Cross-modality person re-identification人员重识别(person Re-ID),是视频监控中的一个重要问题,其目的是在摄像机视点上匹配行人的即时信息,目前,大多应用于RGB图像中,但例如在黑暗环境中,这样是远远不够的,在许多视觉系统中,红外成像是必不可少的。为此,需要将RGB图像与红外图像进行匹配,而红外图像是异构的,具有非常不同的视觉特征。对于人员重识别来说,这是一个非常具有挑战性的跨模态问题。目前已有的跨模态行人重识别大致有如下四个方向:低分辨、近红外、素描和文本。其任务是主要分两个子任务,一个是减少模态间的差异,通过在两个不同模态中寻找出具有同一个身份的人,另一个是单模态ReID任务,即在RGB图像或者红外图像内部通过对身份的约束找到同一个身份的人。
IEEE Transactions on Information Forensics and Security, (2020): 407-419
Visible thermal person re-identification (VT-REID) is a task of matching person images captured by thermal and visible cameras, which is an extremely important issue in night-time surveillance applications. Existing cross-modality recognition works mainly focus on learning sharab...
Cited by35BibtexViews36DOI
0
0
Zelong Zeng, Zhixiang Wang,Zheng Wang,Yinqiang Zheng, Yung-Yu Chuang, Shin'ichi Satoh
IEEE Transactions on Multimedia, (2020)
We show the ability of the proposed Identity Disentanglement network
Cited by7BibtexViews18DOI
0
0
european conference on computer vision, pp.229-247, (2020)
We present a dynamic dual-attentive aggregation learning framework for visible-infrared person re-identification
Cited by1BibtexViews57DOI
0
0
CVPR, pp.13376-13386, (2020)
We proposed a cross-modality sharedspecific feature transfer algorithm for cross-modality person ReID, which can utilize the specific features ignored by conventionally shared feature learning
Cited by0BibtexViews80DOI
0
0
IEEE Transactions on Industrial Electronics, pp.1-1, (2019)
With a large number of video surveillance systems installed for the requirement from industrial security, the task of object tracking, which aims to locate objects of interest in videos, is very important. Although numerous tracking algorithms for RGB videos have been developed i...
Cited by52BibtexViews44DOI
0
0
CVPR, pp.618-626, (2019)
We present Dual-level Discrepancy Reduction Learning network for the Infrared-Visible person REIDentification task that exhibits both modality discrepancy and appearance discrepancy
Cited by34BibtexViews22DOI
0
0
IEEE Transactions on Circuits and Systems for Video Technology, pp.1-1, (2019)
Person re-identification (re-ID) has been a popular topic in computer vision and pattern recognition communities for a decade. Several important milestones such as metric-based and deeply-learned re-ID in recent years have promoted this topic. However, most of existing re-ID work...
Cited by23BibtexViews18
0
0
IEEE Access, (2019): 67761-67771
Infrared sensors have been employed in many video surveillance systems for security monitoring, traffic management, etc..The past
Cited by13BibtexViews47DOI
0
0
CVPR, pp.10254-10263, (2019)
We have proposed a novel hierarchical cross-modality disentanglement method for Visible-infrared person re-identification
Cited by3BibtexViews16DOI
0
0
IEEE Transactions on Image Processing, no. 8 (2019): 3821-3835
Cross-view person identification (CVPI) from multiple temporally synchronized videos taken by multiple wearable cameras from different, varying views is a very challenging but important problem, which has attracted more interest recently. Current state-of-the-art performance of C...
Cited by2BibtexViews40DOI
0
0
Zheng Wang, Zhixiang Wang, Yang Wu, Jingdong Wang,Shin'ichi Satoh
arXiv: Computer Vision and Pattern Recognition, (2019)
When only one ground truth exists in the gallery, or we are caring about the true matches ranking in top positions of the returning list, the Cumulative Matching Characteristics curve is an acceptable selection
Cited by0BibtexViews12DOI
0
0
Jin Kyu Kang, Toan Minh Hoang,Kang Ryoung Park
IEEE Access, pp.1-1, (2019)
The method of combining the multi-scale Retinex with one stream network based on the IPVT-1, IPVT-2, and IIPVT demonstrated superior performance compared to other conventional person ReID methods based on a multimodal camera
Cited by0BibtexViews9DOI
0
0
Sulan Zhai, Pengpeng Shao, Xinyan Liang,Xin Wang
Neurocomputing, (2019): 172-181
This paper studies how to perform RGB-T object tracking in the correlation filter framework. Given the input RGB and thermal videos, we utilize the correlation filter for each modality due to its high performance in both of accuracy and speed. To take the interdependency between ...
Cited by0BibtexViews7DOI
0
0
Jian-Wu Lin, Hao Li
CoRR, (2019)
This paper addresses RGB-IR Re-ID by a feature learning framework based on hard pentaplet loss and identity loss as shown in Fig
Cited by0BibtexViews7DOI
0
0
IJCAI, pp.1092-1099, (2018)
To address the large cross-modality variations, a bi-directional cross-modality constrained top-ranking loss is employed to enhance the discriminability of the learnt feature representation
Cited by74BibtexViews36DOI
0
0
AAAI, (2018)
This paper investigates a cross-modal person reidentification problem, which is an important issue for many specific practical surveillance applications
Cited by63BibtexViews18DOI
0
0
Pingyang Dai,Rongrong Ji,Haibin Wang,Qiong Wu, Yuyu Huang
IJCAI, pp.677-683, (2018)
We propose a novel cross-modality generative adversarial network to learn discriminative common representations for RGB-IR person re-identification problem, which is formulated as cross-modality problem
Cited by53BibtexViews18DOI
0
0
Pattern Recognit. Lett., (2018): 12-20
Abstract To intelligently analyze and understand video content, a key step is to accurately perceive the motion of the interested objects in videos. To this end, the task of object tracking, which aims to determine the position and status of the interested object in consecutive...
Cited by50BibtexViews31
0
0
AAAI, (2018)
Even if some unreliable modality exists under some scenarios, the impact of such kind of modality would be suppressed, and the importance of reliable modality would be enhanced for more effective modality fusion
Cited by41BibtexViews51DOI
0
0
IJCAI, pp.3891-3897, (2018)
This paper focuses on a new issue, i.e., scale-adaptive low resolution person re-identification
Cited by37BibtexViews33DOI
0
0