Learning Directional Local Pairwise Bases with Sparse Coding

Learning Directional Local Pairwise Bases with Sparse Coding
复制标题

DOI：
10.5244/c.24.32
复制
发表时间：
2010-08
期刊：
ArXiv
影响因子：
--
通讯作者：
Nobuyuki Morioka;S. Satoh
Nobuyuki Morioka;S. Satoh
中科院分区：
其他
文献类型：
--
作者：
Nobuyuki Morioka;S. Satoh

文献摘要

被引文献

相似文献

近年来，稀疏编码由于在学习有效码本方面优于k-均值聚类，在目标和场景识别任务中受到了广泛的关注。然而，从经验上看，这种码本需要相对大量的视觉单词，本质上是基础，才能达到较高的识别精度。因此，由于视觉单词的组合爆炸，使用该码本来表示高阶空间特征是不可行的，而高阶空间特征在捕捉场景和对象的不同属性方面同样重要。与以往许多利用高阶空间特征的方法不同，局部成对码本(LPC)是一种简单而有效的学习紧凑聚类的方法，该聚类表示具有k-均值的空间封闭描述符对。在LPC的基础上，提出了方向局部配对基(DLPB)，它利用稀疏编码学习一组紧凑的基集，以捕捉这些描述符之间的相关性，从而避免组合爆炸。此外，对于每个量化方向学习这样的基，从而明确地将方向信息添加到表示。我们已经用几个具有挑战性的对象和场景类别数据集来评估DLPB。我们的实验结果表明，DLPB在所有数据集上的性能都优于基线，并且在某些数据集上达到了最先进的性能。
Recently, sparse coding has been receiving much attention in object and scene recognition tasks because of its superiority in learning an effective codebook over k-means clustering. However, empirically, such codebook requires a relatively large number of visual words, essentially bases, to achieve high recognition accuracy. Therefore, due to the combinatorial explosion of visual words, it is infeasible to use this codebook to represent higher-order spatial features which are equally important in capturing distinct properties of scenes and objects. Contrasted with many previous techniques that exploit higher-order spatial features, Local Pairwise Codebook (LPC) is a simple and effective method to learn a compact set of clusters representing pairs of spatially close descriptors with k-means. Based on LPC, this paper proposes Directional Local Pairwise Bases (DLPB) that applies sparse coding to learn a compact set of bases capturing correlation between these descriptors, so to avoid the combinatorial explosion. Furthermore, such bases are learned for each quantized direction thereby explicitly adding directional information to the representation. We have evaluated DLPB with several challenging object and scene category datasets. Our experimental results show that DLPB outperforms the baselines across all datasets and achieves the state-of-the-art performance on some datasets.