检索结果-内蒙古大学图书馆

Hadamard Attention Recurrent Transformer: A Strong Baseline for Stereo Matching Transformer

学校读者我要写书评

暂无评论

arxiv 2025年

作者： Chen, Ziyang Zhang, Yongjun Li, Wenting Wang, Bingshu Wu, Yabo Zhao, Yong Chen, C.L. Philip College of Computer Science The State Key Laboratory of Public Big Data Guizhou University Guiyang550025 China School of Information Engineering Guizhou University of Commerce Guiyang550021 China School of Software Northwestern Polytechnical University Xi’an710129 China Key Laboratory of Integrated Microsystems Shenzhen Graduate School Peking University Shenzhen518055 China School of Computer Science and Engineering South China University of Technology Guangzhou510641 China

In light of the advancements in transformer technology, extant research posits the construction of stereo transformers as a potential solution to the binocular stereo matching challenge. However, constrained by the low-rank bottleneck and quadratic complexity of attention mechanisms, stereo transformers still fail to demonstrate sufficient nonlinear expressiveness within a reasonable inference time. The lack of focus on key homonymous points renders the representations of such methods vulnerable to challenging conditions, including reflections and weak textures. Furthermore, a slow computing speed is not conducive to the application. To overcome these difficulties, we present the Hadamard Attention Recurrent Stereo Transformer (HART) that incorporates the following components: 1) For faster inference, we present a Hadamard product paradigm for the attention mechanism, achieving linear computational complexity. 2) We designed a Dense Attention Kernel (DAK) to amplify the differences between relevant and irrelevant feature responses. This allows HART to focus on important details. DAK also converts zero elements to non-zero elements to mitigate the reduced expressiveness caused by the low-rank bottleneck. 3) To compensate for the spatial and channel interaction missing in the Hadamard product, we propose MKOI to capture both global and local information through the interleaving of large and small kernel convolutions. Experimental results demonstrate the effectiveness of our HART. In reflective area, HART ranked 1st on the KITTI 2012 benchmark among all published methods at the time of submission. Code is available at https://***/ ZYangChen/HART. Copyright © 2025, The Authors. All rights reserved.

关键词： Hadamard transforms

Joint Optimization of Charging Time and Resource Allocation in Wireless Power Transfer Assisted Federated Learning

学校读者我要写书评

暂无评论

Joint Optimization of Charging Time and Resource Allocation ...

IEEE Conference on computer Communications Workshops, INFOCOM Wksps

作者： Jingjiao Wang Huan Zhou Liang Zhao Deng Meng X. Shouzhi Hubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering College of Computer and Information Technology China Three Gorges University Yichang China School of Computer Science Northwestern Polytechnical University Xi'an China College of Economics and Management China Three Gorges University Yichang China

ISBN: (数字)9798350384475

ISBN: (纸本)9798350384482

As a new distributed machine learning methodology, Federated Learning (FL) allows mobile devices (MDs) to collaboratively train a global model without sharing their raw data in a privacy-preserving manner. However, it is a great challenge to schedule each MD and allocate various resources reasonably. This paper studies the joint optimization of computing resources used by MDs for FL training, the number of local iterations as well as WPT duration of each MD in a Wireless Power Transfer (WPT) assisted FL system, with the goal of maximizing the total utility of all MDs in the entire FL training process. Furthermore, we analyze the problem by using the Karush-Kuhn-Tucker (KKT) conditions and Lagrange dual method, and propose an improved Lagrangian subgradient method to solve this problem. Finally, extensive simulation experiments are conducted under various scenarios to verify the effectiveness of the proposed algorithm. The results show that our proposed algorithm has better performance in terms of the total utility of all MDs compared with other benchmark methods.

关键词： Training Schedules Federated learning Distributed databases Wireless power transfer Benchmark testing Mobile handsets

Stackelberg Game-based and Broker-assisted Computation Offloading in MEC Networks

学校读者我要写书评

暂无评论

Stackelberg Game-based and Broker-assisted Computation Offlo...

IEEE Conference on computer Communications Workshops, INFOCOM Wksps

作者： Deng Meng Jianmeng Guo Liang Zhao Huan Zhou Shouzhi Xu College of Economics and Management China Three Gorges University Yichang China Hubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering College of Computer and Information Technology China Three Gorges University Yichang China School of Computer Science Northwestern Polytechnical University Xi'an China

ISBN: (数字)9798350384475

ISBN: (纸本)9798350384482

Mobile Edge Computing (MEC) can effectively speed up data processing and improve Quality of Service (QoS) by offloading Mobile Users' (MUs') tasks to nearby Edge Servers (ESs). However, due to the individual rationality of entities (i.e., ESs and MUs) in MEC networks, they may be reluctant to participate in the computation offloading process without reasonable resource pricing or compensation. To address the challenge, we propose a Two-stage Stackelberg game-based computation Offloading and Resource Pricing mechanism (TORP). Specifically, we first introduce a broker in MEC, which rents computation resources from ESs and provides services to MUs. Next, we formulate the interactions among the broker, MUs, and ESs as a two-stage Stackelberg game, aiming to maximize their respective utilities. Then, we propose a Gradient-Ascent-Based Dynamic Iterative Search Algorithm (GADISA) and an Alternating Iteration-Based Resource Pricing and Task Offloading Algorithm (AIPOA) to solve the optimization problem. Finally, simulations show that TORP greatly outperforms other benchmarks in improving the utilities of three entities.

关键词： Heuristic algorithms Simulation Pricing Games Quality of service Benchmark testing Search problems

Research on Image Caption Based on Multiple Word Embedding Representations 3

学校读者我要写书评

暂无评论

Research on Image Caption Based on Multiple Word Embedding R...

3rd International Conference on Natural Language Processing, ICNLP 2021

作者： Lin, Zhen-xian Qu, Jia-xin Yang, xiao-Bao Hou, Zhi-Qiang Ding, Chen Xi'An University of Posts and Telecommunications School of Science Xi'an China Xi'An University of Posts and Telecommunications School of Telecommunication and Information Engineering Xi'an China Northwestern Polytechnical University School of Computer Science and Technology Xi'an China Xi'An University of Posts and Telecommunications Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Shaanxi Xi'an China

ISBN: (纸本)9781665414111

Word embedding representation has been a research hot spot in the field of computer vision and natural language processing. The semantic quality of representation severely affects the effects of machine translation and image caption. The key factors affecting the quality of word embedding representation are capture of contextual semantics and solution of the problem of polysemy. In this paper, the principles of the current mainstream word embedding representation methods were explored, such as sparse representation (One-Hot), static representation (Word2Vec, GloVe), dynamic representation (ELMo, BERT) representation and based on position information (***-Position Embedding). The impact of those word embedding representations on image caption task was researched attentively. And those representations methods were compared experimentally using the MSCOCO data based on the same image caption model. The pros and cons of these methods were judged according to the accuracy and richness of the extracted label data. Experimental results show that the BERT model representation with dynamic bidirectional semantics has the best performance in image caption tasks, but it requires longer training time. The *** representation has the second best image semantic description effect, and its training speed is the fastest. © 2021 IEEE.

关键词： Embeddings

Look Inside for More: Internal Spatial Modality Perception for 3D Anomaly Detection

学校读者我要写书评

暂无评论

arxiv 2024年

作者： Liang, Hanzhe xie, Guoyang Hou, Chengbin Wang, Bingshu Gao, Can Wang, Jinbao College of Computer Science and Software Engineering Shenzhen University Shenzhen China Shenzhen Audencia Financial Technology Institute Shenzhen University Shenzhen China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen China Department of Intelligent Manufacturing CATL Ningde China School of Computing and Artificial Intelligence Fuyao University of Science and Technology Fuzhou China School of Software Northwestern Polytechnical University Xi’an China Guangdong Provincial Key Laboratory of Intelligent Information Processing Shenzhen China

3D anomaly detection has recently become a significant focus in computer vision. Several advanced methods have achieved satisfying anomaly detection performance. However, they typically concentrate on the external structure of 3D samples and struggle to leverage the internal information embedded within samples. Inspired by the basic intuition of why not look inside for more, we introduce a straightforward method named Internal Spatial Modality Perception (ISMP) to explore the feature representation from internal views fully. Specifically, our proposed ISMP consists of a critical perception module, Spatial Insight Engine (SIE), which abstracts complex internal information of point clouds into essential global features. Besides, to better align structural information with point data, we propose an enhanced key point feature extraction module for amplifying spatial structure feature representation. Simultaneously, a novel feature filtering module is incorporated to reduce noise and redundant features for further aligning precise spatial structure. Extensive experiments validate the effectiveness of our proposed method, achieving object-level and pixel-level AUROC improvements of 4.2% and 13.1%, respectively, on the Real3D-AD benchmarks. Note that the strong generalization ability of SIE has been theoretically proven and is verified in both classification and segmentation tasks. Copyright © 2024, The Authors. All rights reserved.

关键词： Feature extraction

FaceBench: A Multi-View Multi-Level Facial Attribute VQA dataset for Benchmarking Face Perception MLLMs

学校读者我要写书评

暂无评论

arxiv 2025年

作者： Wang, xiaoqin Ma, Xusen Hou, xianxu Ding, Meidan Li, Yudong Chen, Junliang Chen, Wenting Peng, xiaoyang Shen, Linlin Computer Vision Institute 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 AIAC Xi’an Jiaotong-Liverpool University China Tsinghua University China The Hong Kong Polytechnic University Hong Kong City University of Hong Kong Hong Kong Sun Yat-sen University China

Multimodal large language models (MLLMs) have demonstrated remarkable capabilities in various tasks. However, effectively evaluating these MLLMs on face perception remains largely unexplored. To address this gap, we introduce FaceBench, a dataset featuring hierarchical multi-view and multi-level attributes specifically designed to assess the comprehensive face perception abilities of MLLMs. Initially, we construct a hierarchical facial attribute structure, which encompasses five views with up to three levels of attributes, totaling over 210 attributes and 700 attribute values. Based on the structure, the proposed FaceBench consists of 49,919 visual question-answering (VQA) pairs for evaluation and 23,841 pairs for fine-tuning. Moreover, we further develop a robust face perception MLLM baseline, Face-LLaVA, by training with our proposed face VQA data. Extensive experiments on various mainstream MLLMs and Face-LLaVA are conducted to test their face perception ability, with results also compared against human performance. The results reveal that, the existing MLLMs are far from satisfactory in understanding the fine-grained facial attributes, while our Face-LLaVA significantly outperforms existing open-source models with a small amount of training data and is comparable to commercial ones like GPT-4o and Gemini. The dataset will be released at https://***/CVI-SZU/FaceBench. Copyright © 2025, The Authors. All rights reserved.

关键词：

Robin: A Novel Method to Produce Robust Interpreters for Deep Learning-Based Code Classifiers

学校读者我要写书评

暂无评论

Robin: A Novel Method to Produce Robust Interpreters for Dee...

IEEE International Conference on Automated Software Engineering (ASE)

作者： Zhen Li Ruqian Zhang Deqing Zou Ning Wang Yating Li Shouhuai Xu Chen Chen Hai Jin School of Cyber Science and Engineering Huazhong University of Science and Technology Wuhan China Services Computing Technology and System Lab Hubei Key Laboratory of Distributed System Security Cluster and Grid Computing Lab National Engineering Research Center for Big Data Technology and System Hubei Engineering Research Center on Big Data Security Department of Computer Science University of Colorado Colorado Springs USA Center for Research in Computer Vision University of Central Florida USA School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China

Deep learning has been widely used in source code classification tasks, such as code classification according to their functionalities, code authorship attribution, and vulnerability detection. Unfortunately, the black-box nature of deep learning makes it hard to interpret and understand why a classifier (i.e., classification model) makes a particular prediction on a given example. This lack of interpretability (or explainability) might have hindered their adoption by practitioners because it is not clear when they should or should not trust a classifier's prediction. The lack of interpretability has motivated a number of studies in recent years. However, existing methods are neither robust nor able to cope with out-of-distribution examples. In this paper, we propose a novel method to produce Robust interpreters for a given deep learning-based code classifier; the method is dubbed Robin. The key idea behind Robin is a novel hybrid structure combining an interpreter and two approximators, while leveraging the ideas of adversarial training and data augmentation. Experimental results show that on average the interpreter produced by Robin achieves a 6.11% higher fidelity (evaluated on the classifier), 67.22% higher fidelity (evaluated on the approximator), and 15.87x higher robustness than that of the three existing interpreters we evaluated. Moreover, the interpreter is 47.31% less affected by out-of-distribution examples than that of LEMNA.

关键词：

Perceptive self-supervised learning network for noisy image watermark removal

学校读者我要写书评

暂无评论

arxiv 2024年

作者： Tian, Chunwei Zheng, Menghua Li, Bo Zhang, Yanning Zhang, Shichao Zhang, David School of Software Northwestern Polytechnical University Xi’an710129 China National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology Xi’an710129 China School of Electronics and Information Northwestern Polytechnical University Xi’an710129 China School of Computer Science Northwestern Polytechnical University National Engineering Laboratory for Integrated AeroSpace-Ground-Ocean Big Data Application Technology Xi’an710129 China Guangxi Key Lab of Multisource Information Mining & Security College of Computer Science & Engineering Guangxi Normal University Guilin541004 China Shenzhen518172 China Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen518172 China

Popular methods usually use a degradation model in a supervised way to learn a watermark removal model. However, it is true that reference images are difficult to obtain in the real world, as well as collected images by cameras suffer from noise. To overcome these drawbacks, we propose a perceptive self-supervised learning network for noisy image watermark removal (PSLNet) in this paper. PSLNet depends on a parallel network to remove noise and watermarks. The upper network uses task decomposition ideas to remove noise and watermarks in sequence. The lower network utilizes the degradation model idea to simultaneously remove noise and watermarks. Specifically, mentioned paired watermark images are obtained in a self-supervised way, and paired noisy images (i.e., noisy and reference images) are obtained in a supervised way. To enhance the clarity of obtained images, interacting two sub-networks and fusing obtained clean images are used to improve the effects of image watermark removal in terms of structural information and pixel enhancement. Taking into texture information account, a mixed loss uses obtained images and features to achieve a robust model of noisy image watermark removal. Comprehensive experiments show that our proposed method is very effective in comparison with popular convolutional neural networks (CNNs) for noisy image watermark removal. Codes can be obtained at https://***/hellloxiaotian/PSLNet. Copyright © 2024, The Authors. All rights reserved.

关键词： Image denoising

CrossDiff: Exploring Self-Supervised Representation of Pansharpening via Cross-Predictive Diffusion Model

学校读者我要写书评

暂无评论

arxiv 2024年

作者： xing, Yinghui Qu, Litao Zhang, Shizhou Zhang, Kai Zhang, Yanning Ran, Lingyan Zhang, xiuwei Yin, Hanlin The Shaanxi Provincial Key Laboratory of Speech and Image Information Processing The National Engineering Laboratory for Integrated Aerospace-GroundOcean Big Data Application Technology School of Computer Science Northwestern Polytechnical University Xi’an710072 China The Research Development Institute Northwestern Polytechnical University in Shenzhen Shenzhen518057 China

Fusion of a panchromatic (PAN) image and corresponding multispectral (MS) image is also known as pansharpening, which aims to combine abundant spatial details of PAN and spectral information of MS. Due to the absence of high-resolution MS images, available deep-learning-based methods usually follow the paradigm of training at reduced resolution and testing at both reduced and full resolution. When taking original MS and PAN images as inputs, they always obtain sub-optimal results due to the scale variation. In this paper, we propose to explore the self-supervised representation of pansharpening by designing a cross-predictive diffusion model, named CrossDiff. It has two-stage training. In the first stage, we introduce a cross-predictive pretext task to pre-train the UNet structure based on conditional DDPM, while in the second stage, the encoders of the UNets are frozen to directly extract spatial and spectral features from PAN and MS, and only the fusion head is trained to adapt for pansharpening task. Extensive experiments show the effectiveness and superiority of the proposed model compared with state-of-the-art supervised and unsupervised methods. Besides, the cross-sensor experiments also verify the generalization ability of proposed self-supervised representation learners for other satellite’s datasets. We will release our code for reproducibility. © 2024, CC BY.

关键词： Deep learning