检索结果-内蒙古大学图书馆

BrainCLIP: Brain Representation via CLIP for Generic Natural visual Stimulus Decoding

IEEE Transactions on Medical Imaging 2025年 PP卷 PP页

作者： Ma, Yongqiang Liu, Yulong Chen, Liangjun Zhu, Guibo Chen, Badong Zheng, Nanning Xi'an Jiaotong University National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center of Visual Information and Applications Institute of Artificial Intelligence and Robotics Shaanxi Xi'an710049 China Institute of Automation Beijing China

Functional Magnetic Resonance Imaging (fMRI) presents challenges due to limited paired samples and low signal-to-noise ratios, particularly in tasks involving reconstructing natural images or decoding their semantic content. To address these challenges, we introduce BrainCLIP, an innovative fMRI-based brain decoding model. BrainCLIP leverages Contrastive Language-Image Pre-training's (CLIP) cross-modal generalization abilities to bridge brain activity, images, and text for the first time. Our experiments demonstrate CLIP's effectiveness in diverse brain decoding tasks, including zero-shot visual category decoding, fMRI-image/text alignment, and fMRI-to-image generation. The core objective of BrainCLIP is to train a mapping network that translates fMRI patterns into a unified CLIP embedding space, achieved through visual and textual supervision integration. Our experiments highlight that this approach significantly enhances performance in tasks such as fMRI-text alignment and fMRI-based image generation. Notably, BrainCLIP surpasses BraVL, a recent multi-modal method, in zero-shot visual category decoding. Moreover, BrainCLIP demonstrates strong capability in reconstructing visual stimuli with high semantic fidelity, competing favorably with state-of-the-art methods in capturing high-level semantic features during fMRI-based natural image reconstruction. © 1982-2012 IEEE.

关键词： Signal to noise ratio

来源：评论

学校读者我要写书评

暂无评论

Carrier recombination suppression and transport enhancement enable high-performance self-powered broadband Sb_(2)Se_(3) photodetectors

引用

InfoMat 2023年第4期5卷 60-76页

作者： Shuo Chen Yi Fu Muhammad Ishaq Chuanhao Li Donglou Ren Zhenghua Su Xvsheng Qiao Ping Fan Guangxing Liang Jiang Tang Shenzhen Key Laboratory of Advanced Thin Films and Applications Key Laboratory of Optoelectronic Devices and Systems of Ministry of Education and Guangdong ProvinceCollege of Physics and Optoelectronic EngineeringShenzhen UniversityShenzhenGuangdongthe People's Republic of China MOE Key Laboratory of New Processing Technology for Non-ferrous Metals and Materials Guangxi Key Laboratory of Processing for Non-ferrous Metals and Featured MaterialsGuangxi UniversityNanningGuangxithe People's Republic of China State Key Laboratory of Silicon Materials&School of Materials Science and Engineering Zhejiang UniversityHangzhouZhejiangthe People's Republic of China Wuhan National Laboratory for Optoelectronics(WNLO) School of Optical and Electronic InformationHuazhong University of Science and TechnologyWuhanHubeithe People's Republic of China

Antimony selenide(Sb_(2)Se_(3))is a promising candidate for photodetector applications boasting unique material benefits and remarkable optoelectronic *** high-performance self-powered Sb_(2)Se_(3)photodetector through a synergistic regulation of absorber layer and heterojunction interface demonstrates great potential and needs essential *** this study,an effective two-step thermodynamic/kinetic deposition technique containing sputtered and selenized Sb precursor is implemented to induce self-assembled growth of Sb_(2)Se_(3)light absorbing thin film with large crystal grains and desirable[hk1]orientation,presenting considerable thin-film photodetector ***,aluminum(Al^(3+))cation dopant is introduced to modify the optoelectronic properties of CdS buffer layer,and further optimize the Sb_(2)Se_(3)/CdS(Al)heterojunction interface *** to the suppressed carrier recombination and enhanced carrier transport kinetics,the champion Mo/Sb_(2)Se_(3)/CdS(Al)/ITO/Ag photodetector exhibits self-powered and broadband characteristics,accompanied by simultaneously high responsivity of 0.9 A W^(-1)(at 11 nW cm^(-2)),linear dynamic range of 120 dB,impressive ON/OFF switching ratio over 10^(6)and signal-to-noise ratio of 10^(9),record total noise determined realistic detectivity of 4.78×10^(12)Jones,and ultra-fast response speed with rise/decay time of 24/75 ns,representing the top level for Sb_(2)Se_(3)-based *** intriguing work opens up an avenue for its selfpowered broadband photodetector applications.

关键词： photoresponse recombination suppression Sb_(2)Se_(3) self-powered photodetector transport enhancement

来源：评论

学校读者我要写书评

暂无评论

AMA: An Analytical Approach to Maximizing the Efficiency of Deep Learning on Versal AI Engine

AMA: An Analytical Approach to Maximizing the Efficiency of ...

引用

International Conference on Field Programmable Logic and applications

作者： Xiaodong Deng Shijie Wang Tianyi Gao Jing Liu Longjun Liu Nanning Zheng National Key Laboratory of HumanMachine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications and Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University

ISBN: (数字)9798331530075

ISBN: (纸本)9798331530082

The traditional cache-based multi-core architecture represented by CUDA has been plagued by the “memory wall” problem, and a large number of applications represented by large language model inference are unable to meet the high computational intensity requirements. The Versal AI Engine architecture provides a variety of rich inter-core connections in the processor array, increasing many data reuse opportunities and potentially alleviating the “memory wall” issue. However, traditional parallel programming models cannot be directly applied to this architecture, and how to map computations to achieve high computational utilization becomes a new challenge. To address this, we propose AMA, a hierarchical performance analysis model built on the Versal AI Engine architecture, designed to maximize the efficiency of typical deep learning applications. Experiments show that AMA modeling is accurate and efficient. On the VCK190 platform, we achieved a matrix multiplication throughput of 5867.29 GFLOPS in fp32 and 88.55 TOPS in int8, and a convolution throughput of 99.6770 TOPS. In terms of energy efficiency, AMA achieved 142.68 GFLOPS/W in fp32 precision and 1.416 TOPS/W in int8 matrix multiplication. Compared to the current state-of-the-art methods, we achieved a $\mathbf{1 4. 9 9 \%}$ increase in throughput and a 22.92% increase in energy efficiency, providing new analytical performance model and practical guidance for efficient deep learning deployment on AI Engine.

关键词： Deep learning Analytical models Convolution Computational modeling Computer architecture Throughput Energy efficiency Performance analysis Computational efficiency Engines

来源：评论

学校读者我要写书评

暂无评论

Refiner: Fine-grained Cross-modal Concepts Refinement for Compositional Zero-Shot Learning

Refiner: Fine-grained Cross-modal Concepts Refinement for Co...

引用

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

作者： Xiao Zhang Haodong Jing Hui Chen Yongqiang Ma Nanning Zheng National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications and Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Recent Compositional Zero-Shot Learning (CZSL) methods increasingly adopt the pre-trained vision-language models to capture the contextual relations between image and text spaces. However, the single-class-token design from Transformer-based encoder inevitably captures contextual information from unrelated objects and background, thus hindering the modeling of fine-grained class-specific visual features. Suffering from cross-modal gap, prior methods also struggle to improve compositional recognition performance. To address these issues, we propose a fine-grained cross-modal concepts refinement framework, termed as Refiner, which comprises two pivotal components: (i) the fine-grained concepts refinement of image embeddings to capture state-object context within visual scenes, and (ii) the cross-modal information fusion to mitigate the modality gap. By leveraging learnable query vectors to capture region-specific semantic information pertinent to composition labels, our approach refines visual representations with fine-grained state-object context information. As for cross-modal information fusion, we construct a robust image-to-text mapping by aligning visual embeddings with states, objects, and compositions, respectively. Extensive experiments demonstrate that our Refiner achieves new state-of-the-art performance across all popular benchmarks in both closed- and open-world settings.

关键词： visualization Zero shot learning Semantics Benchmark testing Signal processing Transformers Vectors Acoustics Speech processing Context modeling

来源：评论

学校读者我要写书评

暂无评论

Two-Stream (2+1)D CNN Based on Frame Difference Attention for Driver Behavior Recognition 10

Two-Stream (2+1)D CNN Based on Frame Difference Attention fo...

引用

10th International Conference on Dependable Systems and Their applications, DSA 2023

作者： Zou, Yang Du, Shaoyi Han, Hongcheng Liu, Yuying Tian, Zhiqiang Xi'an Jiaotong University National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications Institute of Artificial Intelligence and Robotics Shaanxi Xi'an China Shunan Academy of Artificial Intelligence Zhejiang Ningbo China Xi'an Jiaotong University School of Software Engineering Shaanxi Xi'an China

ISBN: (纸本)9798350304770

Driving behavior recognition is a key technology in human-machine cooperative driving, which plays an important role in the construction of an intelligent transportation environment. Due to the similarity of drivers ' actions and backgrounds in driving environments, it's challenging for existing behavior recognition methods to recognize behaviors in driving environments precisely. To solve these problems, we propose the two-stream (2+1)D CNN with Frame Difference Attention Module. First, we use two streams to process the information captured from the face camera and road camera and use the Squeeze and Excitation attention module to fuse the stream information. Second, we propose the Frame Difference Attention Module to enhance the changes between adjacent frames so that the subtle actions in driving environments can be captured easier. We test our network on the Brain4Cars dataset and have a good performance on it. © 2023 IEEE.

关键词： Cameras

来源：评论

学校读者我要写书评

暂无评论

SparallaxMPC: High-Speed Model Predictive Control Solver with High Sparsity and Parallelism

SparallaxMPC: High-Speed Model Predictive Control Solver wit...

引用

Chinese Automation Congress (CAC)

作者： ZhengXing Lei Xiaodong Deng Longjun Liu Caigui Jiang National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications and Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University

ISBN: (数字)9798350368604

ISBN: (纸本)9798350368611

Model Predictive Control (MPC) is one of the most popular methods used in control problem. However, MPC is computationally demanding with limited use of sparse and parallel implementations. We present SparallaxMPC, a high-speed MPC solver with parallel and sparse representation. Our approach is based on sparse matrix representation and leverages the structure of the MPC problem for efficiency. We demonstrate the effectiveness of SparallaxMPC by benchmarking it against the state-of-the-art solver CasADi, achieving more than a 3x acceleration.

关键词： Sparse approximation Single instruction multiple data Graphics processing units Kinematics Programming Benchmark testing Predictive models Prediction algorithms Sparse matrices Predictive control

来源：评论

学校读者我要写书评

暂无评论

POAQL: A Partially Observable Altruistic Q-Learning Method for Cooperative Multi-Agent Reinforcement Learning

POAQL: A Partially Observable Altruistic Q-Learning Method f...

引用

IEEE International Conference on Robotics and Automation (ICRA)

作者： Lesong Tao Miao Kang Jinpeng Dong Songyi Zhang Ke Ye Shitao Chen Nanning Zheng National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications and Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University

ISBN: (数字)9798350384574

ISBN: (纸本)9798350384581

Multi-Agent Path Finding (MAPF) is an important issue in multi-agent cooperation. Many studies apply MultiAgent Reinforcement Learning (MARL) to solve MAPF in partially observable settings. The objective of cooperative MARL is to maximize the cumulative team reward. Nevertheless, in partially observable settings, the team reward is misleading due to unpredictable factors from the behavior and state of unobserved agents. To address this issue, we propose a Partially Observable Altruistic Q-learning (POAQL) method. POAQL considers the cumulative reward of the observed subteam instead of the whole team, where Altruistic Q-learning plays an important role in learning the subteam action value. In addition, we design a new conflict resolution without additional guidance to emphasize the cooperative nature of MARL frameworks. Experimental results show that POAQL outperforms existing reinforcement learning methods in terms of efficiency and performance.

关键词： Q-learning Costs Robotics and automation

来源：评论

学校读者我要写书评

暂无评论

LuminAIRe: illumination-aware conditional image repainting for lighting-realistic generation 23

LuminAIRe: illumination-aware conditional image repainting f...

引用

Proceedings of the 37th International Conference on Neural information processing Systems

作者： Jiajun Tang Haofeng Zhong Shuchen Weng Boxin Shi National Key Laboratory for Multimedia Information Processing and National Engineering Research Center of Visual Technology School of Computer Science Peking University

We present the ilLumination-Aware conditional Image Repainting (LuminAIRe) task to address the unrealistic lighting effects in recent conditional image repainting (CIR) methods. The environment lighting and 3D geometry conditions are explicitly estimated from given background images and parsing masks using a parametric lighting representation and learning-based priors. These 3D conditions are then converted into illumination images through the proposed physically-based illumination rendering and illumination attention module. With the injection of illumination images, physically-correct lighting information is fed into the lighting-realistic generation process and repainted images with harmonized lighting effects in both foreground and background regions can be acquired, whose superiority over the results of state-of-the-art methods is confirmed through extensive experiments. For facilitating and validating the LuminAIRe task, a new dataset CAR-LUMINAIRE with lighting annotations and rich appearance variants is collected.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Semantic Graph Embedded Energy Minimization Learning for Scene Graph Generation

Semantic Graph Embedded Energy Minimization Learning for Sce...

引用

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

作者： Jinghang Chen Chi Zhang Yuehu Liu Le Wang Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University Xi’an China National Engineering Research Center for Visual Information and Applications National Key Laboratory of Human-Machine Hybrid Augmented Intelligence

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

The performance of current scene graph generation models is affected by training with cross-entropy loss, exacerbating the problem of prediction bias stemming from biased training data. Energy-based model adopts a learning method for joint image and scene graph to alleviate this challenge. However, this method only focuses on the visual features of images, neglecting the rich relation information contained in the semantic space. To address this issue, we innovatively employ powerful pre-trained large models to realize a simple yet effective semantic graph embedded energy minimization framework for the SGG task. Specifically, we use large models to generate image descriptions and extract relation triplets, which are then transformed into semantic graphs with entities as nodes and relations as edges. Moreover, by mapping these graphs into the same space using GNN to learn the minimal energy value, our approach enables SGG model to learn structural information in both semantic and visual spaces. We validate the effectiveness and efficiency of our method on the SGG benchmark visual Genome dataset. Compared with prevailing models and EBM, we achieve a significant performance improvement of up to 2.99% and 2.24%, respectively.

关键词： Training Learning systems visualization Image edge detection Semantics Training data Signal processing Predictive models Minimization Speech processing

来源：评论

学校读者我要写书评

暂无评论

Learning to Learn Transferable Generative Attack for Person Re-Identification

引用

IEEE Transactions on Image processing 2025年 PP卷 PP页

作者： Bian, Yuan Liu, Min Wang, Xueping Ma, Yunfeng Wang, Yaonan Hunan University National Engineering Research Center of Robot Visual Perception and Control Technology College of Electrical and Information Engineering Hunan Changsha China Hunan Normal University Hunan Provincial Key Laboratory of Intelligent Computing and Language Information Processing College of Information Science and Engineering Hunan Changsha China

Deep learning-based person re-identification (reid) models are widely employed in surveillance systems and inevitably inherit the vulnerability of deep networks to adversarial attacks. Existing attacks merely consider cross-dataset and cross-model transferability, ignoring the cross-test capability to perturb models trained in different domains. To powerfully examine the robustness of real-world re-id models, the Meta Transferable Generative Attack (MTGA) method is proposed, which adopts meta-learning optimization to promote the generative attacker producing highly transferable adversarial examples by learning comprehensively simulated transfer-based crossmodel&dataset&test black-box meta attack tasks. Specifically, cross-model&dataset black-box attack tasks are first mimicked by selecting different re-id models and datasets for meta-train and meta-test attack processes. As different models may focus on different feature regions, the Perturbation Random Erasing module is further devised to prevent the attacker from learning to only corrupt model-specific features. To boost the attacker learning to possess cross-test transferability, the Normalization Mix strategy is introduced to imitate diverse feature embedding spaces by mixing multi-domain statistics of target models. Extensive experiments show the superiority of MTGA, especially in cross-model&dataset and cross-model&dataset&test attacks, our MTGA outperforms the SOTA methods by 20.0% and 11.3% on mean mAP drop rate, respectively. © 1992-2012 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：