ObjectDetectionParameters Class Reference

Class containing a set of parameters for the object detection module. More...

Functions

ObjectDetectionParameters __cinit__ (self, image_sync=True, enable_tracking=True, enable_segmentation=False, detection_model=OBJECT_DETECTION_MODEL.MULTI_CLASS_BOX_FAST, max_range=-1.0, batch_trajectories_parameters=BatchParameters(), filtering_mode=OBJECT_FILTERING_MODE.NMS3D, prediction_timeout_s=0.2, allow_reduced_precision_inference=False, instance_module_id=0)
 Default constructor. More...
 
bool image_sync (self)
 Whether the object detection is synchronized to the image or runs in a separate thread. More...
 
bool enable_tracking (self)
 Whether the object detection system includes object tracking capabilities across a sequence of images. More...
 
bool enable_segmentation (self)
 Whether the object masks will be computed. More...
 
OBJECT_DETECTION_MODEL detection_model (self)
 sl.OBJECT_DETECTION_MODEL to use. More...
 
float max_range (self)
 Upper depth range for detections. More...
 
BatchParameters batch_parameters (self)
 Batching system parameters. More...
 
def filtering_mode (self)
 Filtering mode that should be applied to raw detections. More...
 
float prediction_timeout_s (self)
 Prediction duration of the ZED SDK when an object is not detected anymore before switching its state to sl.OBJECT_TRACKING_STATE.SEARCHING. More...
 
bool allow_reduced_precision_inference (self)
 Whether to allow inference to run at a lower precision to improve runtime and memory usage. More...
 
int instance_module_id (self)
 Id of the module instance. More...
 

Detailed Description

Class containing a set of parameters for the object detection module.

The default constructor sets all parameters to their default settings.

Note
Parameters can be adjusted by the user.

Functions

◆ __cinit__()

ObjectDetectionParameters __cinit__ (   self,
  image_sync = True,
  enable_tracking = True,
  enable_segmentation = False,
  detection_model = OBJECT_DETECTION_MODEL.MULTI_CLASS_BOX_FAST,
  max_range = -1.0,
  batch_trajectories_parameters = BatchParameters(),
  filtering_mode = OBJECT_FILTERING_MODE.NMS3D,
  prediction_timeout_s = 0.2,
  allow_reduced_precision_inference = False,
  instance_module_id = 0 
)

Default constructor.

All the parameters are set to their default values.

Parameters
image_sync: Activates image_sync
enable_tracking: Activates enable_tracking
enable_segmentation: Activates enable_segmentation
detection_model: Chosen detection_model
max_range: Chosen max_range
batch_trajectories_parameters: Chosen batch_parameters
filtering_mode: Chosen filtering_mode
prediction_timeout_s: Chosen prediction_timeout_s
allow_reduced_precision_inference: Activates allow_reduced_precision_inference
instance_module_id: Chosen instance_module_id

◆ image_sync()

bool image_sync (   self)

Whether the object detection is synchronized to the image or runs in a separate thread.

If set to true, the detection is run on every sl.Camera.grab().
Otherwise, the thread runs at its own speed, which can lead to new detection once in a while.
Default: True

◆ enable_tracking()

bool enable_tracking (   self)

Whether the object detection system includes object tracking capabilities across a sequence of images.

Default: True

◆ enable_segmentation()

bool enable_segmentation (   self)

Whether the object masks will be computed.

Default: False

◆ detection_model()

◆ max_range()

float max_range (   self)

Upper depth range for detections.

Default: -1 (value set in sl.InitParameters.depth_maximum_distance)

Note
The value cannot be greater than sl.InitParameters.depth_maximum_distance and its unit is defined in sl.InitParameters.coordinate_units.

◆ batch_parameters()

BatchParameters batch_parameters (   self)

Batching system parameters.

Batching system (introduced in 3.5) performs short-term re-identification with deep-learning and trajectories filtering.
sl.BatchParameters.enable must to be true to use this feature (by default disabled).

◆ filtering_mode()

def filtering_mode (   self)

Filtering mode that should be applied to raw detections.

Default: sl.OBJECT_FILTERING_MODE.NMS_3D (same behavior as previous ZED SDK version)

Note
This parameter is only used in detection model sl.OBJECT_DETECTION_MODEL.MULTI_CLASS_BOX_XXX and sl.OBJECT_DETECTION_MODEL.CUSTOM_BOX_OBJECTS.
For custom object, it is recommended to use sl.OBJECT_FILTERING_MODE.NMS_3D_PER_CLASS or sl.OBJECT_FILTERING_MODE.NONE.
In this case, you might need to add your own NMS filter before ingesting the boxes into the object detection module.

◆ prediction_timeout_s()

float prediction_timeout_s (   self)

Prediction duration of the ZED SDK when an object is not detected anymore before switching its state to sl.OBJECT_TRACKING_STATE.SEARCHING.

It prevents the jittering of the object state when there is a short misdetection.
The user can define their own prediction time duration.
Default: 0.2

Note
During this time, the object will have sl.OBJECT_TRACKING_STATE.OK state even if it is not detected.
The duration is expressed in seconds.
Warning
prediction_timeout_s will be clamped to 1 second as the prediction is getting worse with time.
Setting this parameter to 0 disables the ZED SDK predictions.

◆ allow_reduced_precision_inference()

bool allow_reduced_precision_inference (   self)

Whether to allow inference to run at a lower precision to improve runtime and memory usage.

It might increase the initial optimization time and could include downloading calibration data or calibration cache and slightly reduce the accuracy.

Note
The fp16 is automatically enabled if the GPU is compatible and provides a speed up of almost x2 and reduce memory usage by almost half, no precision loss.
This setting allow int8 precision which can speed up by another x2 factor (compared to fp16, or x4 compared to fp32) and half the fp16 memory usage, however some accuracy could be lost.
The accuracy loss should not exceed 1-2% on the compatible models.
The current compatible models are all sl.AI_MODELS.HUMAN_BODY_XXXX.

◆ instance_module_id()

int instance_module_id (   self)

Id of the module instance.

This is used to identify which object detection module instance is used.