Struct VisualSortOptions 

pub struct VisualSortOptions { /* private fields */ }

Class that is used to configure the Visual Tracker

Implementations

impl VisualSortOptions

pub fn max_idle_epochs(self, n: usize) -> Self

The number of epochs the track remains active.

Lets the Frame Rate per second is 30, setting max_idle_epochs to 30 means that the track in store will be active even if only one new observation was merged with it during the second. If during 30 invocations of predict for the scene, where the track is defined, no observations are merged with it, the track will be marked as wasted, and no further observations will be merged with it.

pub fn kept_history_length(self, n: usize) -> Self

The number of last observations, predictions, and features kept within the track attributes.

The track’s attributes may accumulate last observations, predictions, and features for the caller’s purpose. To protect the system from overflow kept_history_length parameter is used. It forces the track to keep only the last N values instead of unlimited history. The parameter is important when one uses the tracker in offline mode - when the wasted tracks are used to get the history. If the tracker is an online tracker, setting 1 is a reasonable value to keep memory utilization low.

pub fn visual_metric(self, metric: VisualSortMetricType) -> Self

The method is used to calculate the distance for visual_sort feature vectors.

Currently, cosine and euclidean metrics are supported. The one you choose is defined by the ReID model used.

pub fn visual_min_votes(self, n: usize) -> Self

The minimal number of votes that is required to allow a track candidate to surpass the enabling threshold of the visual_sort voting. The maximum allowed number of visual_sort features kept for the track is defined by visual_max_observations.

Don’t confuse visual_max_observations with kept_history_length - they have no relation. The later is only used for caller purposes, not for track prediction.

When the track candidate consisting of the single observation is compared versus tracks kept in the store the system calculates up to N distances (1 X N), where N at most is equal to visual_max_observations, but can be less if the track is short or previous observations were ignored due to quality or other constraints.

Only when N >= visual_min_votes, the track candidate is used in leader selection.

pub fn visual_max_observations(self, n: usize) -> Self

The maximum number of visual_sort observations kept in the track for visual_sort estimations. The features are collected in the track from the candidates, and when the visual_max_observations is reached, the features with lower quality are wiped from the track.

pub fn positional_min_confidence(self, conf: f32) -> Self

Minimal allowed confidence for bounding boxes. If the confidence is less than specified it is corrected to be the minimal

pub fn spatio_temporal_constraints( self, constraints: SpatioTemporalConstraints, ) -> Self

The constraints define how far the candidate is allowed to be from a track’s last box to participate in the selection for the track. If the track candidate is too far from the track kept in the store, it is skipped from the comparison.

pub fn positional_metric(self, metric: PositionalMetricType) -> Self

The parameter defines which positional metric is used to calculate distances between the track candidate and tracks kept in the store. There are two metrics are supported - the Mahalanobis metric and the IoU metric.

pub fn visual_minimal_track_length(self, length: usize) -> Self

The minimally required number of visual_sort features in the track that enables their usage in candidates estimation. If the track is short and there are fewer features collected than visual_minimal_track_length then candidates are estimated against it only by positional distance. Keep in mind that this parameter must be less than or equal to visual_max_observations to have sense.

pub fn visual_minimal_area(self, area: f32) -> Self

The minimal required area of track candidate’s bounding box to use the visual_sort feature in estimation. This parameter protects from the low-quality features received from the smallish boxes.

pub fn visual_minimal_quality_use(self, q: f32) -> Self

The visual_sort quality threshold of a feature that activates the visual_sort estimation of a candidate versus the tracks kept in the store.

pub fn visual_minimal_quality_collect(self, q: f32) -> Self

The visual_sort quality threshold of a feature that activates the adding of the visual_sort feature to the track’s visual_sort features.

pub fn visual_minimal_own_area_percentage_use(self, area: f32) -> Self

The threshold is calculated as solely_owned_area / all_area of the bounding box that prevents low-quality visual_sort features received in a messy environment from being used in visual_sort predictions.

pub fn visual_minimal_own_area_percentage_collect(self, area: f32) -> Self

The threshold is calculated as solely_owned_area / all_area of the bounding box that prevents low-quality visual_sort features received in a messy environment from being collected to a track for making visual_sort predictions.

pub fn kalman_position_weight(self, weight: f32) -> Self

pub fn kalman_velocity_weight(self, weight: f32) -> Self

Trait Implementations

impl Clone for VisualSortOptions

fn clone(&self) -> VisualSortOptions

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for VisualSortOptions

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for VisualSortOptions

fn default() -> Self

Returns the “default value” for a type.