Classes
class	ObjectData
	Class containing data of a detected object such as its bounding_box, label, id and its 3D position. More...

class	CustomBoxObjectData
	Class that store externally detected objects. More...

class	Objects
	Class containing the results of the object detection module. More...

class	ObjectsBatch
	Class containing batched data of a detected objects from the object detection module. More...

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

class	ObjectDetectionParameters
	Structure containing a set of parameters for the object detection module. More...

class	ObjectDetectionRuntimeParameters
	Structure containing a set of runtime parameters for the object detection module. More...

class	AI_Model_status
	Structure containing AI model status. More...

Enumerations
enum class	OBJECT_DETECTION_MODEL
	Lists available models for the object detection module. More...

enum class	AI_MODELS
	Lists available AI models. More...

enum class	OBJECT_CLASS
	Lists available object classes. More...

enum class	OBJECT_SUBCLASS
	List available object subclasses. More...

enum class	OBJECT_TRACKING_STATE
	Lists the different states of object tracking. More...

enum class	OBJECT_ACTION_STATE
	Lists the different states of an object's actions. More...

enum class	OBJECT_FILTERING_MODE
	Lists supported bounding box preprocessing. More...

Functions
AI_Model_status	checkAIModelStatus (AI_MODELS model, int gpu_id=0)
	Checks if a corresponding optimized engine is found for the requested Model based on your rig configuration. More...

ERROR_CODE	downloadAIModel (sl::AI_MODELS model, int gpu_id=0)
	Downloads the requested model. More...

ERROR_CODE	optimizeAIModel (sl::AI_MODELS model, int gpu_id=0)
	Optimizes the requested model (and download the model if it is not present on the host). More...

String	generate_unique_id ()
	Generate a UUID like unique id to help identify and track AI detections. More...

Enumeration Type Documentation

◆ OBJECT_DETECTION_MODEL

enum OBJECT_DETECTION_MODEL

strong

Lists available models for the object detection module.

Enumerator
MULTI_CLASS_BOX_FAST	Any objects, bounding box based.
MULTI_CLASS_BOX_MEDIUM	Any objects, bounding box based, compromise between accuracy and speed.
MULTI_CLASS_BOX_ACCURATE	Any objects, bounding box based, more accurate but slower than the base model.
PERSON_HEAD_BOX_FAST	Bounding box detector specialized in person heads particularly well suited for crowded environments. The person localization is also improved.
PERSON_HEAD_BOX_ACCURATE	Bounding box detector specialized in person heads, particularly well suited for crowded environments. The person localization is also improved, more accurate but slower than the base model.
CUSTOM_BOX_OBJECTS	For external inference, using your own custom model and/or frameworks. This mode disables the internal inference engine, the 2D bounding box detection must be provided.

◆ AI_MODELS

enum AI_MODELS

strong

Lists available AI models.

Enumerator
MULTI_CLASS_DETECTION	Related to sl::OBJECT_DETECTION_MODEL::MULTI_CLASS_BOX_FAST
MULTI_CLASS_MEDIUM_DETECTION	Related to sl::OBJECT_DETECTION_MODEL::MULTI_CLASS_BOX_MEDIUM
MULTI_CLASS_ACCURATE_DETECTION	Related to sl::OBJECT_DETECTION_MODEL::MULTI_CLASS_BOX_ACCURATE
HUMAN_BODY_FAST_DETECTION	Related to sl::BODY_TRACKING_MODEL::HUMAN_BODY_FAST
HUMAN_BODY_MEDIUM_DETECTION	Related to sl::BODY_TRACKING_MODEL::HUMAN_BODY_MEDIUM
HUMAN_BODY_ACCURATE_DETECTION	Related to sl::BODY_TRACKING_MODEL::HUMAN_BODY_ACCURATE
HUMAN_BODY_38_FAST_DETECTION	Related to sl::BODY_TRACKING_MODEL::HUMAN_BODY_FAST
HUMAN_BODY_38_MEDIUM_DETECTION	Related to sl::BODY_TRACKING_MODEL::HUMAN_BODY_FAST
HUMAN_BODY_38_ACCURATE_DETECTION	Related to sl::BODY_TRACKING_MODEL::HUMAN_BODY_FAST
PERSON_HEAD_DETECTION	Related to sl::OBJECT_DETECTION_MODEL::PERSON_HEAD_BOX_FAST
PERSON_HEAD_ACCURATE_DETECTION	Related to sl::OBJECT_DETECTION_MODEL::PERSON_HEAD_BOX_ACCURATE
REID_ASSOCIATION	Related to sl::BatchParameters.enable
NEURAL_DEPTH	Related to sl::DEPTH_MODE::NEURAL
NEURAL_PLUS_DEPTH	Related to sl::DEPTH_MODE::NEURAL_PLUS

◆ OBJECT_CLASS

enum OBJECT_CLASS

strong

Lists available object classes.

Enumerator
PERSON	For people detection
VEHICLE	For vehicle detection (cars, trucks, buses, motorcycles, etc.)
BAG	For bag detection (backpack, handbag, suitcase, etc.)
ANIMAL	For animal detection (cow, sheep, horse, dog, cat, bird, etc.)
ELECTRONICS	For electronic device detection (cellphone, laptop, etc.)
FRUIT_VEGETABLE	For fruit and vegetable detection (banana, apple, orange, carrot, etc.)
SPORT	For sport-related object detection (sport ball, etc.)

◆ OBJECT_SUBCLASS

enum OBJECT_SUBCLASS

strong

List available object subclasses.

Given as hint, when using object tracking an object can change of sl::OBJECT_SUBCLASS while keeping the same sl::OBJECT_CLASS (i.e.: frame n: MOTORBIKE, frame n+1: BICYCLE).

Enumerator
PERSON	PERSON
PERSON_HEAD	PERSON
BICYCLE	VEHICLE
CAR	VEHICLE
MOTORBIKE	VEHICLE
BUS	VEHICLE
TRUCK	VEHICLE
BOAT	VEHICLE
BACKPACK	BAG
HANDBAG	BAG
SUITCASE	BAG
BIRD	ANIMAL
CAT	ANIMAL
DOG	ANIMAL
HORSE	ANIMAL
SHEEP	ANIMAL
COW	ANIMAL
CELLPHONE	ELECTRONICS
LAPTOP	ELECTRONICS
BANANA	FRUIT_VEGETABLE
APPLE	FRUIT_VEGETABLE
ORANGE	FRUIT_VEGETABLE
CARROT	FRUIT_VEGETABLE
SPORTSBALL	SPORT

◆ OBJECT_TRACKING_STATE

enum OBJECT_TRACKING_STATE

strong

Lists the different states of object tracking.

Enumerator
OFF	The tracking is not yet initialized. The object id is not usable.
OK	The object is tracked.
SEARCHING	The object could not be detected in the image and is potentially occluded. The trajectory is estimated.
TERMINATE	This is the last searching state of the track. The track will be deleted in the next sl::Camera.retrieveObjects().

◆ OBJECT_ACTION_STATE

enum OBJECT_ACTION_STATE

strong

Lists the different states of an object's actions.

Enumerator
IDLE	The object is staying static.
MOVING	The object is moving.

◆ OBJECT_FILTERING_MODE

enum OBJECT_FILTERING_MODE

strong

Lists supported bounding box preprocessing.

Enumerator
NONE	The ZED SDK will not apply any preprocessing to the detected objects.
NMS3D	The ZED SDK will remove objects that are in the same 3D position as an already tracked object (independent of class id).
NMS3D_PER_CLASS	The ZED SDK will remove objects that are in the same 3D position as an already tracked object of the same class id.

Function Documentation

◆ checkAIModelStatus()

AI_Model_status sl::checkAIModelStatus	(	AI_MODELS	model,
		int	gpu_id = `0`
	)

Checks if a corresponding optimized engine is found for the requested Model based on your rig configuration.

Returns: The status of the given model for the specified GPU.

◆ downloadAIModel()

ERROR_CODE sl::downloadAIModel	(	sl::AI_MODELS	model,
		int	gpu_id = `0`
	)

Downloads the requested model.

Returns: sl::ERROR_CODE::SUCCESS if the model is already download.

◆ optimizeAIModel()

ERROR_CODE sl::optimizeAIModel	(	sl::AI_MODELS	model,
		int	gpu_id = `0`
	)

Optimizes the requested model (and download the model if it is not present on the host).

Returns: sl::ERROR_CODE::SUCCESS if the model is well optimized.

◆ generate_unique_id()

String sl::generate_unique_id ( )

Generate a UUID like unique id to help identify and track AI detections.

Classes

Enumerations

Functions

Enumeration Type Documentation

◆ OBJECT_DETECTION_MODEL

◆ AI_MODELS

◆ OBJECT_CLASS

◆ OBJECT_SUBCLASS

◆ OBJECT_TRACKING_STATE

◆ OBJECT_ACTION_STATE

◆ OBJECT_FILTERING_MODE

Function Documentation

◆ checkAIModelStatus()

◆ downloadAIModel()

◆ optimizeAIModel()

◆ generate_unique_id()