Usually, object detection is performed directly on (normalized) gray values or gray primitives like gradients or Haar-like features. In that case the learning of relationships between gray primitives, that describe the structure of the object, is the complete responsibility of the classifier. We propose to apply more knowledge about the image structure in the preprocessing step, by computing local isophote directions and curvatures, in order to supply the classifier with much more informative image structure features. However, a periodic feature space, like orientation, is unsuited for common classification methods. Therefore, we split orientation into two more suitable components. Experiments show that the isophote features result in better detection performance than intensities, gradients or Haar-like features.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Isophote properties as features for object detection


    Contributors:


    Publication date :

    2005-01-01


    Size :

    611540 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Head pose estimation using isophote features for driver assistance systems

    Zhang, Xuetao / Zheng, Nanning / Mu, Fan et al. | IEEE | 2009


    Head Pose Estimation Using Isophote Features for Driver Assistance Systems

    Zhang, X. / Zheng, N. / Mu, F. et al. | British Library Conference Proceedings | 2009


    Visualizing Object Detection Features

    Vondrick, C. / Khosla, A. / Pirsiavash, H. et al. | British Library Online Contents | 2016


    New object detection features in the OpenCV library

    Druzhkov, P. N. / Erukhimov, V. L. / Zolotykh, N. Y. et al. | British Library Online Contents | 2011


    Sharing features: efficient boosting procedures for multiclass object detection

    Torralba, A. / Murphy, K.P. / Freeman, W.T. | IEEE | 2004