Laplacian regularized locality-constrained coding for image classification

详细信息    查看全文

• 作者：Huaqing Min^a ; ^{hqmin@scut.edu.cn" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Mingjie Liang^b ; ^{mjie.liang@gmail.com" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Ronghua Luo^b ; ^{rhluo@scut.edu.cn" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Jinhui Zhu^a ; ^{csjhzhu@scut.edu.cn" class="auth_mail" title="E-mail the corresponding author}Author Vitae
• 关键词：Image classification ; Feature coding ; Locality-constrained ; Laplacian regularization
• 刊名：Neurocomputing
• 出版年：2016
• 出版时间：1 January 2016
• 年：2016
• 卷：171
• 期：Complete
• 页码：1486-1495
• 全文大小：1673 K

文摘

Feature coding, which encodes local features extracted from an image with a codebook and generates a set of codes for efficient image representation, has shown very promising results in image classification. Vector quantization is the most simple but widely used method for feature coding. However, it suffers from large quantization errors and leads to dissimilar codes for similar features. To alleviate these problems, we propose Laplacian Regularized Locality-constrained Coding (LapLLC), wherein a locality constraint is used to favor nearby bases for encoding, and Laplacian regularization is integrated to preserve the code consistency of similar features. By incorporating a set of template features, the objective function used by LapLLC can be decomposed, and each feature is encoded by solving a linear system. Additionally, k nearest neighbor technique is employed to construct a much smaller linear system, so that fast approximated coding can be achieved. Therefore, LapLLC provides a novel way for efficient feature coding. Our experiments on a variety of image classification tasks demonstrated the effectiveness of this proposed approach.