An object recognition model with kernel method and its application for detection of rotated faces

摘要

An object recognition model is constructed in feature space into which an input vector is mapped by a particular non-linear function. In this model, recognition process is realized by comparison of the mapped input vector to reconstructed vectors that are generated initially and refined iteratively by linear combination of the basis vectors in feature space. The kernel method allows efficient computation of the model in the high dimensional feature space. Some experiments based on our model show effective detection of rotated facial images and description of rotation angle.