Face Spotting in Color Images using Recursive Neural Networks

Images containing faces are fundamental for vision–based human computer interaction. Several problems concerning the presence of faces in images, including face recognition, pose estimation, and expression recognition have focused the research efforts in the last few years. Moreover , those problems are usually solved having assumed that the face was previously localized, often via heuristics based on prototypes of the whole face or significant details. Therefore, each fully automated system, which is able to analyze the information contained in images of faces, requires an efficient face localization method. In this paper, we propose a novel approach to the solution of the face localization problem using recursive neural networks. In particular, the proposed approach assumes a graph–based representation of images that combines structural and sub–symbolic visual features. Each image is represented by a forest of trees. Such trees are then processed by recursive neural networks, in order to establish the eventual presence and the position of faces inside the image. Some preliminary experiments on snapshots from video sequences are reported, showing very promising results.