Crate norfair_rs 

Norfair - Object Tracking Library

Rust port of the Python norfair library.

Norfair is a customizable lightweight library for real-time multi-object tracking.

Features

• Kalman filter-based tracking with multiple filter implementations
• Pluggable distance functions (Euclidean, IoU, custom)
• Camera motion compensation
• Re-identification (ReID) support
• MOTChallenge metrics evaluation

Example

use norfair_rs::{Tracker, TrackerConfig, Detection, distance_by_name};

// Create tracker
let config = TrackerConfig::new(distance_by_name("euclidean"), 50.0);
let mut tracker = Tracker::new(config).unwrap();

// Process detections
let detections = vec![Detection::new(vec![[100.0, 100.0]])];
let tracked_objects = tracker.update(detections, 1, None);

Re-exports

pub use camera_motion::CoordinateTransformation;

pub use detection::Detection;

pub use distances::distance_by_name;

pub use distances::Distance;

pub use filter::Filter;

pub use filter::FilterFactory;

pub use tracked_object::TrackedObject;

pub use tracked_object::TrackedObjectFactory;

pub use tracker::Tracker;

pub use tracker::TrackerConfig;

Modules

camera_motion

Camera motion compensation module.

detection

Detection struct for input to the tracker.

distances

Distance functions for matching detections to tracked objects.

filter

Kalman filter implementations for object tracking.

matching

Detection-to-object matching algorithms.

metrics

MOTChallenge metrics evaluation module.

tracked_object

TrackedObject struct for tracked objects maintained by the tracker.

tracker

Main tracker implementation.

utils

Utility functions for norfair.

Enums

Error

Errors that can occur in the norfair library

Type Aliases

Result

Result type for norfair operations