A committee machine scheme for feature map fusion under uncertainty: the face detection case

Abstract

Feature map fusion in Visual Attention (VA) models is by definition an uncertain procedure. One of the major impediments in extending the static VA architecture proposed by Itti et al. (2000) to account for motion or other information is the lack of justification on how to integrate the various channels. We propose an innovative committee machine scheme that allows for dynamically changing the committee members (maps) and weighting them according to the confidence level of their estimation. Through this machine we handle the extensions on Itti's model; we add a motion channel and a prior knowledge channel which accounts for the conscious search performed by humans when looking for faces in a scene. The experimental results, obtained when considering face detection, show that the map fusion, through the proposed committee machine, leads to significantly better statistical results when compared with the simple skin-based face detection method.