# Lecture-9 - Alper Yilmaz Target Tracking Using Mean Shift...

This preview shows pages 1–3. Sign up to view the full content.

Alper Yilmaz UCF Computer Vision Lab. 1 Lecture-9 UCF Computer Vision Lab. 2 Target Tracking Using Mean Shift Alper YILMAZ Computer Vision Lab. University of Central Florida UCF Computer Vision Lab. 3 Problem ! Given a target (object) in the first frame track it through all other frames. UCF Computer Vision Lab. 4 Key Concepts ! Mean ! Mean shift ! Probability Density Functions (PDFs) ! Kernel Density ! Gradient of Kernel Density ! Bhatacharya coefficient UCF Computer Vision Lab. 5 Mean Shift Vector Given: Data points and approximate location of the mean of this data. Task: Estimate the exact location of the mean of the data by determining the shift vector from initial mean. UCF Computer Vision Lab. 6 Mean Shift Vector Example ! " # \$ % ( ) ( x n i i x h n M 1 0 ) ( 1 ) ( y x y Mean shift vector always points towards the direction of the maximum increase in the density . n x : number of points y 0 : initial mean location x i : data points

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Alper Yilmaz UCF Computer Vision Lab. 2 UCF Computer Vision Lab. 7 Modified Mean Shift (weighted) 0 0 1 1 0 0 ) ( ) ( ) ( y y x y y ! ! " # \$ \$ % ( ) ) ( ( x x n i i n i i i h w w M n x : number of points in the kernel y 0 : initial mean location x i : data points h : kernel radius Weights are determined using kernels (masks): Uniform, Gaussian or Epanechnikov UCF Computer Vision Lab. 8 Properties of Mean Shift • Mean shift vector has the direction of the gradient of the density estimate . • It is computed iteratively for obtaining the maximum
This is the end of the preview. Sign up to access the rest of the document.

## This note was uploaded on 10/04/2011 for the course CAP 6411 taught by Professor Shah during the Spring '09 term at University of Central Florida.

### Page1 / 5

Lecture-9 - Alper Yilmaz Target Tracking Using Mean Shift...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online