检索结果-内蒙古大学图书馆

BIG-MOE: BYPASS ISOLATED GATING MOE FOR GENERALIZED MULTIMODAL FACE ANTI-SPOOFING

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Ma, Yingjie Yu, Zitong Lin, Xun Xie, Weicheng Shen, Linlin Great Bay University China College of Computer Science and Software Engineering Shenzhen University China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University China Guangdong Provincial Key Laboratory of Intelligent Information Processing China

In the domain of facial recognition security, multimodal Face Anti-Spoofing (FAS) is essential for countering presentation attacks. However, existing technologies encounter challenges due to modality biases and imbalances, as well as domain shifts. Our research introduces a Mixture of Experts (MoE) model to address these issues effectively. We identified three limitations in traditional MoE approaches to multimodal FAS: (1) Coarse-grained experts’ inability to capture nuanced spoofing indicators;(2) Gated networks’ susceptibility to input noise affecting decision-making;(3) MoE’s sensitivity to prompt tokens leading to overfitting with conventional learning methods. To mitigate these, we propose the Bypass Isolated Gating MoE (BIG-MoE) framework, featuring: (1) Fine-grained experts for enhanced detection of subtle spoofing cues;(2) An isolation gating mechanism to counteract input noise;(3) A novel differential convolutional prompt bypass enriching the gating network with critical local features, thereby improving perceptual capabilities. Extensive experiments on four benchmark datasets demonstrate significant generalization performance improvement in multimodal FAS task. The code is released at https://***/murInJ/BIGMoE. Copyright © 2024, The Authors. All rights reserved.

关键词： HTTP

DEGSTalk: Decomposed Per-Embedding Gaussian Fields for Hair-Preserving Talking Face Synthesis

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Deng, Kaijun Zheng, Dezhi Xie, Jindong Wang, Jinbao Xie, Weicheng Shen, Linlin Song, Siyang Computer Vision Institute School of Computer Science and Software Engineering Shenzhen University China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University China Guangdong Provincial Key Laboratory of Intelligent Information Processing China Department of Computer Science University of Exeter United Kingdom

Accurately synthesizing talking face videos and capturing fine facial features for individuals with long hair presents a significant challenge. To tackle these challenges in existing methods, we propose a decomposed per-embedding Gaussian fields (DEGSTalk), a 3D Gaussian Splatting (3DGS)-based talking face synthesis method for generating realistic talking faces with long hairs. Our DEGSTalk employs Deformable Pre-Embedding Gaussian Fields, which dynamically adjust pre-embedding Gaussian primitives using implicit expression coefficients. This enables precise capture of dynamic facial regions and subtle expressions. Additionally, we propose a Dynamic Hair-Preserving Portrait Rendering technique to enhance the realism of long hair motions in the synthesized videos. Results show that DEGSTalk achieves improved realism and synthesis quality compared to existing approaches, particularly in handling complex facial dynamics and hair preservation. Our code will be publicly available at https://***/CVI-SZU/DEGSTalk. Copyright © 2024, The Authors. All rights reserved.

关键词： Gaussian distribution

Sample Reduction Using 1 -Norm Twin Bounded Support Vector Machine 2nd

学校读者我要写书评

暂无评论

Sample Reduction Using 1 -Norm Twin Bounded Support Vector M...

2nd International Conference on Neural Computing for Advanced Applications, NCAA 2021

作者： Zheng, Xiaohan Zhang, Li Yan, Leilei School of Computer Science and Technology Joint International Research Laboratory of Machine Learning and Neuromorphic Computing Soochow University Suzhou215006 China Provincial Key Laboratory for Computer Information Processing Technology Soochow University Suzhou215006 China

ISBN: (纸本)9789811651878

Twin support machine (TSVM) has a lower time complexity than support vector machine (SVM), but it has a poor ability to perform sample reduction. In order to improve the ability of TSVM to reduce sample, we propose an 1 -norm twin bounded support machine (1 -TBSVM) inspired by the sparsity of 1 -norm in feature space. The objective function of 1 -TBSVM contains the hinge loss and the 1 -norm terms, both which can induce sparsity. We solve the primal programming problems of 1 -TBSVM to prevent the disappearance of sparsity and avoid the situation that the inverse of matrix does not exist. Thus, 1 -TBSVM has a good sparsity, or a good ability to reduce sample. Experimental results on synthetic and UCI datasets indicate that 1 -TBSVM has a good ability to perform sample reduction and simultaneously enhances the classification performance. © 2021, Springer Nature Singapore Pte Ltd.

关键词： Support vector machines

FreqBlender: Enhancing DeepFake Detection by Blending Frequency Knowledge

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Li, Hanzhe Zhou, Jiaran Li, Yuezun Wu, Baoyuan Li, Bin Dong, Junyu School of Computer Science and Technology Ocean University of China China School of Data Science The Chinese University of Hong Kong Shenzhen China Guangdong Provincial Key Laboratory of Intelligent Information Processing Shenzhen University Shenzhen China

Generating synthetic fake faces, known as pseudo-fake faces, is an effective way to improve the generalization of DeepFake detection. Existing methods typically generate these faces by blending real or fake faces in spatial domain. While these methods have shown promise, they overlook the simulation of frequency distribution in pseudo-fake faces, limiting the learning of generic forgery traces in-depth. To address this, this paper introduces FreqBlender, a new method that can generate pseudo-fake faces by blending frequency knowledge. Concretely, we investigate the major frequency components and propose a Frequency Parsing Network to adaptively partition frequency components related to forgery traces. Then we blend this frequency knowledge from fake faces into real faces to generate pseudo-fake faces. Since there is no ground truth for frequency components, we describe a dedicated training strategy by leveraging the inner correlations among different frequency knowledge to instruct the learning process. Experimental results demonstrate the effectiveness of our method in enhancing DeepFake detection, making it a potential plug-and-play strategy for other methods. Copyright © 2024, The Authors. All rights reserved.

关键词： Fake detection

Brainnetformer: Decoding Brain Cognitive States with Spatial-Temporal Cross Attention

学校读者我要写书评

暂无评论

Brainnetformer: Decoding Brain Cognitive States with Spatial...

International Conference on Acoustics, Speech, and Signal processing (ICASSP)

作者： Leheng Sheng Wenhan Wang Zhiyi Shi Jichao Zhan Youyong Kong Chien-Shiung Wu College Southeast University China Jiangsu Provincial Joint International Research Laboratory of Medical Information Processing School of Computer Science and Engineering Southeast University China Key Laboratory of Computer Network and Information Integration Southeast University Ministry of Education Nanjing China

Learning about the cognitive state of the brain has always been a popular topic. Based on the fact that fluctuations of brain signals and functional connectome (FC) relate to specific human behaviors, deep learning based methods have shown promising results on the prediction of such behaviors by analyzing biological signals. Existing methods either model from static perspectives or apply spatial-temporal graph convolution to extract dynamic properties. However, the static information and dynamic information can reflect global brain activities and local brain activities respectively. Thus, we propose BrainNetFormer to incorporate both static and dynamic properties for human behavior prediction. To be specific, a spatial cross attention module and a temporal cross attention module are introduced for information fusion. In addition, since a specific behavior of subjects can be decomposed into a series of subtasks, we introduce a sub-task regularization loss to assist in training and empower the model to recognize subtasks at each moment. Experiments on the HCP-Task dataset demonstrate the superior performance of the proposed model.

关键词： Training Convolution Fuses Magnetic resonance imaging Predictive models Brain modeling Behavioral sciences

OCTAMamba: A State-Space Model Approach for Precision OCTA Vasculature Segmentation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zou, Shun Zhang, Zhuo Gao, Guangwei College of Artificial Intelligence Nanjing Agricultural University Nanjing China College of Computer Science and Technology National University of Defense Technology Changsha China Institute of Advanced Technology Nanjing University of Posts and Telecommunications Nanjing China Provincial Key Laboratory for Computer Information Processing Technology Soochow University Suzhou China

Optical Coherence Tomography Angiography (OCTA) is a crucial imaging technique for visualizing retinal vasculature and diagnosing eye diseases such as diabetic retinopathy and glaucoma. However, precise segmentation of OCTA vasculature remains challenging due to the multi-scale vessel structures and noise from poor image quality and eye lesions. In this study, we proposed OCTAMamba, a novel U-shaped network based on the Mamba architecture, designed to segment vasculature in OCTA accurately. OCTAMamba integrates a Quad Stream Efficient Mining Embedding Module for local feature extraction, a Multi-Scale Dilated Asymmetric Convolution Module to capture multi-scale vasculature, and a Focused Feature Recalibration Module to filter noise and highlight target areas. Our method achieves efficient global modeling and local feature extraction while maintaining linear complexity, making it suitable for low-computation medical applications. Extensive experiments on the OCTA 3M, OCTA 6M, and ROSSA datasets demonstrated that OCTAMamba outperforms state-of-the-art methods, providing a new reference for efficient OCTA segmentation. Code is available at https://***/zs1314/OCTAMamba Copyright © 2024, The Authors. All rights reserved.

关键词： Optical coherence tomography

FaceGCN: Structured Priors Inspired Graph Convolutional Networks for Blind Face Restoration

学校读者我要写书评

暂无评论

IEEE Transactions on Circuits and Systems for Video technology 2025年

作者： Yan, Weidan Shao, Wenze Zhang, Dengyin Xiao, Liang Nanjing University of Posts and Telecommunications School of Internet of Things Nanjing210003 China Nanjing University of Posts and Telecommunications Jiangsu Key Laboratory of Intelligent Information Processing and Communication Technology Nanjing210003 China Nanjing University of Science and Technology School of Computer Science and Engineering Nanjing210094 China

Facial image restoration has gained a tremendous progress since the increasing boom of the deep learning methods. Owing to its nature of strong ill-posedness, different categories of a-priori constraints have been harnessed or embedded in the existing deep architectures. While, as it turns to blind face restoration with more complicated degradations, the challenge becomes greater. In this paper, a further insightful step is taken by exploring the potentials of the graph convolutional networks (GCN) in conjunction with the structured priors for the blind problem. Specifically, a lightweight yet physically more intuitive model termed FaceGCN is proposed. On the one hand, a dynamic generator of facial adjacency matrices is constructed assisted by two self-supervised losses, allowing a sparse, accurate, and adaptive construction of case-specific face graphs with facial feature components as nodes. On the other hand, to model well the joint local-nonlocal correlations among various facial feature components, a kind of novel strip-attention GCN modules is correspondingly developed by splitting facial feature maps into intra- and inter-strips in both horizontal and vertical orientations, respectively. Extensive experimental results show that FaceGCN has achieved comparable or even superior performance to state-of-the-art methods, yet at a considerably less computational cost. © 1991-2012 IEEE.

关键词： Convolution

Ssd Optimization Model Based on Shallow Feature Fusion

学校读者我要写书评

暂无评论

SSRN

SSRN 2022年

作者： Yang, Zhe Bu, Ziyu Liu, Chunping Provincial Key Laboratory for Computer Information Processing Technology Suzhou215006 China Department of Computer Science and Technology Soochow University Suzhou215006 China Provincial Key Laboratory for Intelligent Engineering in Big Data Suzhou215006 China

Object detection has been an important research branch in the field of computer vision. The single-shot-detection (SSD) is an object detection model based on deep learning, which can achieve a good balance between the detection accuracy and the detection speed, but has the problem of poor recognition accuracy for small objects. To address this limitation , this paper improves the structure of the SSD feature pyramid and upsamples the shallow feature map with small object information and fuses it with the upper feature map, thus enhancing the ability of the shallow feature map to represent detailed information. In this way, not only the overall detection accuracy of the SSD is improved, but also a relatively high detection speed is maintained. The proposed model is verified by experiments on two common datasets, the Pascal VOC and MS COCO datasets. On the PASCAL VOC07+12, MS COCO14, and VOC07+12 + COCO datasets, the improved model achieves the mean average precision values of 80.1% (+3.3% compared with the conventional model), 49.9% (+6.8%), and 82.1% (+3.0%), respectively. Meanwhile, the proposed model can achieve the detection speed of 42.2 frames per second. © 2022, The Authors. All rights reserved.

关键词： Object detection

Theme-Based Literature Analysis in the Field of Transportation 16th

学校读者我要写书评

暂无评论

Theme-Based Literature Analysis in the Field of Transportati...

16th International Conference on Intelligent information Hiding and Multimedia Signal processing, IIH-MSP 2020 in conjunction with the 13th International Conference on Frontiers of information technology, Applications and Tools, FITAT 2020

作者： Zhang, Sifan Lu, Hao Zhang, Fuquan Zhu, Yifan Wang, Rongrong Computer School Beijing Institute of Technology Beijing10089 China Fujian Provincial Key Laboratory of Information Processing and Intelligent Control Min-Jiang University Fuzhou350121 China

ISBN: (纸本)9789813367562

Transportation is an important research field of engineering science, and its related research covers multiple fields and links in policies, methods, technologies, and applications. In order to sort out the research context and development status of the transportation field and provide scientific and technological information support for scientific researchers in the field, this article analyzes the literature published in SCI journals in the transportation field from 2008 to 2018 through the topic analysis method. The analysis results show that the current main research results in the field of transportation have formed different clusters around keywords such as cognitive radio, electric vehicles, optimization and simulation, and intelligent transportation systems, and the research development process of different countries is different. For example, the research results of China in recent years have gone from energy efficiency, automobile emissions to intelligent transportation, while the United States has been focusing on the research of intelligent transportation systems. Over time, China has gradually focused on the cutting-edge technology in the field of transportation, and its research results have greatly promoted the development of research in the field of transportation. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021.

关键词： Energy efficiency