Linear discrimination dictionary learning for shape descriptors

Meng Wang, Jin Xie, Fan Zhu, Yi Fang

    Research output: Contribution to journalArticle

    Abstract

    The complexity and variation of 3D models have posed a lot of challenges in 3D shape retrieval area, for example, the invariant representation and retrieval of nonrigid and noisy 3D shapes. This paper proposed a supervised dictionary learning scheme called Linear Discrimination Dictionary Learning (LDDL) which can learn shape representations that are insensitive to 3D shape deformations in the same category and different for shapes from different categories in the meantime. Besides, it can extract the subtle differences between 3D shapes for fine-grained shapes. To be specific, in this paper, category-specific dictionaries are learnt to encode subtle visual differences of shapes among different categories, a shared dictionary is learnt to encode common patterns of shapes among all the categories; with the Linear Discriminant Analysis (LDA) constraint on the learnt descriptors, the new descriptors can have small within-class scatter and big between-class scatter. Our method is efficient in training and can obtain promising shape retrieval performance on representative shape benchmark datasets.

    Original languageEnglish (US)
    Pages (from-to)349-356
    Number of pages8
    JournalPattern Recognition Letters
    Volume83
    DOIs
    StatePublished - Nov 1 2016

    Fingerprint

    Glossaries
    Discriminant analysis

    Keywords

    • Dictionary learning
    • Linear discriminant analysis
    • Shape retrieval

    ASJC Scopus subject areas

    • Software
    • Signal Processing
    • Computer Vision and Pattern Recognition
    • Artificial Intelligence

    Cite this

    Linear discrimination dictionary learning for shape descriptors. / Wang, Meng; Xie, Jin; Zhu, Fan; Fang, Yi.

    In: Pattern Recognition Letters, Vol. 83, 01.11.2016, p. 349-356.

    Research output: Contribution to journalArticle

    Wang, Meng ; Xie, Jin ; Zhu, Fan ; Fang, Yi. / Linear discrimination dictionary learning for shape descriptors. In: Pattern Recognition Letters. 2016 ; Vol. 83. pp. 349-356.
    @article{80ede510f5674028a7a1ce6cc46e277d,
    title = "Linear discrimination dictionary learning for shape descriptors",
    abstract = "The complexity and variation of 3D models have posed a lot of challenges in 3D shape retrieval area, for example, the invariant representation and retrieval of nonrigid and noisy 3D shapes. This paper proposed a supervised dictionary learning scheme called Linear Discrimination Dictionary Learning (LDDL) which can learn shape representations that are insensitive to 3D shape deformations in the same category and different for shapes from different categories in the meantime. Besides, it can extract the subtle differences between 3D shapes for fine-grained shapes. To be specific, in this paper, category-specific dictionaries are learnt to encode subtle visual differences of shapes among different categories, a shared dictionary is learnt to encode common patterns of shapes among all the categories; with the Linear Discriminant Analysis (LDA) constraint on the learnt descriptors, the new descriptors can have small within-class scatter and big between-class scatter. Our method is efficient in training and can obtain promising shape retrieval performance on representative shape benchmark datasets.",
    keywords = "Dictionary learning, Linear discriminant analysis, Shape retrieval",
    author = "Meng Wang and Jin Xie and Fan Zhu and Yi Fang",
    year = "2016",
    month = "11",
    day = "1",
    doi = "10.1016/j.patrec.2016.05.028",
    language = "English (US)",
    volume = "83",
    pages = "349--356",
    journal = "Pattern Recognition Letters",
    issn = "0167-8655",
    publisher = "Elsevier",

    }

    TY - JOUR

    T1 - Linear discrimination dictionary learning for shape descriptors

    AU - Wang, Meng

    AU - Xie, Jin

    AU - Zhu, Fan

    AU - Fang, Yi

    PY - 2016/11/1

    Y1 - 2016/11/1

    N2 - The complexity and variation of 3D models have posed a lot of challenges in 3D shape retrieval area, for example, the invariant representation and retrieval of nonrigid and noisy 3D shapes. This paper proposed a supervised dictionary learning scheme called Linear Discrimination Dictionary Learning (LDDL) which can learn shape representations that are insensitive to 3D shape deformations in the same category and different for shapes from different categories in the meantime. Besides, it can extract the subtle differences between 3D shapes for fine-grained shapes. To be specific, in this paper, category-specific dictionaries are learnt to encode subtle visual differences of shapes among different categories, a shared dictionary is learnt to encode common patterns of shapes among all the categories; with the Linear Discriminant Analysis (LDA) constraint on the learnt descriptors, the new descriptors can have small within-class scatter and big between-class scatter. Our method is efficient in training and can obtain promising shape retrieval performance on representative shape benchmark datasets.

    AB - The complexity and variation of 3D models have posed a lot of challenges in 3D shape retrieval area, for example, the invariant representation and retrieval of nonrigid and noisy 3D shapes. This paper proposed a supervised dictionary learning scheme called Linear Discrimination Dictionary Learning (LDDL) which can learn shape representations that are insensitive to 3D shape deformations in the same category and different for shapes from different categories in the meantime. Besides, it can extract the subtle differences between 3D shapes for fine-grained shapes. To be specific, in this paper, category-specific dictionaries are learnt to encode subtle visual differences of shapes among different categories, a shared dictionary is learnt to encode common patterns of shapes among all the categories; with the Linear Discriminant Analysis (LDA) constraint on the learnt descriptors, the new descriptors can have small within-class scatter and big between-class scatter. Our method is efficient in training and can obtain promising shape retrieval performance on representative shape benchmark datasets.

    KW - Dictionary learning

    KW - Linear discriminant analysis

    KW - Shape retrieval

    UR - http://www.scopus.com/inward/record.url?scp=84978805277&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84978805277&partnerID=8YFLogxK

    U2 - 10.1016/j.patrec.2016.05.028

    DO - 10.1016/j.patrec.2016.05.028

    M3 - Article

    VL - 83

    SP - 349

    EP - 356

    JO - Pattern Recognition Letters

    JF - Pattern Recognition Letters

    SN - 0167-8655

    ER -