Geintra

Departamento de electronica Universidad de Alcala

Research lines

Access information on the Geintra research activity structure. More information


Work with us

Access to our current offer of grants and contracts. More information


Contact

You can contact us using different means. More information

    Robust System for Partially Occluded People Detection in RGB Images

    TitleRobust System for Partially Occluded People Detection in RGB Images
    Publication TypeConference Paper
    Año de publicación2017
    AutoresBaptista, M, Marron, M, Losada-Gutiérrez, C, Angel Cruz-Lozano, J, del Abril, A
    Idioma de publicaciónEnglish
    Conference NameInternational Conference on Computer Vision Theory and ApplicationsProceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications
    Volumen4
    Numero de volúmenes4
    Páginas532 - 539
    EditorialSCITEPRESS - Science and Technology Publications
    Conference LocationPorto, Portugal
    Fecha de publicación03/2017
    Numero ISBN978-989-758-225-7
    Palabras claveHistogram of Oriented Gradients (HOG), Partial Occlusion, People Detector, Support Vector Machine (SVM).
    DOI10.5220/0006165005320539
    Abstract

    This work presents a robust system for people detection in RGB images. The proposal increases the robustness of previous approaches against partial occlusions, and it is based on a bank of individual detectors whose results are combined using a multimodal association algorithm. Each individual detector is trained for a different body part (full body, half top, half bottom, half left and half right body parts). It consists of two elements: a feature extractor that obtains a Histogram of Oriented Gradients (HOG) descriptor, and a Support Vector Machine (SVM) for classification. Several experimental tests have been carried out in order to validate the proposal, using INRIA and CAVIAR datasets, that have been widely used by the scientific community.
    The obtained results show that the association of all the body part detections presents a better accuracy that any of the parts individually. Regarding the body parts, the best results have been obtained for the full body and half top body.

    DOI10.5220/0006165005320539
    AttachmentSize
    robust_system_for_partially_occluded_people_detection_2017_visapp.pdf1.63 MB

    Geintra © 2008-2024

    Diseño web por Hazhistoria