Histograms of Oriented Gradients for Human Detection

Navneet Dalal 1 Bill Triggs 1
1 LEAR - Learning and recognition in vision
GRAVIR - IMAG - Graphisme, Vision et Robotique, Inria Grenoble - Rhône-Alpes, CNRS - Centre National de la Recherche Scientifique : FR71
Abstract : We study the question of feature sets for robust visual object recognition, adopting linear SVM based human detection as a test case. After reviewing existing edge and gradient based descriptors, we show experimentally that grids of Histograms of Oriented Gradient (HOG) descriptors significantly outperform existing feature sets for human detection. We study the influence of each stage of the computation on performance, concluding that fine-scale gradients, fine orientation binning, relatively coarse spatial binning, and high-quality local contrast normalization in overlapping descriptor blocks are all important for good results. The new approach gives near-perfect separation on the original MIT pedestrian database, so we introduce a more challenging dataset containing over 1800 annotated human images with a large range of pose variations and backgrounds.
Type de document :
Communication dans un congrès
Cordelia Schmid and Stefano Soatto and Carlo Tomasi. International Conference on Computer Vision & Pattern Recognition (CVPR '05), Jun 2005, San Diego, United States. IEEE Computer Society, 1, pp.886--893, 2005, <http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=1467360>. <10.1109/CVPR.2005.177>
Liste complète des métadonnées

https://hal.inria.fr/inria-00548512
Contributeur : Thoth Team <>
Soumis le : lundi 20 décembre 2010 - 09:08:29
Dernière modification le : vendredi 7 janvier 2011 - 15:51:48
Document(s) archivé(s) le : lundi 21 mars 2011 - 03:07:17

Identifiants

Collections

IMAG | INRIA | UGA

Citation

Navneet Dalal, Bill Triggs. Histograms of Oriented Gradients for Human Detection. Cordelia Schmid and Stefano Soatto and Carlo Tomasi. International Conference on Computer Vision & Pattern Recognition (CVPR '05), Jun 2005, San Diego, United States. IEEE Computer Society, 1, pp.886--893, 2005, <http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=1467360>. <10.1109/CVPR.2005.177>. <inria-00548512>

Partager

Métriques

Consultations de
la notice

2179

Téléchargements du document

23422