Title :
SCaLE: Supervised and Cascaded Laplacian Eigenmaps for Visual Object Recognition Based on Nearest Neighbors
Author :
Ruobing Wu ; Yizhou Yu ; Wenping Wang
Author_Institution :
Dept. of Comput. Sci., Univ. of Hong Kong, Hong Kong, China
Abstract :
Recognizing the category of a visual object remains a challenging computer vision problem. In this paper we develop a novel deep learning method that facilitates example-based visual object category recognition. Our deep learning architecture consists of multiple stacked layers and computes an intermediate representation that can be fed to a nearest-neighbor classifier. This intermediate representation is discriminative and structure-preserving. It is also capable of extracting essential characteristics shared by objects in the same category while filtering out nonessential 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 aforementioned properties. The first step computes a discrete mapping called supervised Laplacian Eigenmap. The second 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 recognition rates on a number of benchmark datasets.
Keywords :
computer vision; eigenvalues and eigenfunctions; image recognition; learning (artificial intelligence); regression analysis; benchmark datasets; computer vision problem; continuous mapping; deep learning method; discrete mapping; discriminative preserving; example-based visual object category recognition; intermediate representation; multiple stacked layers; nearest-neighbor classifier; nonlinear mapping; nonlinear regression; sequential steps; structure-preserving; supervised and cascaded Laplacian eigenmaps; Feature extraction; Kernel; Laplace equations; Optimization; Training; Vectors; Visualization; deep leanring; feature combination; image classification; object recognition;
Conference_Titel :
Computer Vision and Pattern Recognition (CVPR), 2013 IEEE Conference on
Conference_Location :
Portland, OR
DOI :
10.1109/CVPR.2013.117