Article
Computer Science, Software Engineering
Cong Yang, Bipin Indurkhya, John See, Marcin Grzegorzek
Summary: This article introduces a novel approach to automatically generate visually promising skeletons without manual tuning. The approach generates backbone and dense skeletons from shape input, extends backbone branches via skeleton grafting, and introduces potential functions to improve accuracy of skeleton-based matching. Experimental results show that the proposed potential functions effectively reduce the number of incorrect matches.
IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS
(2021)
Article
Computer Science, Artificial Intelligence
Abd Errahmane Kiouche, Hamida Seba, Karima Amrouche
Summary: This paper introduces an effective dissimilarity measure for geometric graphs representing shapes, which combines sparsification and node embedding methods to improve shape matching performance and reduce overall matching time. Experimental results demonstrate that the proposed approach outperforms existing methods in shape matching.
PATTERN RECOGNITION LETTERS
(2021)
Article
Computer Science, Artificial Intelligence
Xue Liu, Wei Wei, Xiangnan Feng, Xiaobo Cao, Dan Sun
Summary: The paper introduces a novel graph embedding algorithm named GraphCSC, which integrates skeleton information and component information of graphs into embeddings for classification. Experiments demonstrate that the algorithm outperforms state-of-the-art baselines in graph classification tasks.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Computer Science, Information Systems
Basheer Alwaely, Charith Abhayaratne
Summary: This article presents a new method for 2D/3D shape recognition based on graph spectral domain handcrafted features, which capture both the global outline and structural descriptions of the shape. The proposed method outperforms state-of-the-art studies with significant improvements in accuracy rates for 2D static hand gestures, 2D shapes, and 3D shapes.
ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Jiayi Ma, Xingyu Jiang, Aoxiang Fan, Junjun Jiang, Junchi Yan
Summary: Image matching is a fundamental task in various visual applications, and with the development of deep learning techniques, there has been an increasing number of methods proposed in this field. However, the challenge remains in choosing the suitable method for specific applications and designing image matching methods with superior performance. This comprehensive review and analysis provide insights into classical and latest techniques, and offer prospects for future development in image matching technologies.
INTERNATIONAL JOURNAL OF COMPUTER VISION
(2021)
Article
Computer Science, Artificial Intelligence
Narges Mirehi, Maryam Tahmasbi, Alireza Tavakoli Targhi
Summary: This paper introduces a graph-based method for shape recognition, which can extract features robust to noise, rotation, scale variation, and articulation, showing advantages in different datasets.
Article
Computer Science, Artificial Intelligence
Zhenyue Qin, Yang Liu, Pan Ji, Dongwoo Kim, Lei Wang, R. McKay, Saeed Anwar, Tom Gedeon
Summary: Skeleton sequences are effective for action recognition on edge devices due to their lightweight and compact nature. By incorporating third-order features in the form of angular encoding into modern architectures, recognition performance can be improved while reducing parameters and run time. This fusion approach has achieved new state-of-the-art accuracy in large benchmarks such as NTU60 and NTU120.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Computer Science, Software Engineering
Jiangce Chen, Horea T. Ilies, Caiwen Ding
Summary: This paper presents a shape analysis method for different geometric representations, using graph analysis tools and neural networks to achieve similarity, retrieval, and substructure matching for geometric models. The method shows comparable performance with state-of-the-art methods across different geometric representations.
COMPUTER-AIDED DESIGN
(2022)
Article
Chemistry, Multidisciplinary
Kai Hu, Yiwu Ding, Junlan Jin, Liguo Weng, Min Xia
Summary: This paper proposes a novel multi-scale time sampling module and a deep spatiotemporal feature extraction module to enhance the accuracy of human motion recognition network. Comparative experiments show that the proposed method achieves performance improvement on two datasets.
APPLIED SCIENCES-BASEL
(2022)
Article
Computer Science, Artificial Intelligence
Cunling Bian, Wei Feng, Liang Wan, Song Wang
Summary: This research explores the feasibility of using low-quality skeletons for action recognition and proposes a structural knowledge distillation scheme to minimize accuracy degradations and improve model robustness. The proposed scheme demonstrates effectiveness in various action recognition datasets.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Chemistry, Analytical
Meng Xu, Zhihuang Zhang, Yuanhao Gong, Stefan Poslad
Summary: In this paper, a novel camera pose regression framework is proposed, which utilizes global features with rotation consistency and local features with rotation invariance. Experimental results demonstrate that the proposed method achieves high accuracy in pose estimation and image matching tasks.
Article
Mathematics, Applied
Alberto Pepe, Joan Lasenby
Summary: Protein structure prediction involves predicting the folding of a protein from its amino acid sequence. This paper demonstrates the relevance of geometric algebra (GA) in instantiating orientational features for protein structure prediction. The authors propose two GA-based metrics which contain information on relative orientations of amino acid residues and show that these features yield comparable results to classical angle-based metrics in terms of accuracy of the predicted coordinates.
MATHEMATICAL METHODS IN THE APPLIED SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Liangliang Zhu, Xinwen Zhu, Xiurui Geng
Summary: In recent years, multi-graph matching has become a popular yet challenging task in graph theory. The cycle-consistency problem and the high time and space complexity problem are two major challenges in multi-graph matching. Pairwise-based methods have low complexity but require additional constraints for cycle-consistency, while tensor-based methods can avoid the cycle-consistency problem but have high complexity. This paper proposes a new multi-graph matching method by finding the equivalence between pairwise-based and tensor-based methods under specific circumstances, reducing complexity and improving performance.
PATTERN RECOGNITION
(2023)
Article
Statistics & Probability
Xiaoyang Guo, Aditi Basu Bal, Tom Needham, Anuj Srivastava
Summary: This paper presents a method for mathematically representing and statistically analyzing the shapes of arterial networks in the human brain. The study findings suggest that age has a clear, quantifiable effect on the shapes of these networks.
ANNALS OF APPLIED STATISTICS
(2022)
Article
Computer Science, Artificial Intelligence
Runzhong Wang, Junchi Yan, Xiaokang Yang
Summary: This study proposes an unsupervised framework for graph matching, which can match two or multiple graphs and handle graphs with a mixture of modes. The framework is trained by minimizing the discrepancy between a second-order classic solver and a first-order differentiable Sinkhorn net. Experimental results show that our method performs well in real-world applications such as natural image matching.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)