Article
Engineering, Electrical & Electronic
Jiawei Chen, Junlin Hu
Summary: Metric learning aims to learn a mapping relationship to reduce intraclass distance and increase interclass distance. The proposed weakly supervised compositional metric learning method effectively optimizes weight combination and verification accuracy.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Engineering, Electrical & Electronic
Xin Yuan, Xin Xu, Zheng Wang, Kai Zhang, Wei Liu, Ruimin Hu
Summary: By automatically designing loss functions with specific evaluation metrics, the accuracy of identifying objects across multiple cameras is improved, outperforming other relevant loss functions.
IEEE JOURNAL OF SELECTED TOPICS IN SIGNAL PROCESSING
(2023)
Article
Computer Science, Artificial Intelligence
Wanyin Wu, Dapeng Tao, Hao Li, Zhao Yang, Jun Cheng
Summary: This study summarizes different types of features and metric learning approaches for person re-identification from a label attributes perspective. By combining advanced methods in data enhancement and feature extraction, comprehensive experiments were conducted on metric learning methods using two datasets, revealing the relationships between loss functions, deep feature space, and metric learning.
PATTERN RECOGNITION
(2021)
Article
Computer Science, Information Systems
Guofeng Zou, Guixia Fu, Xiang Peng, Yue Liu, Mingliang Gao, Zheng Liu
Summary: Person re-identification is a challenging research issue in computer vision, with feature extraction and distance metric being critical technologies. While feature extraction has been well summarized, there is a lack of systematic analysis on distance metric methods. Effective and reliable distance metric is crucial for improving the accuracy of person re-identification.
MULTIMEDIA TOOLS AND APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Yixiu Liu, Yunzhou Zhang, Bir Bhanu, Sonya Coleman, Dermot Kerr
Summary: This paper introduces a multi-level cross-view consistent feature learning framework for person re-identification. The framework extracts multi-level features, conducts dictionary learning and metric learning to achieve accuracy in person re-identification.
Article
Computer Science, Artificial Intelligence
Mang Ye, Jianbing Shen, Gaojie Lin, Tao Xiang, Ling Shao, Steven C. H. Hoi
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
Maria Jose Gomez-Silva, Arturo de la Escalera, Jose Maria Armingol
Summary: The automatic re-identification of individuals across video surveillance cameras is challenging due to the learning of discriminative features and distance metrics affected by appearance variations. This article focuses on finding discriminative descriptors to reflect appearance differences independently of acquisition point variations through Mahalanobis distance learning in a Deep Neural Re-Identification model.
INTEGRATED COMPUTER-AIDED ENGINEERING
(2021)
Review
Computer Science, Artificial Intelligence
Xiaoting Wu, Xiaoyi Feng, Xiaochun Cao, Xin Xu, Dewen Hu, Miguel Bordallo Lopez, Li Liu
Summary: This paper provides a comprehensive review of the problem of Facial Kinship Verification (FKV), covering various aspects such as problem definition, challenges, applications, benchmark datasets, taxonomy of methods, and state-of-the-art performance. The paper also identifies gaps in current research and suggests potential future research directions.
INTERNATIONAL JOURNAL OF COMPUTER VISION
(2022)
Article
Mathematics
Zining Wang, Jiawei Chen, Junlin Hu
Summary: In pattern recognition, it is desirable to utilize the information from multiple views to complement each other. However, existing methods are mostly developed for single-view representation and are not suitable for multi-view data. In this paper, we propose a multi-view cosine similarity learning approach to efficiently utilize multi-view data for face verification, achieving superior performance in fine-grained face verification and kinship verification tasks.
Article
Computer Science, Artificial Intelligence
Nianchang Huang, Jianan Liu, Yunqi Miao, Qiang Zhang, Jungong Han
Summary: This paper provides a comprehensive review of the visible-infrared cross-modality person re-identification (VI-ReID) approaches, including the importance, definition, and challenges of VI-ReID. The motivations and methodologies of existing VI-ReID methods are analyzed in detail. A comprehensive taxonomy of state-of-the-art VI-ReID models is provided. Widely used datasets and evaluation metrics are also discussed. The limitations of current methods are pointed out based on comprehensive comparisons. Lastly, the challenges and future research trends in this field are outlined.
INFORMATION FUSION
(2023)
Article
Computer Science, Artificial Intelligence
Shilin Zhang, Cong Lin, Siming Ma
Summary: This paper introduces a novel large margin metric learning scheme for vehicle re-identification, aiming to maximize the distance margins among different vehicles; The method utilizes different loss functions and sampling methods in two stages of training process to improve performance.
Article
Computer Science, Theory & Methods
Jianyang Gu, Weihua Chen, Hao Luo, Fan Wang, Hao Li, Wei Jiang, Weijie Mao
Summary: In this work, a Multi-view Evolutionary Training (MET) method is proposed to effectively reduce noises in clustering results for person re-identification (Re-ID) tasks. The method includes a Multi-view Diffusion (MvD) module to improve clustering accuracy and an Evolutionary Local Refinement (ELR) module to enhance temporal consistency.
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
(2022)
Article
Computer Science, Artificial Intelligence
Zhangqiang Ming, Min Zhu, Xiangkun Wang, Jiamin Zhu, Junlong Cheng, Chengrui Gao, Yong Yang, Xiaoyong Wei
Summary: This paper introduces the research progress in person re-identification (Re-ID) field in recent years, categorizes deep learning-based methods, and discusses the challenges and future research directions in this field.
IMAGE AND VISION COMPUTING
(2022)
Article
Engineering, Electrical & Electronic
Zongheng Huang, Botao He, Bo Yang, Changxin Gao, Nong Sang
Summary: This paper proposes a Norm-Aware Margin Assignment (NAMA) scheme to dynamically adjust the weight of each sample during training, which improves the robustness of feature embedding by assigning larger margins to more recognizable samples. A margin re-balance mechanism is introduced to align the expectation of learned margins to a pre-defined value.
IEEE SIGNAL PROCESSING LETTERS
(2022)
Article
Computer Science, Information Systems
Xin Xu, Xin Yuan, Zheng Wang, Kai Zhang, Ruimin Hu
Summary: This article proposes a new differentiable retrieval-sort loss (DRSL) to optimize re-ID models. The DRSL maintains inter-class distance distribution and preserves intra-class similarity structure. Experimental results demonstrate that the proposed method improves the performance of re-ID models.
ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Haoxin Li, Wei-Shi Zheng, Jianguo Zhang, Haifeng Hu, Jiwen Lu, Jian-Huang Lai
Summary: This study proposes a weakly supervised model for egocentric action recognition, which automatically localizes interactors and establishes explicit relation models for recognition without using annotations or prior knowledge. Extensive experiments on egocentric video datasets demonstrate the effectiveness of the proposed method.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Wenzhao Zheng, Jiwen Lu, Jie Zhou
Summary: This paper proposes a deep metric learning method called DML-DC, which utilizes adaptively generated dynamic constraints for image retrieval and clustering. The method employs a learnable constraint generator to produce dynamic constraints and trains the metric towards better generalization. It formulates the deep metric learning objective under a proxy collection, pair sampling, tuple construction, and tuple weighting paradigm.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Ziwei Wang, Han Xiao, Yueqi Duan, Jie Zhou, Jiwen Lu
Summary: In this paper, we propose a GraphBit method for learning unsupervised deep binary descriptors to efficiently represent images. The method reduces the uncertainty of binary codes by maximizing the mutual information with input and related bits, allowing reliable binarization of ambiguous bits. Additionally, a differentiable search method called GraphBit+ is introduced to mine bitwise interaction in continuous space, reducing the computational cost of reinforcement learning. To address the issue of inaccurate instructions from fixed bitwise interaction, the unsupervised binary descriptor learning method D-GraphBit is proposed, which utilizes a graph convolutional network to reason the optimal bitwise interaction for each input sample.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Robotics
Sichao Huang, Ziwei Wang, Jie Zhou, Jiwen Lu
Summary: Object packing by autonomous robots is a significant challenge in warehouses and logistics industry. This paper proposes a deep hierarchical reinforcement learning approach to simultaneously plan the packing sequence and placement for irregular objects. The approach utilizes two networks, a top manager network to infer the packing sequence and a bottom worker network to predict the placement position and orientation, which are trained hierarchically in a self-supervised Q-Learning framework.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2023)
Article
Computer Science, Artificial Intelligence
Mantang Guo, Junhui Hou, Jing Jin, Hui Liu, Huanqiang Zeng, Jiwen Lu
Summary: This paper proposes a content-aware warping method that adaptsively learns the interpolation weights for pixels from their contextual information via a lightweight neural network. Based on this learnable warping module, a new end-to-end learning-based framework is proposed for novel view synthesis, which includes two additional modules to address occlusion and spatial correlation issues. Experimental results demonstrate that the proposed method outperforms state-of-the-art methods both quantitatively and visually.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Yi Wei, Shaohui Liu, Jie Zhou, Jiwen Lu
Summary: In this work, a new multi-view depth estimation method called NerfingMVS is presented, which combines conventional reconstruction and learning-based priors with neural radiance fields (NeRF). It directly optimizes over implicit volumes, eliminating the need for pixel matching in indoor scenes. The key is using learning-based priors to guide the optimization process of NeRF. The proposed method achieves state-of-the-art performances and improves rendering quality on both seen and novel views.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Yongming Rao, Wenliang Zhao, Zheng Zhu, Jie Zhou, Jiwen Lu
Summary: We present GFNet, a conceptually simple yet computationally efficient architecture that learns long-term spatial dependencies in the frequency domain. GFNet outperforms Transformer-based models and CNNs in terms of efficiency, generalization ability, and robustness. We provide a series of isotropic and hierarchical models based on GFNet design.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Hui Li, Tianyang Xu, Xiao-Jun Wu, Jiwen Lu, Josef Kittler
Summary: Deep learning based fusion methods have achieved promising performance in image fusion tasks due to the importance of network architecture. However, designing fusion networks is still a challenging task. In this paper, the fusion task is mathematically formulated and a connection between the optimal solution and network architecture is established. This leads to the proposal of a lightweight fusion network based on a learnable representation approach.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Yongming Rao, Zuyan Liu, Wenliang Zhao, Jie Zhou, Jiwen Lu
Summary: In this paper, a new approach for model acceleration by exploiting spatial sparsity in visual data is presented. A dynamic token sparsification framework is proposed, which prunes redundant tokens progressively and dynamically based on the input to accelerate vision Transformers. The framework extends to hierarchical models and more complex dense prediction tasks, offering a new and more effective dimension for model acceleration. Promising results are achieved on various architectures and visual tasks, demonstrating the effectiveness of the proposed framework.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Automation & Control Systems
Sheng Huang, Jingkai Lin, Luwen Huangfu, Yun Xing, Junlin Hu, Daniel Dajun Zeng
Summary: Facial image-based kinship verification is a rapidly growing field in computer vision and biometrics. This study proposes a novel deep learning model called AWk-TMN that leverages high-order cross-pair features to enhance the performance of kinship verification.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)
Article
Computer Science, Artificial Intelligence
Bingyao Yu, Xiu Li, Wanhua Li, Jie Zhou, Jiwen Lu
Summary: In this paper, a discrepancy-aware meta-learning approach for zero-shot face manipulation detection is proposed. The approach aims to learn a discriminative model that maximizes generalization to unseen face manipulation attacks with the guidance of the discrepancy map. Unlike existing methods, the detection of face manipulation is defined as a zero-shot problem, where algorithmic solutions are presented for known face manipulation attacks. The learning process is formulated as meta-learning and zero-shot face manipulation tasks are generated to learn diversified attack meta-knowledge. The discrepancy map is utilized to focus the model on generalized optimization directions during meta-learning, and a center loss is incorporated to better guide the model in exploring more effective meta-knowledge. Experimental results on widely used face manipulation datasets demonstrate competitive performance under the zero-shot setting.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2023)
Article
Computer Science, Information Systems
Bin Fan, Yuzhu Yang, Wensen Feng, Fuchao Wu, Jiwen Lu, Hongmin Liu
Summary: This paper proposes an adversarial learning based solution to extract robust local features and descriptions across day-night images. By training a discriminator to distinguish day and night images and adjusting the feature extraction network to fool the discriminator, the network can extract domain invariant keypoints and descriptors. Compared to existing methods, this approach only requires additional easily captured night images to improve the domain invariance of learned features.
IEEE TRANSACTIONS ON MULTIMEDIA
(2023)
Article
Computer Science, Artificial Intelligence
Ziwei Wang, Changyuan Wang, Xiuwei Xu, Jie Zhou, Jiwen Lu
Summary: In this article, the authors propose Quantformer, a type of extremely low-precision vision transformers for efficient inference. They address the limitations of conventional network quantization methods by considering the properties of transformer architectures and implementing capacity-aware distribution and group-wise discretization strategies. Experimental results show that Quantformer outperforms state-of-the-art methods in image classification and object detection across various vision transformer architectures. The authors also integrate Quantformer with mixed-precision quantization to further enhance performance.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Computer Science, Theory & Methods
Yongjie Duan, Jianjiang Feng, Jiwen Lu, Jie Zhou
Summary: In this study, a fusion of voting strategy and deep network is proposed to estimate fingerprint center and direction. Experimental results show that this approach can achieve consistent fingerprint pose estimations, improve performance of fingerprint indexing and verification, and be robust to different sensing technologies and impression types.
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
(2023)
Article
Computer Science, Artificial Intelligence
Shuai Shen, Wanhua Li, Zheng Zhu, Jie Zhou, Jiwen Lu
Summary: This paper proposes a new face clustering method, called STructure-AwaRe Face Clustering (STAR-FC), which addresses the dilemma of large-scale training and efficient inference by designing a structure-preserving subgraph sampling strategy and a novel hierarchical GCN training paradigm. During inference, the STAR-FC performs efficient full-graph clustering with two steps: graph parsing and graph refinement, and introduces the concept of node intimacy to mine the local structural information. The experimental results demonstrate that this method achieves superior performance and efficiency.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)