计算机科学与探索 ›› 2021, Vol. 15 ›› Issue (9): 1740-1752.DOI: 10.3778/j.issn.1673-9418.2006104

• 图形图像 • 上一篇    下一篇

融合衣服特征迁移的行人重识别方法

苑春苗,牛瑛,郭涛,李鑫   

  1. 1. 天津工业大学 天津市自主智能技术与系统重点实验室,天津 300387
    2. 天津工业大学 计算机科学与技术学院,天津 300387
    3. 天津工业大学 电子与信息工程学院,天津 300387
    4. 天津工业大学 经济与管理学院,天津 300387
  • 出版日期:2021-09-01 发布日期:2021-09-06

Person Re-identification Combined with Clothing Information Transfer

YUAN Chunmiao, NIU Ying, GUO Tao, LI Xin   

  1. 1. Key Laboratory of Independent Intelligent Technology and System, Tiangong University, Tianjin 300387, China
    2. School of Computer Science and Technology, Tiangong University, Tianjin 300387, China
    3. School of Electronics and Information Engineering, Tiangong University, Tianjin 300387, China
    4. School of Economics and Management, Tiangong University, Tianjin 300387, China
  • Online:2021-09-01 Published:2021-09-06

摘要:

行人重新识别(Re-ID)是一种基于人体整体特征的身份识别方法,是指在图像或者视频序列中判断是否存在特定行人的技术。对于行人重识别方法来说,行人所穿着的衣服特征具有十分重要的影响。然而在某些特殊情况下(例如罪犯为了摆脱跟踪,更换衣服后逃跑等),目标行人在查询图像和图像库中的衣服特征是不同的,这使得目前的行人重识别方法难以将目标人物的图像与图像库中的人物相匹配。为了解决这一问题,提出了一种基于衣服特征迁移的行人重识别方法。给定一个目标人物的查询图像,将该查询图像行人的衣服特征迁移到图像库中的所有行人身上,以此消除衣服特征的差异。最后,该模型在数据集PRDDC上进行了验证,实验结果表明,与没有服装特征转移模块的算法相比,该算法有显著的改进。

关键词: 行人重识别(Re-ID), 特征迁移, 图像匹配

Abstract:

Person re-identification (Re-ID) is an identification method based on the overall characteristics of human body, which is usually used to judge whether there is a specific person in the image or video sequence. The clothing characteristics of persons have very important impact on Re-ID methods. But in some special scenarios, for example, a criminal maybe change his clothes in order to escape tracking, then his clothes will be different in the query image from those in the gallery image, which will cause the existing Re-ID algorithms difficult to match the target person in the gallery image. In order to solve this problem, a Re-ID model based on clothing information transfer is proposed in this paper. Given a query image of a target person, all the persons are re-dressed in the gallery image with the clothing feature in the query image, therefore the difference of clothing feature in person matching process can be eliminated. Finally, the model is verified on the dataset PRDDC (person Re-ID with different clothes), and experimental results show the significant improvement of the proposed algorithm compared with the algorithms without clothing feature transfer module.

Key words: person re-identification (Re-ID), feature transfer, image matching