Article
Geochemistry & Geophysics
Zhi He, Dan He, Xinyuan Li, Jiani Xu
Summary: Recent studies have shown that unsupervised deep learning methods can improve the performance of video satellite super-resolution. This research proposes a single video-based approach that does not rely on any prior high-resolution or low-resolution pairs, making it highly practical.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Computer Science, Information Systems
Junyu Gao, Xiaoshan Yang, Yingying Zhang, Changsheng Xu
Summary: This research proposes to learn relation-aware hard assignments for selecting key video clips in an unsupervised manner by utilizing clip-clip relations. By constructing an assignment-learning graph and optimizing the whole framework, the approach achieves favorable performance on popular benchmarks.
IEEE TRANSACTIONS ON MULTIMEDIA
(2021)
Article
Engineering, Electrical & Electronic
Li Tao, Xueting Wang, Toshihiko Yamasaki
Summary: This paper proposes a self-supervised contrastive learning method for learning video feature representations. By introducing intra-negative samples and utilizing strong data augmentations, the proposed method achieves significant improvements in video retrieval and video recognition tasks.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Mingjie Li, Po-Yao Huang, Xiaojun Chang, Junjie Hu, Yi Yang, Alex Hauptmann
Summary: The main challenge in unsupervised machine translation is to associate source-target sentences in the latent space. Various unsupervised multi-modal machine translation models have been proposed to improve performance by employing visual contents in natural images. However, current state-of-the-art methods are sensitive to spurious correlations as they do not explicitly model object relations.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Engineering, Electrical & Electronic
Qinghai Zheng, Jihua Zhu, Zhongyu Li
Summary: This paper focuses on the challenging problem of unsupervised multi-view representation learning and introduces a novel method, Collaborative Unsupervised Multi-view Representation Learning (CUMRL), which benefits from the high-order view correlations of multi-view data by introducing a collaborative learning strategy. Experiments demonstrate the effectiveness and competitiveness of the multi-view representation achieved by the proposed method for different learning tasks.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Engineering, Electrical & Electronic
Xuejun Huang, Jinshan Ding, Qinghua Guo
Summary: This article introduces an unsupervised image registration approach for video SAR moving target detection, utilizing a cascade of two convolutional neural networks. The method compensates for local deformations between images by predicting a displacement field, thereby preventing false alarms in moving target detection.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2021)
Article
Computer Science, Artificial Intelligence
Wei Huang, Shuzhou Sun, Xiao Lin, Ping Li, Lei Zhu, Jihong Wang, C. L. Philip Chen, Bin Sheng
Summary: This article proposes a feature-matching-based uncertainty method to alleviate the data bias issue by resampling selected uncertainty data. To meet the requirement of no additional costs, the authors specially designed an unsupervised fusion feature matching (UFFM), and redesigned classic uncertainty methods for more complex visual tasks. Experimental results demonstrate that the proposed method outperforms other similar techniques.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Huihui Song, Wenjie Xu, Dong Liu, Bo Liu, Qingshan Liu, Dimitris N. Metaxas
Summary: The paper introduces a multi-stage feature fusion network for video super-resolution task, which fuses temporally aligned features of supporting frames and spatial features of reference frame at different stages to enhance features from low-resolution to high-resolution. Experimental results demonstrate that the proposed method achieves state-of-the-art performance on VSR task.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Geochemistry & Geophysics
Huilin Xu, Wei He, Liangpei Zhang, Hongyan Zhang
Summary: This paper proposes an unsupervised deep semantic feature learning network (S3FN) for hyperspectral images (HSIs), which learns spectral-spatial features from a high-level semantic perspective and aligns these features using a contrastive loss function. Experimental results show that S3FN can achieve promising classification results with lower time cost compared to other state-of-the-art unsupervised FE methods.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Engineering, Electrical & Electronic
Muheng Li, Lei Chen, Jiwen Lu, Jianjiang Feng, Jie Zhou
Summary: In this paper, a weakly-supervised approach called Order-Constrained Representation Learning (OCRL) is proposed to predict future actions from instructional videos by observing incomplete steps of actions. The approach learns video representations from step order-rearranged trimmed video clips and integrates shared semantic information between step order and task semantics. The results show improvements in comparison to conventional prediction methods.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Shixuan Zhou, Peng Song, Yanwei Yu, Wenming Zheng
Summary: Multi-view unsupervised feature selection (MUFS) is a popular research topic that aims to select a compact representative feature subset from multi-view data. However, most existing MUFS methods overlook the discriminative ability of multi-view data. This paper proposes a novel MUFS method called structural regularization based discriminative multi-view unsupervised feature selection (SDFS), which addresses these limitations and outperforms state-of-the-art MUFS models.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Engineering, Electrical & Electronic
Yifeng Zhou, Xing Xu, Fumin Shen, Xiaofeng Zhu, Heng Tao Shen
Summary: This paper presents a novel model called Flow Edge-based Motion-Attentive Network (FEM-Net) for addressing the problem of unsupervised video object segmentation. Experimental results show that the proposed FEM-Net outperforms existing methods on two challenging public benchmarks.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Xuanzhao Wang, Zhengping Che, Bo Jiang, Ning Xiao, Ke Yang, Jian Tang, Jieping Ye, Jingyu Wang, Qi Qi
Summary: This article proposes a novel video anomaly detection method based on frame prediction, with better performance and noise tolerance loss, which outperforms existing state-of-the-art methods as confirmed by experiments.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Computer Science, Information Systems
Elham Ravanbakhsh, Yongqing Liang, J. Ramanujam, Xin Li
Summary: This paper reviews representation learning for videos, discussing recent spatio-temporal feature learning methods and comparing their advantages and disadvantages for general video analysis. It emphasizes the importance of building effective video features in computer vision tasks and summarizes the effectiveness and challenges of existing spatial and temporal features.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Zhuping Wang, Xinke Dai, Zhanyu Guo, Chao Huang, Hao Zhang
Summary: In this article, an unsupervised monocular depth and camera motion estimation framework is proposed using unlabeled monocular videos. The method utilizes photometric loss, channelwise attention mechanism, and spatialwise attention mechanism to achieve state-of-the-art results on the KITTI benchmark and great generalization performance on the Make3D dataset.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
C. L. Philip Chen, Chun-Yang Zhang, Long Chen, Min Gan
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2015)
Article
Green & Sustainable Science & Technology
Chun-Yang Zhang, C. L. Philip Chen, Min Gan, Long Chen
IEEE TRANSACTIONS ON SUSTAINABLE ENERGY
(2015)
Article
Automation & Control Systems
Wenxi Liu, Chun-Yang Zhang, Genggeng Liu, Yaru Su, Neil N. Xiong
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2019)
Article
Automation & Control Systems
Chun-Yang Zhang, Zhi-Liang Yao, Hong-Yu Yao, Feng Huang, C. L. Philip Chen
Summary: This article introduces a dynamic graph representation learning model called DynGNN, which embeds an RNN into a graph neural network to better capture the temporal dynamics and topological correlations of graphs and achieves significant improvements in multiple tasks.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2023)
Article
Engineering, Multidisciplinary
Chun-Yang Zhang, Hai-Chun Cai, C. L. Philip Chen, Yue-Na Lin, Wu-Peng Fang
Summary: Contrastive learning is widely used in graph representation learning, but most existing models ignore the diversity of node attributes and network topologies. To address this, we propose a novel graph representation learning model called GRAM, which uses an adaptive metric to generate appropriate similarity scores for node pairs based on the significance of each dimension in their embedding vectors and data distribution. Experimental results demonstrate that GRAM is highly competitive in multiple tasks.
IEEE TRANSACTIONS ON NETWORK SCIENCE AND ENGINEERING
(2023)
Article
Computer Science, Cybernetics
Chun-Yang Zhang, Wu-Peng Fang, Hai-Chun Cai, C. L. Philip Chen, Yue-Na Lin
Summary: This article proposes a Sparse Graph Transformer model (SGTC) for graph representation learning. It removes redundant topological information using centrality measures, learns node representations in a contrastive manner, and introduces a sparse attention mechanism to capture structural features of graphs. Experimental results demonstrate that SGTC outperforms existing baselines in terms of performance.
IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Chun-Yang Zhang, Yue-Na Lin, C. L. Philip Chen, Hong-Yu Yao, Hai-Chun Cai, Wu-Peng Fang
Summary: In recent years, there has been a significant increase in graph representation learning. This article proposes an unsupervised fuzzy representation learning model that improves the expressiveness of graphs and networks by making crisp representations fuzzy. The model utilizes fuzzy logic to fully explore feature-level uncertainties and generate fuzzy representations that are more expressive than crisp ones.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Huibin Lin, Hai-Tao Fu, Chun-Yang Zhang, C. L. Philip Chen
Summary: This study proposes a new robust unsupervised person re-identification model that solves the problem of misclassification caused by clustering operations, achieving a better exploration of implicit relationships between samples.
INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS
(2023)
Article
Automation & Control Systems
Hong-Yu Yao, Yuan-Long Yu, Chun-Yang Zhang, Yue-Na Lin, Shang-Jia Li
Summary: This paper proposes a new dynamic graph representation learning model called FuzzyDGL, which incorporates fuzzy representation learning to handle uncertainties in dynamic graphs. Experimental results show that FuzzyDGL has strong competitiveness and generalization in tasks like link prediction and node classification.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2023)
Article
Engineering, Electrical & Electronic
Feng-Jie Li, Chun-Yang Zhang, C. L. Philip Chen
Summary: Accurate prediction of vehicle trajectories is crucial for autonomous vehicles. Existing models have limitations in capturing both spatial and temporal dependencies. Therefore, a novel dynamic graph neural network is proposed to jointly extract spatial-temporal features and consider low-order and high-order dynamics collaboratively.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Computer Science, Artificial Intelligence
Shuang Feng, C. L. Philip Chen, Chun-Yang Zhang
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2020)