Article
Computer Science, Artificial Intelligence
Guanjie Huang, Xiangyu Luo, Shaowei Wang, Tianlong Gu, Kaile Su
Summary: Chinese character recognition is a challenging task due to the variety and complexity of the characters. To tackle this problem, researchers propose a novel network model called HCRN, which can recognize unseen Chinese characters by training only part of the radicals. Experimental results demonstrate that HCRN achieves high accuracy in recognizing both printed and handwritten characters.
PATTERN RECOGNITION
(2022)
Article
Computer Science, Artificial Intelligence
Guo-Feng Luo, Da-Han Wang, Xia Du, Hua-Yi Yin, Xu-Yao Zhang, Shunzhi Zhu
Summary: Zero-shot Chinese character recognition (ZSCCR) is an important research topic that aims to recognize unseen Chinese characters. This paper proposes a new method called self-information of radicals (SIR) to measure the importance of radicals in Chinese character recognition, and demonstrates its effectiveness and high extensibility through comprehensive experiments.
PATTERN RECOGNITION
(2023)
Article
Computer Science, Information Systems
Zhili Qin, Han Wang, Cobbinah Bernard Mawuli, Wei Han, Rui Zhang, Qinli Yang, Junming Shao
Summary: This paper proposes an improvement for attention mechanisms in few-shot learning algorithms. By splitting the image into patches and adding a new patch dimension, the proposed method achieves consistent improvement on multiple real-world datasets and reduces the number of attention module parameters.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Information Systems
Haoxing Chen, Huaxiong Li, Yaohui Li, Chunlin Chen
Summary: In this paper, a novel neural network architecture called sparse spatial transformers (SSFormers) is proposed, which can find task-relevant features and suppress task-irrelevant features. By dividing images into patches and finding spatial correspondence, the model retains contextual information and achieves precise classification.
SCIENCE CHINA-INFORMATION SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Zhe Wang, Pingchuan Ma, Ziqiu Chi, Dongdong Li, Hai Yang, Wenli Du
Summary: The purpose of few-shot learning is to achieve good generalization with limited samples. We propose the MAMD framework that uses multiple attention mechanisms for feature extraction and mutual learning for aggregating features. The method combines distributed learning and achieves expected results in few-shot learning benchmarks.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Jun Zhou, Qiujie Lv, Calvin Yu-Chian Chen
Summary: This paper proposes a dynamic concept-aware network (DCAN) that efficiently encodes task-specific structural concepts and adaptively dynamic alignment. Through dynamic prototype task awareness and cross-correlation dynamic alignment, DCAN outperforms existing methods on several FSL classification benchmarks.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Shuang Wu, Mohan Kankanhalli, Anthony K. H. Tung
Summary: This research is inspired by how humans recognize novel objects based on prior knowledge in addition to visual information. The study introduces a novel framework called Sup-Net to address the few-shot learning problem. By extracting knowledge and learning superclass features, the framework improves classification accuracy and has been validated through experiments.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2022)
Article
Engineering, Electrical & Electronic
Jiaoyan Chen, Yuxia Geng, Zhuo Chen, Jeff Z. Z. Pan, Yuan He, Wen Zhang, Ian Horrocks, Huajun Chen
Summary: Machine learning (ML), especially deep neural networks, has achieved great success, but the reliance on labeled samples is a challenge. Researchers have been investigating ML with sample shortage and many studies utilize auxiliary information, such as knowledge graph (KG), to reduce reliance on labeled samples. In this survey, over 90 articles about KG-aware research for zero-shot learning (ZSL) and few-shot learning (FSL) are reviewed comprehensively, covering construction methods and different paradigms of KG-aware methods. Various applications and evaluation resources are presented, along with discussions on challenges and open problems.
PROCEEDINGS OF THE IEEE
(2023)
Article
Computer Science, Information Systems
Nagwa Elaraby, Sherif Barakat, Amira Rezk
Summary: This paper introduces a deep metric learning method, the Siamese architecture, for few/zero-shot learning in the task of handwritten character recognition (HCR). By utilizing transfer learning and contrastive loss, the proposed network achieves high recognition accuracy and reduces training time compared to traditional Siamese CNNs.
CMC-COMPUTERS MATERIALS & CONTINUA
(2023)
Article
Computer Science, Information Systems
Jun He, Richang Hong, Xueliang Liu, Mingliang Xu, Qianru Sun
Summary: Few-shot classification focuses on adapting deep learning models to new classes with limited samples. We propose a method called DCAP, which improves feature embedding quality using Dense Classification and Attentive Pooling. Experimental results demonstrate that the proposed approach outperforms other methods in multiple few-shot settings.
ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
De Cheng, Gerong Wang, Bo Wang, Qiang Zhang, Jungong Han, Dingwen Zhang
Summary: Zero-shot learning aims to recognize unseen image semantics based on the training of data with seen semantics. This paper proposes a hybrid routing transformer model called HRT, which uses active attention and static routing to align visual features with attributes and generate class label predictions. Experimental results demonstrate the effectiveness of the proposed method.
PATTERN RECOGNITION
(2023)
Article
Computer Science, Artificial Intelligence
Yunxiao Qin, Weiguo Zhang, Chenxu Zhao, Zezheng Wang, Xiangyu Zhu, Jingping Shi, Guojun Qi, Zhen Lei
Summary: This study introduces a novel meta-learning approach that leverages prior-knowledge and attention mechanism to reduce the few-shot cognition burden of meta-learners, while also addressing the Task-Over-Fitting issue.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Environmental Sciences
Xin Zhao, Xiaoling Lv, Jinlei Cai, Jiayi Guo, Yueting Zhang, Xiaolan Qiu, Yirong Wu
Summary: This paper proposes an instance-aware transformer (IAT) model for few-shot synthetic aperture radar automatic target recognition (SAR-ATR). The IAT model effectively utilizes the power of all instances by constructing attention maps and aligns the features of query and support images using shared cross-transformers. Experimental results demonstrate the superiority of the proposed method.
Article
Computer Science, Artificial Intelligence
Heda Song, Bowen Deng, Michael Pound, Ender Ozcan, Isaac Triguero
Summary: Few-shot learning is a challenging problem in computer vision that aims to learn new visual concepts from very limited data. The proposed fusion spatial attention approach addresses the uncertainty introduced by the small training set through spatial attention in both image and embedding spaces. This method has proven effective and competitive performance on widely used few-shot learning benchmarks.
INFORMATION FUSION
(2022)
Article
Computer Science, Theory & Methods
Yisheng Song, Ting Wang, Puyu Cai, Subrota K. Mondal, Jyoti Prakash Sahoo
Summary: This survey investigates the latest advances in few-shot learning and provides a fair comparison of the strengths and weaknesses of existing work. By elaborating and contrasting relevant concepts, the prior knowledge is extracted and summarized in the form of a pyramid. In-depth analysis and discussions are presented for each subsection, highlighting the important application of FSL in computer vision.
ACM COMPUTING SURVEYS
(2023)
Article
Computer Science, Artificial Intelligence
Guang-Hai Liu, Zuo-Yong Li, Jing-Yu Yang, David Zhang
Summary: This article introduces a novel image retrieval method that improves retrieval performance by using sublimated deep features. The method incorporates orientation-selective features and color perceptual features, effectively mimicking these mechanisms to provide a more discriminating representation.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Fengguang Peng, Zihan Ding, Ziming Chen, Gang Wang, Tianrui Hui, Si Liu, Hang Shi
Summary: RGB-Thermal (RGB-T) semantic segmentation is an emerging task that aims to improve the robustness of segmentation methods under extreme imaging conditions by using thermal infrared modality. The challenges of foreground-background distinguishment and complementary information mining are addressed by proposing a cross modulation process with two collaborative components. Experimental results show that the proposed method achieves state-of-the-art performances on current RGB-T segmentation benchmarks.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Baihong Han, Xiaoyan Jiang, Zhijun Fang, Hamido Fujita, Yongbin Gao
Summary: This paper proposes a novel automatic prompt generation method called F-SCP, which focuses on generating accurate prompts for low-accuracy classes and similar classes. Experimental results show that our approach outperforms state-of-the-art methods on six multi-domain datasets.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Huikai Liu, Ao Zhang, Wenqian Zhu, Bin Fu, Bingjian Ding, Shengwu Xiong
Summary: Adverse weather conditions present challenges for computer vision tasks, and image de-weathering is an important component of image restoration. This paper proposes a multi-patch skip-forward structure and a Residual Deformable Convolutional module to improve feature extraction and pixel-wise reconstruction.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Oliver M. Crook, Mihai Cucuringu, Tim Hurst, Carola-Bibiane Schonlieb, Matthew Thorpe, Konstantinos C. Zygalakis
Summary: The transportation LP distance (TLP) is a generalization of the Wasserstein WP distance that can be applied directly to color or multi-channelled images, as well as multivariate time-series. TLP interprets signals as functions, while WP interprets signals as measures. Although both distances are powerful tools in modeling data with spatial or temporal perturbations, their computational cost can be prohibitively high for moderate pattern recognition tasks. The linear Wasserstein distance offers a method for projecting signals into a Euclidean space, and in this study, we propose linear versions of the TLP distance (LTLP) that show significant improvement over the linear WP distance in signal processing tasks while being several orders of magnitude faster to compute than the TLP distance.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Haitao Tian, Shiru Qu, Pierre Payeur
Summary: This paper proposes a method of target-dependent classifier, which optimizes the joint hypothesis of domain adaptation into a target-dependent hypothesis that better fits with the target domain clusters through an unsupervised fine-tuning strategy and the concept of meta-learning. Experimental results demonstrate that this method outperforms existing techniques in synthetic-to-real adaptation and cross-city adaptation benchmarks.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Qingsen Yan, Axi Niu, Chaoqun Wang, Wei Dong, Marcin Wozniak, Yanning Zhang
Summary: Deep learning-based methods have achieved remarkable results in the field of super-resolution. However, the limitation of paired training image sets has led researchers to explore self-supervised learning. However, the assumption of inaccurate downscaling kernel functions often leads to degraded results. To address this issue, this paper introduces KGSR, a kernel-guided network that trains both upscaling and downscaling networks to generate high-quality high-resolution images even without knowing the actual downscaling process.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Yifan Chen, Xuelong Li
Summary: Gait recognition is a popular technology for identification due to its ability to capture gait features over long distances without cooperation. However, current methods face challenges as they use a single network to extract both temporal and spatial features. To solve this problem, we propose a two-branch network that focuses on spatial and temporal feature extraction separately. By combining these features, we can effectively learn the spatio-temporal information of gait sequences.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Wei Shi, Wentao Zhang, Wei-shi Zheng, Ruixuan Wang
Summary: This article proposes a simple yet effective visualization framework called PAMI, which does not require detailed model structure and parameters to obtain visualization results. It can be applied to various prediction tasks with different model backbones and input formats.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Xiaobo Hu, Jianbo Su, Jun Zhang
Summary: This paper reviews the latest technologies in pattern recognition, highlighting their instabilities and failures in practical applications. From a control perspective, the significance of disturbance rejection in pattern recognition is discussed, and the existing problems are summarized. Finally, potential solutions related to the application of compensation on features are discussed to emphasize future research directions.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Andres Felipe Posada-Moreno, Nikita Surya, Sebastian Trimpe
Summary: Convolutional neural networks are widely used in critical systems, and explainable artificial intelligence has proposed methods for generating high-level explanations. However, these methods lack the ability to determine the location of concepts. To address this, we propose a novel method for automatic concept extraction and localization based on pixel-wise aggregations, and validate it using synthetic datasets.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Peng Bao, Jianian Li, Rong Yan, Zhongyi Liu
Summary: In this paper, a novel Dynamic Graph Contrastive Learning framework, DyGCL, is proposed to capture the temporal consistency in dynamic graphs and achieve good performance in node representation learning.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Kristian Schultz, Saptarshi Bej, Waldemar Hahn, Markus Wolfien, Prashant Srivastava, Olaf Wolkenhauer
Summary: Research indicates that deep generative models perform poorly compared to linear interpolation-based methods for synthetic data generation on small, imbalanced tabular datasets. To address this, a new approach called ConvGeN, combining convex space learning with deep generative models, has been proposed. ConvGeN improves imbalanced classification on small datasets while remaining competitive with existing linear interpolation methods.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Khondaker Tasrif Noor, Antonio Robles-Kelly
Summary: In this paper, the authors propose H-CapsNet, a capsule network designed for hierarchical image classification. The network effectively captures hierarchical relationships using dedicated capsules for each class hierarchy. A modified hinge loss is utilized to enforce consistency among the involved hierarchies. Additionally, a strategy for dynamically adjusting training parameters is presented to achieve better balance between the class hierarchies. Experimental results demonstrate that H-CapsNet outperforms competing hierarchical classification networks.
PATTERN RECOGNITION
(2024)
Article
Computer Science, Artificial Intelligence
Lei Liu, Guorun Li, Yuefeng Du, Xiaoyu Li, Xiuheng Wu, Zhi Qiao, Tianyi Wang
Summary: This study proposes a new agricultural image segmentation model called CS-Net, which uses Simple-Attention Block and Simpleformer to improve accuracy and inference speed, and addresses the issue of performance collapse of Transformers in agricultural image processing.
PATTERN RECOGNITION
(2024)