Heterogeneous Distance Learning Based on Kernel Analysis-Synthesis Dictionary for Semi-Supervised Image to Video Person Re-Identification

Heterogeneous Distance Learning Based on Kernel Analysis-Synthesis Dictionary for Semi-Supervised Image to Video Person Re-Identification
复制标题

基于核分析-合成字典的半监督图像到视频行人重识别的异构远程学习

DOI:
10.1109/access.2020.3024289
复制
发表时间:
2020
期刊:
影响因子:
3.9
通讯作者:
张帆
张帆
中科院分区:
计算机科学3区
文献类型:
--
作者:
朱小柯;叶鹏飞;荆晓远;张新玉;崔翔;陈小潘;张帆

文献摘要

参考文献

相似文献

图像到视频个人重新识别(IVPR),即,行人视频与图像的匹配是一项重要的实际工作。虽然已经提出了几种方法用于IVPR,但这些方法大多在监督设置下研究IVPR问题,并且需要大量标记的图像-视频对进行训练。在这篇文章中,我们研究了半监督环境下的IVPR问题,并提出了一种基于核分析-综合字典的异构远程学习(KADDL)方法。具体来说,KADDL首先从核空间中的标记和未标记的训练图像-视频数据中学习两对核分析-合成字典。利用学习的字典对,可以将异构的图像和视频特征转换为同一表示空间的编码系数,从而弥补图像和视频之间的差距。然后,KADDL在变换后的编码系数上学习一个有区别的距离度量,使得正图像-视频对的编码系数相似,而负图像-视频对的编码系数不相似。为了更好地利用未标记数据,我们进一步设计了一个基于可靠性的半监督策略KADDL。在几个公开的行人序列数据集上的实验证明了该方法的有效性。
Image to video person re-identification (IVPR), i.e., matching between pedestrian video and image, is an important task in practice. Although several methods have been presented for IVPR, most of these methods investigate the IVPR problem under the supervised setting, and require a large number of labeled image-video pairs for training. In this article, we study the IVPR problem under the semi-supervised setting, and propose a Kernel Analysis-synthesis Dictionary based heterogeneous Distance Learning (KADDL) approach. Specifically, KADDL first learns two pairs of kernel analysis-synthesis dictionaries from the labeled and unlabeled training image-video data in the kernel space. With the learned dictionary pairs, the heterogeneous image and video features can be transformed into coding coefficients of the same representation space, such that the gap between image and video can be bridged. Then, KADDL learns a discriminative distance metric over the transformed coding coefficients, to make the coding coefficients of positive image-video pair become similar, while those of negative image-video pair dissimilar. To make better use of the unlabeled data, we further designed a reliability-based semi-supervised strategy for KADDL. Experiments on several publicly available pedestrian sequence datasets demonstrate the effectiveness of the proposed approach.
CGAN-TM:一种用于人员重新识别的新型域到域传输方法
DOI: 10.1109/tip.2020.2985545
发表时间: 2020-04
影响因子: 10.6
作者:
Yingzhi Tang;Xi Yang;Nannan Wang;Bin Song;Xinbo Gao
通讯作者: Xinbo Gao
DOI: 10.1109/tpami.2015.2453984
发表时间: 2016-03
影响因子: 23.6
作者:
Weishi Zheng;S. Gong;T. Xiang
通讯作者: Weishi Zheng;S. Gong;T. Xiang
通过同时学习视频内和视频间距离度量来进行基于视频的人员重新识别
DOI: 10.1109/tip.2018.2861366
发表时间: 2016-07
期刊: IEEE TRANSACTIONS ON IMAGE PROCESSING,
影响因子: --
作者:
Zhu Xiaoke;Jing Xiao-Yuan;You Xinge;Zhang Xinyu;Zhang Taiping
通讯作者: Zhang Taiping
DOI: 10.1109/cvpr.2019.00375
发表时间: 2019-06
期刊: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
影响因子: --
作者:
Q. Yang;Hong-Xing Yu;Ancong Wu;Weishi Zheng
通讯作者: Q. Yang;Hong-Xing Yu;Ancong Wu;Weishi Zheng
DOI: --
发表时间: 2018-10
期刊: ArXiv
影响因子: --
作者:
Zhongwei Xie;Lin Li;X. Zhong;Luo Zhong
通讯作者: Zhongwei Xie;Lin Li;X. Zhong;Luo Zhong