Tracking Module Reference

Provides object tracking and background subtraction algorithms.

Tracker

pub enum TrackerType {
    KCF,
    CSRT,
    MOSSE,
}

pub struct Tracker<B: Backend> {
    pub tracker_type: TrackerType,
    pub bbox: Option<Rect<usize>>,
}

impl<B: Backend> Tracker<B> {
    pub fn new(tracker_type: TrackerType) -> Self;
    pub fn init(&mut self, image: &Image<B>, bbox: Rect<usize>) -> Result<()>;
    pub fn update(&mut self, image: &Image<B>) -> Result<Rect<usize>>;
}

BackgroundSubtractor

pub struct BackgroundSubtractor<B: Backend> {
    learning_rate: f32,
    threshold: f32,
    background: Option<Image<B>>,
}

impl<B: Backend> BackgroundSubtractor<B> {
    pub fn new(learning_rate: f32, threshold: f32) -> Self;
    pub fn apply(&mut self, frame: &Image<B>) -> Result<Image<B>>;
}

Example

use iris::prelude::*;
use burn::backend::wgpu::Wgpu;

type Backend = Wgpu;
let device = Default::default();

// Background subtraction
let mut subtractor = BackgroundSubtractor::new(0.1, 0.05);
let mask = subtractor.apply(&frame)?;

// Object tracking
let mut tracker = Tracker::new(TrackerType::MOSSE);
let init_bbox = Rect::new(50, 50, 100, 100);
tracker.init(&frame1, init_bbox)?;

// Track across frames
let bbox = tracker.update(&frame2)?;
println!("Object at: {:?}", bbox);

MeanShift Tracker

Iterative kernel-based tracker that locates the object by shifting the search window toward the peak of the color histogram probability distribution.

pub struct MeanShiftTracker {
    // internal state
}

impl MeanShiftTracker {
    pub fn new() -> Self;
    pub fn init<B: Backend>(&mut self, image: &Image<B>, roi: Rect<usize>) -> Result<()>;
    pub fn update<B: Backend>(&mut self, image: &Image<B>) -> Result<Rect<usize>>;
}

Methods

Method	Description
new()	Creates a new MeanShift tracker with default parameters.
init(image, roi)	Initializes the tracker with the target region of interest (histogram model).
update(image)	Performs one MeanShift iteration and returns the updated bounding box.

Example

use iris::prelude::*;

let device = WgpuDevice::default();
let frame1 = Image::<Wgpu>::open("frame1.jpg", &device)?;
let frame2 = Image::<Wgpu>::open("frame2.jpg", &device)?;

let mut tracker = MeanShiftTracker::new();
let roi = Rect::new(120, 80, 64, 64);
tracker.init(&frame1, roi)?;

let bbox = tracker.update(&frame2)?;
println!("Tracked object at: {:?}", bbox);