An extension of Separable Lattice 2-D HMMS for rotational data variations
An extension of Separable Lattice 2-D HMMS for rotational data variations
复制标题
DOI：
10.1109/icassp.2010.5495735
复制
发表时间：
2010-03
期刊：
影响因子：
--
通讯作者：
Akira Tamamori;Yoshihiko Nankaku;K. Tokuda
中科院分区：
文献类型：
--
作者：
Akira Tamamori;Yoshihiko Nankaku;K. Tokuda
本文通过扩展可分晶格二维hmm (sl2d - hmm)，提出了一种新的处理旋转数据变化的生成模型。在图像识别中，诸如大小、位置和旋转等几何变化会降低性能，因此需要进行归一化。sl2d - hmm可以在水平和垂直方向上进行弹性匹配；这使得对大小和位置的不变性进行建模成为可能。为了处理旋转变化，我们引入了额外的HMM状态，这些状态表示观察线的状态排列在特定方向上的移动。人脸识别实验表明，该方法显著提高了旋转变化数据的识别性能。
This paper proposes a new generative model which can deal with rotational data variations by extending Separable Lattice 2-D HMMs (SL2D-HMMs). In image recognition, geometrical variations such as size, location and rotation degrade the performance, therefore normalization is required. SL2D-HMMs can perform an elastic matching in both horizontal and vertical directions; this makes it possible to model invariances to size and location. To deal with rotational variations, we introduce additional HMM states which represent the shifts of the state alignments of the observation lines in a particular direction. Face recognition experiments show that the proposed method improves the performance significantly for rotational variation data.