Window detection in facades

Haider Ali, Christin Seifert, Nitin Jindal, Lucas Paletta, Gerhard Paar

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    42 Citations (Scopus)

    Abstract

    This work is about a novel methodology for window detection in urban environments and its multiple use in vision system applications. The presented method for window detection includes appropriate early image processing, provides a multi-scale Haar wavelet representation for the determination of image tiles which is then fed into a cascaded classifier for the task of window detection. The classifier is learned from a Gentle Adaboost driven cascaded decision tree on masked information from training imagery and is tested towards window based ground truth information which is together with the original building image databases publicly available. The experimental results demonstrate that single window detection is to a sufficient degree successful, e.g., for the purpose of building recognition, and, furthermore, that the classifier is in general capable to provide a region of interest operator for the interpretation of urban environments. The extraction of this categorical information is beneficial to index into search spaces for urban object recognition as well as aiming towards providing a semantic focus for accurate post-processing in 3D information processing systems. Targeted applications are (i) mobile services on uncalibrated imagery, e.g. , for tourist guidance, (ii) sparse 3D city modeling, and (iii) deformation analysis from high resolution imagery.
    Original languageEnglish
    Title of host publicationProceedings - 14th International conference on Image Analysis and Processing, ICIAP 2007
    Place of PublicationPiscataway, NJ
    PublisherIEEE
    Pages837-842
    Number of pages6
    ISBN (Print)978-0-7695-2877-9
    DOIs
    Publication statusPublished - 2007
    Event14th International Conference on Image Analysis and Processing, ICIAP 2007 - Modena, Italy
    Duration: 10 Sep 200714 Sep 2007
    Conference number: 14
    http://imagelab.ing.unimore.it/iciap07/

    Conference

    Conference14th International Conference on Image Analysis and Processing, ICIAP 2007
    Abbreviated titleICIAP
    CountryItaly
    CityModena
    Period10/09/0714/09/07
    Internet address

    Fingerprint

    Facades
    Classifiers
    Adaptive boosting
    Object recognition
    Tile
    Decision trees
    Image processing
    Semantics
    Processing

    Keywords

    • Building recognition
    • Cascaded classifiers
    • Mobile vision systems
    • Urban environments
    • Window detection

    Cite this

    Ali, H., Seifert, C., Jindal, N., Paletta, L., & Paar, G. (2007). Window detection in facades. In Proceedings - 14th International conference on Image Analysis and Processing, ICIAP 2007 (pp. 837-842). Piscataway, NJ: IEEE. https://doi.org/10.1109/ICIAP.2007.4362880
    Ali, Haider ; Seifert, Christin ; Jindal, Nitin ; Paletta, Lucas ; Paar, Gerhard. / Window detection in facades. Proceedings - 14th International conference on Image Analysis and Processing, ICIAP 2007. Piscataway, NJ : IEEE, 2007. pp. 837-842
    @inproceedings{f46980aab6db4cf18a0298138aef9284,
    title = "Window detection in facades",
    abstract = "This work is about a novel methodology for window detection in urban environments and its multiple use in vision system applications. The presented method for window detection includes appropriate early image processing, provides a multi-scale Haar wavelet representation for the determination of image tiles which is then fed into a cascaded classifier for the task of window detection. The classifier is learned from a Gentle Adaboost driven cascaded decision tree on masked information from training imagery and is tested towards window based ground truth information which is together with the original building image databases publicly available. The experimental results demonstrate that single window detection is to a sufficient degree successful, e.g., for the purpose of building recognition, and, furthermore, that the classifier is in general capable to provide a region of interest operator for the interpretation of urban environments. The extraction of this categorical information is beneficial to index into search spaces for urban object recognition as well as aiming towards providing a semantic focus for accurate post-processing in 3D information processing systems. Targeted applications are (i) mobile services on uncalibrated imagery, e.g. , for tourist guidance, (ii) sparse 3D city modeling, and (iii) deformation analysis from high resolution imagery.",
    keywords = "Building recognition, Cascaded classifiers, Mobile vision systems, Urban environments, Window detection",
    author = "Haider Ali and Christin Seifert and Nitin Jindal and Lucas Paletta and Gerhard Paar",
    year = "2007",
    doi = "10.1109/ICIAP.2007.4362880",
    language = "English",
    isbn = "978-0-7695-2877-9",
    pages = "837--842",
    booktitle = "Proceedings - 14th International conference on Image Analysis and Processing, ICIAP 2007",
    publisher = "IEEE",
    address = "United States",

    }

    Ali, H, Seifert, C, Jindal, N, Paletta, L & Paar, G 2007, Window detection in facades. in Proceedings - 14th International conference on Image Analysis and Processing, ICIAP 2007. IEEE, Piscataway, NJ, pp. 837-842, 14th International Conference on Image Analysis and Processing, ICIAP 2007, Modena, Italy, 10/09/07. https://doi.org/10.1109/ICIAP.2007.4362880

    Window detection in facades. / Ali, Haider; Seifert, Christin; Jindal, Nitin; Paletta, Lucas; Paar, Gerhard.

    Proceedings - 14th International conference on Image Analysis and Processing, ICIAP 2007. Piscataway, NJ : IEEE, 2007. p. 837-842.

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    TY - GEN

    T1 - Window detection in facades

    AU - Ali, Haider

    AU - Seifert, Christin

    AU - Jindal, Nitin

    AU - Paletta, Lucas

    AU - Paar, Gerhard

    PY - 2007

    Y1 - 2007

    N2 - This work is about a novel methodology for window detection in urban environments and its multiple use in vision system applications. The presented method for window detection includes appropriate early image processing, provides a multi-scale Haar wavelet representation for the determination of image tiles which is then fed into a cascaded classifier for the task of window detection. The classifier is learned from a Gentle Adaboost driven cascaded decision tree on masked information from training imagery and is tested towards window based ground truth information which is together with the original building image databases publicly available. The experimental results demonstrate that single window detection is to a sufficient degree successful, e.g., for the purpose of building recognition, and, furthermore, that the classifier is in general capable to provide a region of interest operator for the interpretation of urban environments. The extraction of this categorical information is beneficial to index into search spaces for urban object recognition as well as aiming towards providing a semantic focus for accurate post-processing in 3D information processing systems. Targeted applications are (i) mobile services on uncalibrated imagery, e.g. , for tourist guidance, (ii) sparse 3D city modeling, and (iii) deformation analysis from high resolution imagery.

    AB - This work is about a novel methodology for window detection in urban environments and its multiple use in vision system applications. The presented method for window detection includes appropriate early image processing, provides a multi-scale Haar wavelet representation for the determination of image tiles which is then fed into a cascaded classifier for the task of window detection. The classifier is learned from a Gentle Adaboost driven cascaded decision tree on masked information from training imagery and is tested towards window based ground truth information which is together with the original building image databases publicly available. The experimental results demonstrate that single window detection is to a sufficient degree successful, e.g., for the purpose of building recognition, and, furthermore, that the classifier is in general capable to provide a region of interest operator for the interpretation of urban environments. The extraction of this categorical information is beneficial to index into search spaces for urban object recognition as well as aiming towards providing a semantic focus for accurate post-processing in 3D information processing systems. Targeted applications are (i) mobile services on uncalibrated imagery, e.g. , for tourist guidance, (ii) sparse 3D city modeling, and (iii) deformation analysis from high resolution imagery.

    KW - Building recognition

    KW - Cascaded classifiers

    KW - Mobile vision systems

    KW - Urban environments

    KW - Window detection

    U2 - 10.1109/ICIAP.2007.4362880

    DO - 10.1109/ICIAP.2007.4362880

    M3 - Conference contribution

    SN - 978-0-7695-2877-9

    SP - 837

    EP - 842

    BT - Proceedings - 14th International conference on Image Analysis and Processing, ICIAP 2007

    PB - IEEE

    CY - Piscataway, NJ

    ER -

    Ali H, Seifert C, Jindal N, Paletta L, Paar G. Window detection in facades. In Proceedings - 14th International conference on Image Analysis and Processing, ICIAP 2007. Piscataway, NJ: IEEE. 2007. p. 837-842 https://doi.org/10.1109/ICIAP.2007.4362880