An Efficient Search Method Based on Dynamic Attention Map by Ising Model

Abstract

This paper presents Dynamic Attention Map by Ising model for detection. In general, a detector can not know where faces there are and how many faces there are in advance. Therefore, the detector must search the whole regions on the image and requires much computational time. To speed the search, the information obtained at previous search points should be used effectively. In order to use the likelihood of obtained at previous search points effectively, Ising model is adopted to detection. Ising model has the two-state spins; up and down. The state of a spin is updated by depending on the neighboring spins and an external magnetic field. Ising spins are assigned to face and non-face states of detection. In addition, the measured likelihood of is integrated into the energy function of Ising model as the external magnetic field. It is confirmed that candidates would be reduced effectively by spin flip dynamics. To improve the search performance further, the single level Ising search method is extended to the multilevel Ising search. The interactions between two layers which are characterized by the renormalization group method is used to reduce the candidates. The effectiveness of the multilevel Ising search method is also confirmed by the comparison with the single level Ising search method.