SE 263 R. Venkatesh Babu. Object Tracking. R. Venkatesh Babu

Similar documents
Robert Collins CSE598G. Intro to Template Matching and the Lucas-Kanade Method

Recall: Derivative of Gaussian Filter. Lecture 7: Correspondence Matching. Observe and Generalize. Observe and Generalize. Observe and Generalize

Autonomous Navigation for Flying Robots

Dense Image-based Motion Estimation Algorithms & Optical Flow

Visual Tracking (1) Tracking of Feature Points and Planar Rigid Objects

Visual Tracking (1) Feature Point Tracking and Block Matching

Feature Tracking and Optical Flow

Visual Tracking. Image Processing Laboratory Dipartimento di Matematica e Informatica Università degli studi di Catania.

Automatic Image Alignment (feature-based)

COMPUTER VISION > OPTICAL FLOW UTRECHT UNIVERSITY RONALD POPPE

Feature Tracking and Optical Flow

Image processing and features

Tracking in image sequences

Visual Tracking (1) Pixel-intensity-based methods

Visual Tracking. Antonino Furnari. Image Processing Lab Dipartimento di Matematica e Informatica Università degli Studi di Catania

BSB663 Image Processing Pinar Duygulu. Slides are adapted from Selim Aksoy

Augmented Reality VU. Computer Vision 3D Registration (2) Prof. Vincent Lepetit

CS201: Computer Vision Introduction to Tracking

Harder case. Image matching. Even harder case. Harder still? by Diva Sian. by swashford

Face Tracking : An implementation of the Kanade-Lucas-Tomasi Tracking algorithm

Peripheral drift illusion

Leow Wee Kheng CS4243 Computer Vision and Pattern Recognition. Motion Tracking. CS4243 Motion Tracking 1

Optical flow and tracking

Harder case. Image matching. Even harder case. Harder still? by Diva Sian. by swashford

Outline 7/2/201011/6/

EE795: Computer Vision and Intelligent Systems

Outline. Data Association Scenarios. Data Association Scenarios. Data Association Scenarios

Local Feature Detectors

Computer Vision Lecture 20

Lecture 16: Computer Vision

Lecture 16: Computer Vision

Computer Vision Lecture 20

Image matching. Announcements. Harder case. Even harder case. Project 1 Out today Help session at the end of class. by Diva Sian.

Chapter 3 Image Registration. Chapter 3 Image Registration

Learning to Detect Faces. A Large-Scale Application of Machine Learning

Computer Vision: Homework 3 Lucas-Kanade Tracking & Background Subtraction

CS 4495 Computer Vision Motion and Optic Flow

Edge and corner detection

Matching. Compare region of image to region of image. Today, simplest kind of matching. Intensities similar.

Computer Vision II Lecture 4

Tracking Computer Vision Spring 2018, Lecture 24

CS 378: Autonomous Intelligent Robotics. Instructor: Jivko Sinapov

Motion and Optical Flow. Slides from Ce Liu, Steve Seitz, Larry Zitnick, Ali Farhadi

CS6670: Computer Vision

SIFT: SCALE INVARIANT FEATURE TRANSFORM SURF: SPEEDED UP ROBUST FEATURES BASHAR ALSADIK EOS DEPT. TOPMAP M13 3D GEOINFORMATION FROM IMAGES 2014

Eppur si muove ( And yet it moves )

Image Features. Work on project 1. All is Vanity, by C. Allan Gilbert,

Multi-stable Perception. Necker Cube

Local features: detection and description. Local invariant features

Automatic Image Alignment

Wikipedia - Mysid

CS5670: Computer Vision

Deep Learning: Image Registration. Steven Chen and Ty Nguyen

SUMMARY: DISTINCTIVE IMAGE FEATURES FROM SCALE- INVARIANT KEYPOINTS

16720 Computer Vision: Homework 3 Template Tracking and Layered Motion.

CS231A Section 6: Problem Set 3

Automatic Image Alignment

Motion. 1 Introduction. 2 Optical Flow. Sohaib A Khan. 2.1 Brightness Constancy Equation

Displacement estimation

Automatic Image Alignment (direct) with a lot of slides stolen from Steve Seitz and Rick Szeliski

EE795: Computer Vision and Intelligent Systems

Lecture 28 Intro to Tracking

Recall: Blob Merge/Split Lecture 28

Motion Estimation and Optical Flow Tracking

Announcements. Computer Vision I. Motion Field Equation. Revisiting the small motion assumption. Visual Tracking. CSE252A Lecture 19.

Chapter 9 Object Tracking an Overview

Computer Vision for HCI. Topics of This Lecture

Capturing, Modeling, Rendering 3D Structures

COMPUTATIONAL INTELLIGENCE

Prof. Feng Liu. Spring /26/2017

OPPA European Social Fund Prague & EU: We invest in your future.

Direct Methods in Visual Odometry

Global Flow Estimation. Lecture 9

Object Recognition with Invariant Features

Local features: detection and description May 12 th, 2015

Computer Vision II Lecture 4

Automatic Initialization of the TLD Object Tracker: Milestone Update

Midterm Wed. Local features: detection and description. Today. Last time. Local features: main components. Goal: interest operator repeatability

Visual motion. Many slides adapted from S. Seitz, R. Szeliski, M. Pollefeys

The Lucas & Kanade Algorithm

Segmentation and Tracking of Partial Planar Templates

Object Tracking Using Local Binary Descriptors

C18 Computer vision. C18 Computer Vision. This time... Introduction. Outline.

Tracking. Establish where an object is, other aspects of state, using time sequence Biggest problem -- Data Association

Feature descriptors. Alain Pagani Prof. Didier Stricker. Computer Vision: Object and People Tracking

CAP 5415 Computer Vision Fall 2012

Introduction to Medical Imaging (5XSA0) Module 5

Computer Vision I - Filtering and Feature detection

Modern Medical Image Analysis 8DC00 Exam

Overview. Video. Overview 4/7/2008. Optical flow. Why estimate motion? Motion estimation: Optical flow. Motion Magnification Colorization.

Mosaics. Today s Readings

Corner Detection. Harvey Rhody Chester F. Carlson Center for Imaging Science Rochester Institute of Technology

Assignment: Backgrounding and Optical Flow.

Computer Vision. Exercise 3 Panorama Stitching 09/12/2013. Compute Vision : Exercise 3 Panorama Stitching

CS4670: Computer Vision

Review for the Final

Gradient-Based Differential Approach for Patient Motion Compensation in 2D/3D Overlay

ROBUST OBJECT TRACKING BY SIMULTANEOUS GENERATION OF AN OBJECT MODEL

IN computer vision develop mathematical techniques in

Image Processing: Final Exam November 10, :30 10:30

Transcription:

Object Tracking R. Venkatesh Babu

Primitive tracking Appearance based - Template Matching Assumptions: Object description derived from first frame No change in object appearance Movement only 2D translation Restrictive approach Not suitable for real-world videos

Basic Template Matching Assumptions: a snapshot of object from first frame can be used to describe appearance Object will look nearly identical in new image Movement is nearly pure 2D translation The last two are very restrictive.

Template Matching Is a search problem: Given an intensity patch element in the left (t) image, search for the corresponding patch in the right image (t+1). We will typically need geometric constraints to reduce the size of the search space

Matching Functions (SAD) (SSD) (Correlation)

Example (Correlation) SE 263 R. Venkatesh Babu

C = SE 263 R. Venkatesh Babu

Problem with Correlation of Raw Image Templates > Solution: Subtract off the mean value of the template. Now, the correlation score is higher only when darker parts of the template overlap darker parts of the image, and brighter parts of the template overlap brighter parts of the image.

SSD or block matching (Sum of Squared Differences) Best match (highest score) in image coincides with correct match in this case!

Practical Issues Object shape might not be well described by a scan line-oriented bounding rectangle End up including lots of background pixels Solution : Gaussian Weighting? Segmentation? Search Range Need some estimate of object motion. Appearance change Solution: Adaptive template update Problem : Drift

Gradient Descent Tracking Need a more efficient method than explicit search over some large window If we have a good estimate of object position already, we can efficiently refine it using gradient descent. Assumption: Our estimate of position must be very close to where the object actually is! (however, we can relax this using multi-scale techniques image pyramids)

Harris Detector : Mathematics Consider the SSD Function This tells how match score changes if you shift the template by [u,v]

Gradient Descent Method Taylor Series for 2D funcions

Lucas-Kanade Derivation SE 263 R. Venkatesh Babu

Lucas-Kanade Tracking Traditional Lucas-Kanade is typically run on small, corner-like features (e.g. 5x5) to compute optic flow. Limitations : Assumption of constant flow (pure translation) for all pixels in a larger window is unreasonable. can easily generalize Lucas-Kanade approach to other 2D parametric motion models (like affine or projective)

JY Bouguet, Pyramidal Implementation of the Lucas Kanade Feature Tracker

Robust Object Tracking With RBF Networks Machine Learning-based object modeling Fast learning RBFN Classifier Use two classifiers (object and background) for reliable modeling Use of posterior probability measure for tracking R. Venkatesh Babu, S. Suresh and A. Makur, Robust Visual Tracking with Online Adaptive RBF-Networks CVIU, March 2010.

Object Modeling Training Phase Object Initialization Object Classifier Object/ Background Separation Feature Extraction Object Model Non-Object Classifier

Object Localization - Testing Phase Object Classifier Current Frame Feature Extraction Object Localization Object Position Non-Object Classifier Object Model

Foreground/Background separation Histogram of pixels in object window and background window L x i log max max h h o b x x i i,, Threshold L(x i ) to classify each pixel as foreground / background Robustness to clutter

Classifier - Architecture SE 263 R. Venkatesh Babu

Fast Learning RBFN Classifier The output of RBFN with K neurons: In matrix form

Computing Output Weights The output weights are estimated as: Where Ф K is Hidden Layer Output Matrix, (pseudo-inverse)

Classifier Development Select K (No. of hidden neurons) Assign (μ,σ) for the Gaussian neurons arbitrarily Compute output weights ( ) Select the object/background classifiers from a set of classifiers.

Object Modeling Posterior probability obtained using object Classifier: Posterior probability obtained using non-object Classifier: The final object model is obtained as:

Object Model Object Classifier (p o ) Non-Object Classifier (1-p b ) Object Model (p t )

Object Localization Object center is estimated by iteratively seeking the mode of posterior probability weighted by target model.

Online Adaptation For each new sample (U n,t n ) Compute Adapt output weight ( ) using RLS Where,

Online Adaptation Only few samples are used for model update

Results Proposed System Solid Yellow Meanshift Dashed Blue

Results With adaptation Solid Yellow Without adaptation Dashed Red Meanshift Dashed Blue

Results Proposed System Solid Yellow Meanshift Dashed Blue