Ghost-free high dynamic range imaging via rank minimization

Chul Lee, Yuelong Li, Vishal Monga

    Research output: Contribution to journalArticle

    43 Citations (Scopus)

    Abstract

    We propose a ghost-free high dynamic range (HDR) image synthesis algorithm using a low-rank matrix completion framework, which we call RM-HDR. Based on the assumption that irradiance maps are linearly related to low dynamic range (LDR) image exposures, we formulate ghost region detection as a rank minimization problem. We incorporate constraints on moving objects, i.e., sparsity, connectivity, and priors on under-and over-exposed regions into the framework. Experiments on real image collections show that the RM-HDR can often provide significant gains in synthesized HDR image quality over state-of-the-art approaches. Additionally, a complexity analysis is performed which reveals computational merits of RM-HDR over recent advances in deghosting for HDR.

    Original languageEnglish (US)
    Article number6814772
    Pages (from-to)1045-1049
    Number of pages5
    JournalIEEE Signal Processing Letters
    Volume21
    Issue number9
    DOIs
    StatePublished - Sep 2014

    Fingerprint

    Range Imaging
    High Dynamic Range
    Imaging techniques
    Range Image
    Matrix Completion
    Low-rank Matrices
    Complexity Analysis
    Irradiance
    Dynamic Range
    Moving Objects
    Sparsity
    Image Quality
    Minimization Problem
    Connectivity
    Image quality
    Linearly
    Synthesis
    Experiment

    All Science Journal Classification (ASJC) codes

    • Signal Processing
    • Electrical and Electronic Engineering
    • Applied Mathematics

    Cite this

    Lee, Chul ; Li, Yuelong ; Monga, Vishal. / Ghost-free high dynamic range imaging via rank minimization. In: IEEE Signal Processing Letters. 2014 ; Vol. 21, No. 9. pp. 1045-1049.
    @article{b5ebfa94b3094c5bbbdd496642d6b81e,
    title = "Ghost-free high dynamic range imaging via rank minimization",
    abstract = "We propose a ghost-free high dynamic range (HDR) image synthesis algorithm using a low-rank matrix completion framework, which we call RM-HDR. Based on the assumption that irradiance maps are linearly related to low dynamic range (LDR) image exposures, we formulate ghost region detection as a rank minimization problem. We incorporate constraints on moving objects, i.e., sparsity, connectivity, and priors on under-and over-exposed regions into the framework. Experiments on real image collections show that the RM-HDR can often provide significant gains in synthesized HDR image quality over state-of-the-art approaches. Additionally, a complexity analysis is performed which reveals computational merits of RM-HDR over recent advances in deghosting for HDR.",
    author = "Chul Lee and Yuelong Li and Vishal Monga",
    year = "2014",
    month = "9",
    doi = "10.1109/LSP.2014.2323404",
    language = "English (US)",
    volume = "21",
    pages = "1045--1049",
    journal = "IEEE Signal Processing Letters",
    issn = "1070-9908",
    publisher = "Institute of Electrical and Electronics Engineers Inc.",
    number = "9",

    }

    Ghost-free high dynamic range imaging via rank minimization. / Lee, Chul; Li, Yuelong; Monga, Vishal.

    In: IEEE Signal Processing Letters, Vol. 21, No. 9, 6814772, 09.2014, p. 1045-1049.

    Research output: Contribution to journalArticle

    TY - JOUR

    T1 - Ghost-free high dynamic range imaging via rank minimization

    AU - Lee, Chul

    AU - Li, Yuelong

    AU - Monga, Vishal

    PY - 2014/9

    Y1 - 2014/9

    N2 - We propose a ghost-free high dynamic range (HDR) image synthesis algorithm using a low-rank matrix completion framework, which we call RM-HDR. Based on the assumption that irradiance maps are linearly related to low dynamic range (LDR) image exposures, we formulate ghost region detection as a rank minimization problem. We incorporate constraints on moving objects, i.e., sparsity, connectivity, and priors on under-and over-exposed regions into the framework. Experiments on real image collections show that the RM-HDR can often provide significant gains in synthesized HDR image quality over state-of-the-art approaches. Additionally, a complexity analysis is performed which reveals computational merits of RM-HDR over recent advances in deghosting for HDR.

    AB - We propose a ghost-free high dynamic range (HDR) image synthesis algorithm using a low-rank matrix completion framework, which we call RM-HDR. Based on the assumption that irradiance maps are linearly related to low dynamic range (LDR) image exposures, we formulate ghost region detection as a rank minimization problem. We incorporate constraints on moving objects, i.e., sparsity, connectivity, and priors on under-and over-exposed regions into the framework. Experiments on real image collections show that the RM-HDR can often provide significant gains in synthesized HDR image quality over state-of-the-art approaches. Additionally, a complexity analysis is performed which reveals computational merits of RM-HDR over recent advances in deghosting for HDR.

    UR - http://www.scopus.com/inward/record.url?scp=84901428516&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84901428516&partnerID=8YFLogxK

    U2 - 10.1109/LSP.2014.2323404

    DO - 10.1109/LSP.2014.2323404

    M3 - Article

    AN - SCOPUS:84901428516

    VL - 21

    SP - 1045

    EP - 1049

    JO - IEEE Signal Processing Letters

    JF - IEEE Signal Processing Letters

    SN - 1070-9908

    IS - 9

    M1 - 6814772

    ER -