A discriminative representation of convolutional features for indoor scene recognition

Salman H. Khan, Munawar Hayat, Mohammed Bennamoun, Roberto Togneri, Ferdous Sohel

    Research output: Contribution to journalArticle

    31 Citations (Scopus)
    4 Downloads (Pure)

    Abstract

    Indoor scene recognition is a multi-faceted and challenging problem due to the diverse intra-class variations and the confusing inter-class similarities that characterize such scenes. This paper presents a novel approach that exploits rich mid-level convolutional features to categorize indoor scenes. Traditional convolutional features retain the global spatial structure, which is a desirable property for general object recognition. We, however, argue that the structure-preserving property of the convolutional neural network activations is not of substantial help in the presence of large variations in scene layouts, e.g., in indoor scenes. We propose to transform the structured convolutional activations to another highly discriminative feature space. The representation in the transformed space not only incorporates the discriminative aspects of the target data set but also encodes the features in terms of the general object categories that are present in indoor scenes. To this end, we introduce a new large-scale data set of 1300 object categories that are commonly present in indoor scenes. Our proposed approach achieves a significant performance boost over the previous state-of-the-art approaches on five major scene classification data sets
    Original languageEnglish
    Pages (from-to)3372-3383
    Number of pages12
    JournalIEEE Transactions on Image Processing
    Volume25
    Issue number7
    DOIs
    Publication statusPublished - 2016

    Fingerprint

    Chemical activation
    Object recognition
    Neural networks

    Cite this

    Khan, Salman H. ; Hayat, Munawar ; Bennamoun, Mohammed ; Togneri, Roberto ; Sohel, Ferdous. / A discriminative representation of convolutional features for indoor scene recognition. In: IEEE Transactions on Image Processing. 2016 ; Vol. 25, No. 7. pp. 3372-3383.
    @article{848bfe002f1345fe9140199411d3eeb6,
    title = "A discriminative representation of convolutional features for indoor scene recognition",
    abstract = "Indoor scene recognition is a multi-faceted and challenging problem due to the diverse intra-class variations and the confusing inter-class similarities that characterize such scenes. This paper presents a novel approach that exploits rich mid-level convolutional features to categorize indoor scenes. Traditional convolutional features retain the global spatial structure, which is a desirable property for general object recognition. We, however, argue that the structure-preserving property of the convolutional neural network activations is not of substantial help in the presence of large variations in scene layouts, e.g., in indoor scenes. We propose to transform the structured convolutional activations to another highly discriminative feature space. The representation in the transformed space not only incorporates the discriminative aspects of the target data set but also encodes the features in terms of the general object categories that are present in indoor scenes. To this end, we introduce a new large-scale data set of 1300 object categories that are commonly present in indoor scenes. Our proposed approach achieves a significant performance boost over the previous state-of-the-art approaches on five major scene classification data sets",
    keywords = "deep-learning, convolutional-neural-networks, indoor-scene-classification",
    author = "Khan, {Salman H.} and Munawar Hayat and Mohammed Bennamoun and Roberto Togneri and Ferdous Sohel",
    year = "2016",
    doi = "10.1109/tip.2016.2567076",
    language = "English",
    volume = "25",
    pages = "3372--3383",
    journal = "IEEE Transactions on Image Processing",
    issn = "1057-7149",
    publisher = "IEEE, Institute of Electrical and Electronics Engineers",
    number = "7",

    }

    A discriminative representation of convolutional features for indoor scene recognition. / Khan, Salman H.; Hayat, Munawar; Bennamoun, Mohammed; Togneri, Roberto; Sohel, Ferdous.

    In: IEEE Transactions on Image Processing, Vol. 25, No. 7, 2016, p. 3372-3383.

    Research output: Contribution to journalArticle

    TY - JOUR

    T1 - A discriminative representation of convolutional features for indoor scene recognition

    AU - Khan, Salman H.

    AU - Hayat, Munawar

    AU - Bennamoun, Mohammed

    AU - Togneri, Roberto

    AU - Sohel, Ferdous

    PY - 2016

    Y1 - 2016

    N2 - Indoor scene recognition is a multi-faceted and challenging problem due to the diverse intra-class variations and the confusing inter-class similarities that characterize such scenes. This paper presents a novel approach that exploits rich mid-level convolutional features to categorize indoor scenes. Traditional convolutional features retain the global spatial structure, which is a desirable property for general object recognition. We, however, argue that the structure-preserving property of the convolutional neural network activations is not of substantial help in the presence of large variations in scene layouts, e.g., in indoor scenes. We propose to transform the structured convolutional activations to another highly discriminative feature space. The representation in the transformed space not only incorporates the discriminative aspects of the target data set but also encodes the features in terms of the general object categories that are present in indoor scenes. To this end, we introduce a new large-scale data set of 1300 object categories that are commonly present in indoor scenes. Our proposed approach achieves a significant performance boost over the previous state-of-the-art approaches on five major scene classification data sets

    AB - Indoor scene recognition is a multi-faceted and challenging problem due to the diverse intra-class variations and the confusing inter-class similarities that characterize such scenes. This paper presents a novel approach that exploits rich mid-level convolutional features to categorize indoor scenes. Traditional convolutional features retain the global spatial structure, which is a desirable property for general object recognition. We, however, argue that the structure-preserving property of the convolutional neural network activations is not of substantial help in the presence of large variations in scene layouts, e.g., in indoor scenes. We propose to transform the structured convolutional activations to another highly discriminative feature space. The representation in the transformed space not only incorporates the discriminative aspects of the target data set but also encodes the features in terms of the general object categories that are present in indoor scenes. To this end, we introduce a new large-scale data set of 1300 object categories that are commonly present in indoor scenes. Our proposed approach achieves a significant performance boost over the previous state-of-the-art approaches on five major scene classification data sets

    KW - deep-learning

    KW - convolutional-neural-networks

    KW - indoor-scene-classification

    U2 - 10.1109/tip.2016.2567076

    DO - 10.1109/tip.2016.2567076

    M3 - Article

    VL - 25

    SP - 3372

    EP - 3383

    JO - IEEE Transactions on Image Processing

    JF - IEEE Transactions on Image Processing

    SN - 1057-7149

    IS - 7

    ER -