BatchParameters Class Reference

Structure containing a set of parameters for batch object detection. More...

Functions
	BatchParameters (bool enable=false, float id_retention_time=240.f, float batch_duration=2.f)
	Default constructor. More...
bool	operator== (const BatchParameters &param1) const
bool	operator!= (const BatchParameters &param1) const

Attributes
bool	enable = false
	Whether to enable the batch option in the object detection module. More...
float	id_retention_time = 240
	Max retention time in seconds of a detected object. More...
float	latency = 2.f
	Trajectories will be output in batch with the desired latency in seconds. More...

Detailed Description

Structure containing a set of parameters for batch object detection.

The default constructor sets all parameters to their default settings.

Note

Parameters can be adjusted by the user.

Constructor and Destructor

BatchParameters()

BatchParameters	(	bool	enable = false,
		float	id_retention_time = 240.f,
		float	batch_duration = 2.f
	)

Default constructor.

All the parameters are set to their default values.

Functions

operator==()

bool operator==

(

const BatchParameters &

param1

)

const

Comparison operator ==

Parameters

BatchParameters

to compare

Returns

true if the two struct are identical

operator!=()

bool operator!=

(

const BatchParameters &

param1

)

const

Comparison operator !=

Parameters

BatchParameters

to compare

Returns

true if the two struct are different

Variables

enable

bool enable = false

Whether to enable the batch option in the object detection module.

Batch queueing system provides:

• deep-learning based re-identification
• trajectory smoothing and filtering

Note

To activate this option, enable must be set to true.

id_retention_time

float id_retention_time = 240

Max retention time in seconds of a detected object.

After this time, the same object will mostly have a different id.

latency

float latency = 2.f

Trajectories will be output in batch with the desired latency in seconds.

During this waiting time, re-identification of objects is done in the background.

Note

Specifying a short latency will limit the search (falling in timeout) for previously seen object ids but will be closer to real time output.

Specifying a long latency will reduce the change of timeout in re-identification but increase difference with live output.