Skip to main content ITU
Logo
  • Programmes
    • BSc Programmes
    • BSc in Global Business Informatics
    • BSc in Digital Design and Interactive Technologies
    • BSc in Software Development
    • BSc in Data Science
    • Applying for a BSc programme
    • MSc Programmes
    • MSc in Digital Innovation & Management
    • MSc in Digital Design and Interactive Technologies
    • MSc in Software Design
    • MSc in Data Science
    • MSc in Computer Science
    • MSc in Games
    • Applying for an MSc programme
    • Student Life
    • Practical information for international students
    • Ask a student
    • Women in tech
    • Student organisations at ITU
    • Study start
    • Labs for students
    • Special Educational Support (SPS)
    • Study and Career Guidance
    • Exchange student
    • Become an exchange student
    • Guest Students
    • Who can be a guest student?
    • ITU Summer University
    • Open House
    • Open House - BSc programmes
    • Open House - MSc programmes
  • Professional Education
    • Master in IT Management
    • Master in IT Management
    • Admission and entry requirements
    • Contact
    • Single Subjects
    • About single subjects
    • Admission and entry requirements
    • Contact
    • Short courses | ITU Professional Courses
    • See all short courses
    • Contact
    • Contact
    • Contact us here
  • Research
    • Sections
    • Data Science
    • Data, Systems, and Robotics
    • Digital Business Innovation
    • Digitalization Democracy and Governance
    • Human-Computer Interaction and Design
    • Play Culture and AI
    • Software Engineering
    • Technologies in Practice
    • Theoretical Computer Science
    • Research Centres
    • Centre for Digital Play
    • Center for Climate IT
    • Center for Computing Education Research
    • Centre for Digital Welfare
    • Centre for Information Security and Trust
    • Research Centre for Government IT
    • Danish Institute for IT Program Management
    • Research entities
    • Research centers
    • Sections
    • Research groups
    • Labs
    • ITU Research Portal
    • Find Researcher
    • Find Research
    • Research Ethics and Integrity
    • Good Scientific Practice
    • Technical Reports
    • Technical Reports
    • PhD Programme
    • About the PhD Programme
    • PhD Courses
    • PhD Defences
    • PhD Positions
    • Types of Enrolment
    • PhD Admission Requirements
    • PhD Handbook
    • PhD Support
  • Collaboration
    • Collaboration with students
    • Project collaboration
    • Project Market
    • Student worker
    • Project postings
    • Job and Project bank
    • Employer Branding
    • IT Match Making
    • Hiring an ITU student or graduate
    • Make a post in the job bank
    • Research collaboration
    • Read more about research collaboration at ITU
    • Industrial PhD
    • Hire an Industrial PhD
    • Maritime Hub
    • Innovation and entrepreneurship
    • ITU Business Development
    • ITU NextGen
  • About ITU
    • About ITU
    • Press
    • Vacancies
    • Contact
  • DK
ITU
ITU  /  Research  /  Technical Reports  /  Technical Reports Archive  /  2005  /  Semi-Automatic Foreground Extraction For Natural Images
  • Research
    • Research Sections
    • Research Ethics and Integrity
    • Good Scientific Practice
    • Research centers
    • Research groups
    • Labs
    • Technical Reports
      • Technical Reports Archive
        • 2024
        • 2023
        • 2021
        • 2018
        • 2017
        • 2016
        • 2015
        • 2014
        • 2013
        • 2012
        • 2011
        • 2010
        • 2009
        • 2008
        • 2007
        • 2006
        • 2005
          • Scalable Computation of Acyclic Joins
          • Bigraphical Models of Context-aware Systems
          • Pre-Symmetry Set Based Shape Matching
          • Axiomatizing Binding Bigraphs (revised)
          • Bigraphical Semantics of Higher-Order Mobile Embedded Resources with Local Names
          • BI Hyperdoctrines, Higher-Order Separation Logic, and Abstraction
          • Interactive Reconfiguration in Power Supply Restoration
          • Interactive Configuration Based on Linear Programming
          • Asymmetric k-Center with Minimum Coverage
          • Matching 2D Shapes Using Their Symmetry Sets
          • Semi-Automatic Foreground Extraction For Natural Images
            • Axiomatizing Binding Bigraphs
            • Distributed Reactive XML: an XML-centric coordination middleware
            • Bigraphs by Example
            • Parametric Completion for Models of Polymorphic Linear / Intuitionistic Lambda Calculus
            • Synthetic Domain Theory and Models of Linear Abadi & Plotkin Logic
            • Categorical Models of PILL
            • Parametric Domain-theoretic models of Linear Abadi & Plotkin Logic
            • Bigraphs and (Reactive) XML - an XML-centric model of computation
            • Probabilistic models for concurrency - Notes for a minicourse
            • The Tree Inclusion Problem: In Optimal Space and Faster
          • 2004
          • 2003
          • 2002
          • 2001
          • 2000
      • PhD Programme

    Semi-Automatic Foreground Extraction For Natural Images

    TR-2005-64, Authors: Andreas Rishede Hyllested and Martin Wallengren Nilsson

     

    Semi-Automatic Foreground Extraction For Natural Images

    Andreas Rishede Hyllested

    Martin Wallengren Nilsson

    March 2005

    Abstract

    The work described in this thesis was carried out at the IT University of Copenhagen, Denmark from September 2004 to March 2005. The focus of the work has been defined together with the collaboration partner Laerdal-Sophus A/S who seeks to replace a current manual segmentation method with a semi-automatic segmentation method.

    The thesis produces a method for and describes the framework of interactive segmentation of foreground objects in natural images. The basis technology is segmentation by graph cut. To increase interactivity speed the graph construction is preceded by a toboggan watershed segmentation of the input image. The watershed segmentation is implemented in a multi-scale framework where different normalisation methods are tested. Based on the watershed segmented image is built a graph, in which every watershed segment corresponds to one node. The minimum-cut is efficiently computed using the new augmenting path based Boykov-Kolmogorov algorithm. To increase performance of the algorithm an initial pre-augmentation of all terminal links is proposed.

    For fine tuning of the segmentation we propose a new method that uses a local graph representation to perform a pixel-based minimum-cut in specified areas. The method has automatic input of seed points. For difficult parts of the object boundary the locally found minimum-cut improves the result by overriding the global segmentation.

    Two new methods to enter seed points for the min-cut/max-flow algorithm are demonstrated. The first method is based on finding shortest paths through elongated image structures and converting the found paths into dense rows of seed points. The other method automatically enters background seed points by subtraction of a background image.

    The advanced Bayesian framework for alpha matting proposed by Chuang et al. is tested. It shows to have the potential for producing good alpha mattes even for difficult segments, but it occurs to be too slow to match Laerdal-Sophus A/S’s need for interactivity. Instead, by applying Gaussian low-pass filtering to the alpha channel we obtain decent alpha transitions for simple borders in a few seconds. This simple approach can be allowed, at least, in the specific case Laerdal-Sophus A/S, because their image segments usually have simple borders.

    All implementation is done in C++ to integrate with Picture Factory, which is an image editing application currently used by Laerdal-Sophus A/S.


    Technical report TR-2005-64 in IT University Technical Report Series, March 2005.

    Available as PDF.


    Contact us

    Phone
    +45 7218 5000
    E-mail
    itu@itu.dk

    All contact information

    Web Accessibility Statement

    Find us

    IT University of Copenhagen
    Rued Langgaards Vej 7
    DK-2300 Copenhagen S
    Denmark
    How to get here

    Follow us

    ITU Student /
    Privacy /
    EAN-nr. 5798000417878/
    CVR-nr. 29 05 77 53 /
    P-nummer 1005162959

    This page is printed from https://en.itu.dk/

    Fejl i tilmelding