检索结果-内蒙古大学图书馆

FLIP-80M: 80 Million Visual-Linguistic Pairs for Facial Language-Image Pre-Training 24

学校读者我要写书评

暂无评论

FLIP-80M: 80 Million Visual-Linguistic Pairs for Facial Lang...

32nd ACM International Conference on Multimedia, MM 2024

作者： Li, Yudong Hou, Xianxu Dezhi, Zheng Shen, Linlin Zhao, Zhe School of Computer Science and Software Engineering Shenzhen University Shenzhen China Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen China School of AI and Advanced Computing Xi'an Jiaotong-Liverpool University Shenzhen China Guangdong Provincial Key Laboratory of Intelligent Information Processing Shenzhen University Shenzhen China Tencent AI Lab Beijing China

ISBN: (纸本)9798400706868

While significant progress has been made in multi-modal learning driven by large-scale image-text datasets, there is still a noticeable gap in the availability of such datasets within the facial domain. To facilitate and advance the field of facial representation learning, we present FLIP-80M, a large-scale visual-linguistic dataset comprising over 80 million face images paired with text descriptions. FLIP-80M is constructed by leveraging the large openly available image-text-pair dataset LAION-5B and a mixed-method approach to filter face-related pairs from both visual and linguistic perspectives. Our curation process involves face detection, face caption classification, text de-noising, and synthesis-based image augmentation. As a result, FLIP-80M stands as the largest face-text dataset to date. To evaluate the potential of our dataset, we fine-tune the CLIP model using the proposed FLIP-80M, to create FLIP (Facial Language-Image Pretraining) and assess its representation capabilities across various downstream tasks. Our experiments demonstrate that our FLIP model achieves state-of-the-art results in a range of face analysis tasks, including face parsing, face alignment, and face attribute classification. The dataset and models are available at https://***/ydli-ai/FLIP. © 2024 ACM.

关键词： Face recognition

No-Reference Image Quality Assessment Based on Active Reasoning Module 6

学校读者我要写书评

暂无评论

No-Reference Image Quality Assessment Based on Active Reason...

6th IEEE International Conference on Electronic information and Communication Technology, ICEICT 2023

作者： Qi, Junwei Deng, Yuhao Wang, Qingchun Yang, Zhen Li, Yingsong College of Information and Communication Engineering Harbin Engineering University Harbin150001 China Shanghai Yanding Information Technology Co. Ltd. Shanghai201210 China College of Computer Science and Technology Harbin Engineering University Harbin150001 China Anhui University Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education China

ISBN: (纸本)9798350399059

We present a no-reference image-quality - assessment algorithm based on active reasoning module. This algorithm has three modules: the feature extraction module, the active reasoning module, and the quality assessment module. The active reasoning module incorporates the generator component of the generative adversarial network and enhances its structure with the Res2Net architecture. By integrating this module into the backbone feature extraction network, we improve the receptive field of each convolutional layer, enabling the network to capture information at different scales of the image. To preserve the texture information of the image, we input the gradient map of the distorted image, the distorted image itself, and the image features generated by the generator into the quality assessment module, which has a multi-feature regression networks. This module establishes a mapping model from image-feature to image-quality scores. We conducted experimental analysis of this algorithm on three widely used public datasets, confirming the excellent performance and superiorities of the presented algorithm. The results validate the effectiveness of the designed algorithm and its ability to assess image quality accurately. © 2023 IEEE.

关键词： Image quality

An Entity Alignment Algorithm Based on Ontology Enhancement and Multi-Feature Fusion

学校读者我要写书评

暂无评论

An Entity Alignment Algorithm Based on Ontology Enhancement ...

Frontiers Technology of information and Computer (ICFTIC), IEEE International Conference on

作者： Hao Song Yuxia Lei Kangli Zi Fuyuan Quan Tianhang Zhang Qi Li School of Computer Science Qufu Normal University Rizhao Shandong China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China

ISBN: (数字)9798331541750

ISBN: (纸本)9798331541767

Entity alignment plays a crucial role in the integration of knowledge graphs. However, current knowledge graphs often suffer from inconsistent construction standards, structural heterogeneity, and imprecise semantic representation of entities, which can lead to suboptimal embedding of structural information. Moreover, most existing entity alignment methods rely heavily on structural information, which limits alignment efficiency. To address these challenges, this paper presents Onto-IMF, an ontology-enhanced entity alignment method based on multi-feature fusion. This approach aims to enhance the semantic representation of entities by incorporating ontology information, thereby reducing structural discrepancies among heterogeneous graphs and improving the embedding of structural information. The multi-feature fusion entity alignment leverages diverse semantic features, including relations, attributes, and entity names, to achieve more accurate alignment. Experimental results on various DBP15K subsets indicate that the proposed method achieves Hits@l scores of 86.6%, 93.0%, and 97.7%, respectively, representing an average improvement of 7.9 % over the best-performing baseline models.

关键词： Accuracy Semantics Knowledge graphs Contrastive learning Ontologies Data models Standards

An Efficient Saddle Search Method for Ordered Phase Transitions Involving Translational Invariance

学校读者我要写书评

暂无评论

SSRN

SSRN 2024年

作者： Cui, Gang Jiang, Kai Zhou, Tiejun Hunan Key Laboratory for Computation and Simulation in Science and Engineering Key Laboratory of Intelligent Computing and Information Processing Ministry of Education School of Mathematics and Computational Science Xiangtan University Hunan Xiangtan411105 China

In this work, we propose an efficient nullspace-preserving saddle search (NPSS) method for a class of phase transitions involving translational invariance, where the critical states are often degenerate. The NPSS method includes two stages, escaping from the basin and searching for the index-1 generalized saddle point. The NPSS method climbs upward from the generalized local minimum in segments to overcome the challenges of degeneracy. In each segment, an effective ascent direction is ensured by keeping this direction orthogonal to the nullspace of the initial state in this segment. This method can escape the basin quickly and converge to the transition states efficiently. We apply the NPSS method to the phase transitions between crystals, and between crystal and quasicrystal, based on the Landau-Brazovskii and Lifshitz-Petrich free energy functionals. Numerical results show a good performance of the NPSS method. © 2024, The Authors. All rights reserved.

关键词： Free energy

A Hybrid Iterative Neural Solver Based on Spectral Analysis for Parametric PDEs

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Cui, Chen Jiang, Kai Liu, Yun Shu, Shi Hunan Key Laboratory for Computation and Simulation in Science and Engineering Key Laboratory of Intelligent Computing and Information Processing of Ministry of Education School of Mathematics and Computational Science Xiangtan University Hunan Xiangtan411105 China

Recently, deep learning-based hybrid iterative methods (DL-HIM) have emerged as a promising approach for designing fast neural solvers to tackle large-scale sparse linear systems. DL-HIM combine the smoothing effect of simple iterative methods with the spectral bias of neural networks, which allows them to effectively eliminate both high-frequency and low-frequency error components. However, their efficiency may decrease if simple iterative methods can not provide effective smoothing, making it difficult for the neural network to learn mid-frequency and high-frequency components. This paper first conducts a convergence analysis for general DL-HIM from a spectral viewpoint, concluding that under reasonable assumptions, DL-HIM exhibit a convergence rate independent of grid size h and physical parameters µ. To meet these assumptions, we design a neural network from an eigen perspective, focusing on learning the eigenvalues and eigenvectors corresponding to error components that simple iterative methods struggle to eliminate. Specifically, the eigenvalues are learned by a meta subnet, while the eigenvectors are approximated using Fourier modes with a transition matrix provided by another meta subnet. The resulting DL-HIM, termed the Fourier Neural Solver (FNS), can be trained to achieve a convergence rate independent of grid size and PDE parameters by designing a loss function that ensures the neural network complements the smoothing effect of Jacobi iterative methods. We verify the performance of FNS on six types of linear parametric *** Codes 65F10, 65F08, 68T07, 65N22, 90C06 Copyright © 2024, The Authors. All rights reserved.

关键词： Eigenvalues and eigenfunctions

Self-Training with Contrastive Learning for Adversarial Domain Adaptation

学校读者我要写书评

暂无评论

Self-Training with Contrastive Learning for Adversarial Doma...

International Joint Conference on Neural Networks (IJCNN)

作者： Shanshan Wang Minbin Hu Xingyi Zhang Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Institutes of Physical Science and Information Technology Anhui University Hefei China School of Computer Science and Technology Anhui University Hefei China

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

Unsupervised domain adaptation (UDA) aims to transfer the knowledge learned from the labeled source domain to the unlabeled target domain. Traditional methods often focus on minimizing the distribution gap between feature spaces of the two domains to achieve domain-invariant representations. However, this approach may not fully leverage the inherent class-specific information and could adversely affect decision boundaries in the target domain. In this paper, we introduce an adversarial contrastive self-training framework. The model integrates both intra-class and inter-class domain differences for better alignment between the source and target domains. Initially, we augment target domain samples using Fourier transformations, obtaining reliable samples and pseudo-labels through self-training. We then employ intra-domain contrastive learning on the reliable samples and pseudo-labels, alongside labeled source domain samples, to achieve intra-domain compactness. Finally, we utilize adversarial learning to achieve global alignment. This comprehensive approach ensures improved adaptation performance while also addressing the limitations of previous methods.

关键词： Adaptation models Neural networks Contrastive learning Adversarial machine learning Reliability

Momentum-Accelerated Richardson(m) and Their Multilevel Neural Solvers

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Wang, Zhen Liu, Yun Cui, Chen Shu, Shi Hunan Key Laboratory for Computation and Simulation in Science and Engineering Key Laboratory of Intelligent Computing and Information Processing of Ministry of Education School of Mathematics and Computational Science Xiangtan University Hunan Xiangtan411105 China

Recently, designing neural solvers for large-scale linear systems of equations has emerged as a promising approach in scientific and engineering computing. This paper first introduce the Richardson(m) neural solver by employing a meta network to predict the weights of the long-step Richardson iterative method. Next, by incorporating momentum and preconditioning techniques, we further enhance convergence. Numerical experiments on anisotropic second-order elliptic equations demonstrate that these new solvers achieve faster convergence and lower computational complexity compared to both the Chebyshev iterative method with optimal weights and the Chebyshev semi-iteration method. To address the strong dependence of the aforementioned single-level neural solvers on PDE parameters and grid size, we integrate them with two multilevel neural solvers developed in recent years. Using alternating optimization techniques, we construct Richardson(m)FNS for anisotropic equations and NAG-Richardson(m)-WANS for the Helmholtz equation. Numerical experiments show that these two multilevel neural solvers effectively overcome the drawback of single-level methods, providing better robustness and computational efficiency. Copyright © 2024, The Authors. All rights reserved.

关键词： Iterative methods

Convergent analysis of algebraic multigrid method with data-driven parameter learning for non-selfadjoint elliptic problems

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhang, Juan Luo, Junyue Key Laboratory of Intelligent Computing and Information Processing of Ministry of Education Hunan Key Laboratory for Computation and Simulation in Science and Engineering School of Mathematics and Computational Science Xiangtan University Hunan Xiangtan411105 China

In this paper, we apply the practical GADI-HS iteration as a smoother in algebraic multigrid (AMG) method for solving second-order non-selfadjoint elliptic problem. Additionally, we prove the convergence of the derived algorithm and introduce a data-driven parameter learing method called Gaussian process regression (GPR) to predict optimal parameters. Numerical experimental results show that using GPR to predict parameters can save a significant amount of time cost and approach the optimal parameters accurately. © 2024, CC BY.

关键词： Algebra

Conservative nonconforming virtual element method for stationary incompressible magnetohydrodynamics

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Dong, Xiaojing Huang, Yunqing Wang, Tianwen Hunan Key Laboratory for Computation and Simulation in Science and Engineering Key Laboratory of Intelligent Computing & Information Processing of Ministry of Education School of Mathematics and Computational Science Xiangtan University Hunan Xiangtan411105 China

In this paper, we propose a conservative nonconforming virtual element method for the full stationary incompressible magnetohydrodynamics model. We leverage the virtual element satisfactory divergence-free property to ensure mass conservation for the velocity field. The condition of the well-posedness of the proposed method, as well as the stability are derived. We establish optimal error estimates in the discrete energy norm for both the velocity and magnetic field. Furthermore, by employing a new technique, we obtain the optimal error estimates in L2-norm without any additional conditions. Finally, numerical experiments are presented to validate the theoretical analysis. In the implementation process, we adopt the effective Oseen iteration to handle the nonlinear system. © 2024, CC BY-NC-ND.

关键词： Magnetohydrodynamics