Publication

Enhancing real-time human detection based on histograms of oriented gradients

Conference Article

Conference

International Conference on Computer Recognition Systems (CORES)

Edition

5th

Pages

739-746

Doc link

http://dx.doi.org/10.1007/978-3-540-75175-5_91

File

Download the digital copy of the doc pdf document

Authors

Projects associated

Abstract

In this paper we propose a human detection framework based on an enhanced version of Histogram of Oriented Gradients (HOG) features. These feature descriptors are computed with the help of a precalculated histogram of square-blocks. This novel method outperforms the integral of oriented histograms allowing the calculation of a single feature four times faster. Using Adaboost for HOG feature selection and Support Vector Machine as weak classifier, we build up a real-time human classifier with an excellent detection rate.

Categories

computer vision.

Scientific reference

M. Pedersoli, J. Gonzàlez, B. Chakraborty and J.J. Villanueva. Enhancing real-time human detection based on histograms of oriented gradients, 5th International Conference on Computer Recognition Systems, 2007, Wroclaw, Polònia, in Computer Recognition Systems 2, Vol 45 of Advances in Soft Computing, pp. 739-746, 2007, Springer.