

Punts to contained instances.

displayState()

def displayState	(		self,
			dState = None
	)

Display the perceiver state and activity state per configuration specification.

Parameters

[in]

dState

Monitor state to display (optional). Default is current state.

emptyState()

def emptyState

(

self

)

Returns an empty activity state structure.

Snags default empty state from the activity instance.

Returns

Empty activity state structure.

free()

def free

(

self

)

Destructor.

Just in case other stuff needs to be done.

getState()

def getState

(

self

)

Returns the current activity state structure.

Default is to pass request along to the activity detector.

Parameters

cstate

The current state structure.

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
	)

Run activity detection process to generate activity state measurement.

If perceiver has no measurement/observation, then does nothing.

Parameters

[in]

I

Image to process. Depending on implementation, might be optional.

predict()

def predict

(

self

)

Predict next measurement, if applicable.

Todo:

Is this proper way to conceive of monitor predict?

process()

def process	(		self,
			I
	)

Run perceive + activity recognize pipeline for one step/image measurement.

setState()

def setState	(		self,
			nstate
	)

Sets the state of the tracker.

Sends passed state along to the activity detector.

Parameters

[in]

nstate

The new state structure.

Member Data Documentation

activity

activity

Activity detection/recognition instance.

params

params

perceiver

perceiver

Perceiver instance.

reporter

reporter

Takes activity outcomes and creates report out.

---

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

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