相关参考文献
注意:仅列出部分参考文献,下载原文获取全部文献信息。
Article
Computer Science, Artificial Intelligence
Haojie Liu et al.
Summary: Visible-Infrared person reidentification is a challenging matching problem. Existing methods usually bridge the modality gap with only feature-level constraints, ignoring pixel-level variations. This article proposes a novel spectrum-aware feature augmentation network, SFANet, to address this problem. Improvements are made at the feature level and in the loss function, achieving very competitive results.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Mang Ye et al.
Summary: Person re-identification (Re-ID) has gained significant interest in the computer vision community, with the advancement of deep neural networks. It is categorized into closed-world and open-world settings. While closed-world setting has achieved inspiring success, the research focus has shifted to the more challenging open-world setting. We summarize the open-world Re-ID in five different aspects and introduce a new evaluation metric. This metric provides an additional criteria for evaluating Re-ID systems in real applications.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Article
Computer Science, Artificial Intelligence
Ziyu Wei et al.
Summary: The proposed method in this article utilizes adversarial learning and triplet loss-based metric learning to improve the performance of person re-identification between visible and infrared modalities. By competing methods such as adversarial learning and triplet loss, the model obtains more effective modality-shareable features, shrinking the cross-modality gap and enhancing feature discriminability.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Kaiyang Zhou et al.
Summary: In this paper, we propose an effective person re-identification (re-ID) model that can distinguish similar-looking people and can be deployed across datasets without any adaptation. The model consists of an omni-scale network (OSNet) for feature learning and instance normalisation (IN) layers for improving generalisation. Experimental results show that the proposed model outperforms existing re-ID models in terms of performance, both in the same-dataset and cross-dataset settings.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Article
Computer Science, Artificial Intelligence
Yang Zhao et al.
Summary: This paper proposes a deep part-aware representation learning method for person retrieval, which achieves performance improvement through an improved triplet loss and a localization branch.
PATTERN RECOGNITION
(2021)
Article
Computer Science, Artificial Intelligence
Peng Zhang et al.
Summary: VI-reID aims to automatically retrieve pedestrian of interest exposed to sensors in different modalities, but existing work mainly focuses on tackling the modality difference without investigating discriminant information at a fine-grained level. The proposed DAPR framework addresses this issue by simultaneously alleviating modality bias and mining different levels of discriminant representations.
IMAGE AND VISION COMPUTING
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Hyunjong Park et al.
Summary: In this paper, we propose a novel feature learning framework for visible-infrared person re-identification (VI-reID) by exploiting dense correspondences between cross-modal person images. This approach effectively addresses cross-modal discrepancies and encourages pixel-wise associations between cross-modal local features, further promoting discriminative feature learning for VI-reID. Extensive experiments and analyses show that our approach significantly outperforms the state of the art on standard VI-reID benchmarks.
2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021)
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Xudong Tian et al.
Summary: The paper introduces a new strategy called VSD to help IB principle better understand the correlation between representation and label, and enhance the robustness of representation through two strategies, VCD and VML. Experimental results demonstrate superior performance in cross-modal person Re-ID.
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Yehansen Chen et al.
Summary: In this paper, a novel Neural Feature Search (NFS) method is proposed to automate feature selection in RGB-IR person re-identification (RGB-IR ReID) problem, achieving superior performance compared to existing techniques. NFS combines dual-level feature search space and differentiable search strategy to adaptively filter noise and focus on informative parts of human bodies.
2021 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2021
(2021)
Article
Computer Science, Artificial Intelligence
Yuanxin Zhu et al.
Article
Computer Science, Artificial Intelligence
Haijun Liu et al.
Article
Computer Science, Artificial Intelligence
Ziyang Wang et al.
Article
Computer Science, Artificial Intelligence
Xiang Bai et al.
PATTERN RECOGNITION
(2020)
Article
Computer Science, Artificial Intelligence
Pingyu Wang et al.
PATTERN RECOGNITION LETTERS
(2020)
Proceedings Paper
Computer Science, Artificial Intelligence
Jinxian Liu et al.
2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2018)
Article
Chemistry, Analytical
Dat Tien Nguyen et al.
Proceedings Paper
Computer Science, Artificial Intelligence
Xuelin Qian et al.
2017 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV)
(2017)