ObjectsBatch Class Reference

Contains batched datas of a detected object. More...

Attributes

int id
 the trajectory id More...
 
OBJECT_CLASS label
 Object category. Identify the object type. More...
 
OBJECT_SUBCLASS sublabel
 Object subclass. More...
 
OBJECT_TRACKING_STATE tracking_state
 Defines the object tracking state. More...
 
std::vector< sl::float3positions
 a sample of 3d position More...
 
std::vector< std::array< float, 6 > > position_covariances
 a sample of the associated position covariance More...
 
std::vector< sl::float3velocities
 a sample of 3d velocity More...
 
std::vector< sl::Timestamptimestamps
 the associated position timestamp More...
 
std::vector< std::vector< sl::float3 > > bounding_boxes
 a sample of 3d bounding boxes More...
 
std::vector< std::vector< sl::uint2 > > bounding_boxes_2d
 2D bounding box of the person represented as four 2D points starting at the top left corner and rotation clockwise. Expressed in pixels on the original image resolution, [0,0] is the top left corner. More...
 
std::vector< float > confidences
 a sample of object detection confidence More...
 
std::vector< OBJECT_ACTION_STATEaction_states
 a sample of the object action state More...
 
std::vector< std::vector< sl::float2 > > keypoints_2d
 a sample of 2d person keypoints. More...
 
std::vector< std::vector< sl::float3 > > keypoints
 a sample of 3d person keypoints More...
 
std::vector< std::vector< sl::uint2 > > head_bounding_boxes_2d
 bounds the head with four 2D points. Expressed in pixels on the original image resolution. More...
 
std::vector< std::vector< sl::float3 > > head_bounding_boxes
 bounds the head with eight 3D points. Defined in sl:InitParameters::UNIT, expressed in RuntimeParameters::measure3D_reference_frame. More...
 
std::vector< sl::float3head_positions
 3D head centroid. Defined in sl:InitParameters::UNIT, expressed in RuntimeParameters::measure3D_reference_frame. More...
 
std::vector< std::vector< float > > keypoint_confidences
 Per keypoint detection confidence, can not be lower than the ObjectDetectionRuntimeParameters::detection_confidence_threshold. More...
 

Detailed Description

Contains batched datas of a detected object.

Variables

◆ id

int id

the trajectory id

◆ label

OBJECT_CLASS label

Object category. Identify the object type.

◆ sublabel

OBJECT_SUBCLASS sublabel

Object subclass.

◆ tracking_state

OBJECT_TRACKING_STATE tracking_state

Defines the object tracking state.

◆ positions

std::vector<sl::float3> positions

a sample of 3d position

◆ position_covariances

std::vector<std::array<float, 6 > > position_covariances

a sample of the associated position covariance

◆ velocities

std::vector<sl::float3> velocities

a sample of 3d velocity

◆ timestamps

std::vector<sl::Timestamp> timestamps

the associated position timestamp

◆ bounding_boxes

std::vector<std::vector<sl::float3> > bounding_boxes

a sample of 3d bounding boxes

◆ bounding_boxes_2d

std::vector<std::vector<sl::uint2> > bounding_boxes_2d

2D bounding box of the person represented as four 2D points starting at the top left corner and rotation clockwise. Expressed in pixels on the original image resolution, [0,0] is the top left corner.

A ------ B
| Object |
D ------ C

◆ confidences

std::vector<float> confidences

a sample of object detection confidence

◆ action_states

std::vector<OBJECT_ACTION_STATE> action_states

a sample of the object action state

◆ keypoints_2d

std::vector<std::vector<sl::float2> > keypoints_2d

a sample of 2d person keypoints.

Note
Not available with DETECTION_MODEL::MULTI_CLASS_BOX.
Warning
in some cases, eg. body partially out of the image or missing depth data, some keypoint can not be detected, they will have non finite values.

◆ keypoints

std::vector<std::vector<sl::float3> > keypoints

a sample of 3d person keypoints

Note
Not available with DETECTION_MODEL::MULTI_CLASS_BOX.
Warning
in some cases, eg. body partially out of the image or missing depth data, some keypoint can not be detected, they will have non finite values.

◆ head_bounding_boxes_2d

std::vector<std::vector<sl::uint2> > head_bounding_boxes_2d

bounds the head with four 2D points. Expressed in pixels on the original image resolution.

Note
Not available with DETECTION_MODEL::MULTI_CLASS_BOX.

◆ head_bounding_boxes

std::vector<std::vector<sl::float3> > head_bounding_boxes

bounds the head with eight 3D points. Defined in sl:InitParameters::UNIT, expressed in RuntimeParameters::measure3D_reference_frame.

Note
Not available with DETECTION_MODEL::MULTI_CLASS_BOX.

◆ head_positions

std::vector<sl::float3> head_positions

3D head centroid. Defined in sl:InitParameters::UNIT, expressed in RuntimeParameters::measure3D_reference_frame.

Note
Not available with DETECTION_MODEL::MULTI_CLASS_BOX.

◆ keypoint_confidences

std::vector<std::vector<float> > keypoint_confidences

Per keypoint detection confidence, can not be lower than the ObjectDetectionRuntimeParameters::detection_confidence_threshold.

Note
Not available with DETECTION_MODEL::MULTI_CLASS_BOX.
Warning
in some cases, eg. body partially out of the image or missing depth data, some keypoint can not be detected, they will have non finite values.