The multi-scale covariance descriptor: Performances analysis in human detection

Walid Ayedi (1) , Hichem Snoussi (1) , Fethi Smach (2) , Mohamed Abid (2)

1 LM2S - Laboratoire Modélisation et Sûreté des Systèmes
2 Université de Sfax - University of Sfax

This paper presents a study on human detection using the multi-scale covariance descriptor (MSCOV) proposed in a previous work [1] in which we showed the performance of this descriptor for human re-identification. In this work, we evaluate its performance in human detection. We propose a fast tree based method for multi-scale features covariance computation. This method considerably speed up the image scan process for fast object detection. Furthermore, we experimentally evaluate the human detection performance using region covariance descriptor (COV), multi-scale covariance descriptor (MSCOV) and histogram of oriented gradients (HOG). In term of classifier, we consider the popular Support Vector Machines (SVM). The experiments are performed on both benchmarking datasets INRIA and MIT CBCL. Experiments on both datasets show the high detection performance of the MSCOV based detector.

Domains: Life Sciences [q-bio]; Bioengineering

Walid Ayedi, Hichem Snoussi, Fethi Smach, Mohamed Abid. The multi-scale covariance descriptor: Performances analysis in human detection. 2012 IEEE Workshop on Biometric Measurements and Systems for Security and Medical Applications (BIOMS), Sep 2012, Salerno, Italy. pp.1-5, ⟨10.1109/BIOMS.2012.6345773⟩. ⟨hal-02365400⟩