SCaLE: Supervised and Cascaded Laplacian Eigenmaps for Visual Object Recognition Based on Nearest Neighbors

Recognizing the category of a visual object remains a challenging computer vision problem. In this paper we de-velop a novel deep learning method that facilitates example-based visual object category recognition. Our deep learn-ing architecture consists of multiple stacked layers and computes an intermediate representation that can be fed to a nearest-neighbor classifier. This intermediate repre-sentation is discriminative and structure-preserving. It is also capable of extracting essential characteristics shared by objects in the same category while filtering out nonessen-tial differences among them. Each layer in our model is a nonlinear mapping, whose parameters are learned through two sequential steps that are designed to achieve the afore-mentioned properties. The first step computes a discrete mapping called supervised Laplacian Eigenmap. The sec-ond step computes a continuous mapping from the discrete version through nonlinear regression. We have extensively tested our method and it achieves state-of-the-art recogni-tion rates on a number of benchmark datasets. 1