针对行人再识别中目标的类间显著性外观特征不稳定的问题, 在多显著性融合基础上提出一种新的行人再识别算法。采用两级流形排序算法学习图像的内在显著性特征, 并与已有的类间显著性特征进行融合, 提出了一种更为准确的多显著性融合特征描述方法。方法不仅考虑了特征块与其它行人图像特征块间的类间显著性, 而且考虑了特征块在所在图像中的内在显著性, 从而加强对特征块的显著性描述。实验结果表明, 与同类方法相比, 本文算法能获取更为准确的目标显著性描述, 具有较高的行人再识别率。

Aiming for objects' unsteadied inter-salience properties in person re-identification, a new algorithm of person re-identification is proposed in the base of multi-salience fusion. Two-stage Manifold Ranking (MR) is established to obtain image intra-salience. A more accurate multi-salience fusion description method is presented after combining intra-salience with existing inter-salience. The approach considers not only inter-salience of the patch with other person images' patches, but also the intra-salience of the patch in the own image. It improves the description of the patch in salience. Compared with the similar algorithm, the method can describe the salience of objects more accurately, reaching high re-identification rate.