检索结果-内蒙古大学图书馆

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Xiaohan Wang Yuehu Liu Xinhang Song Yuyi Liu Sixian Zhang Shuqiang Jiang Institute of Artificial Intelligence and Robotics Xi’ an Jiaotong University Xi'an Key Lab of Intelligent Information Processing Laboratory of the Chinese Academy of Sciences (CAS) Institute of Computing Technology Beijing University of Chinese Academy of Sciences Beijing

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

Visual navigation is to let the agent reach the target according to the continuous visual input. In most previous works, visual navigation is usually assumed to be done in a static and ideal environment: the target is always reachable with no need to alter the environment. However, the “messy” environments are more general and practical in our daily lives, where the agent may get blocked by obstacles. Thus Interactive Navigation (InterNav) is introduced to navigate to the objects in more realistic “messy” environments according to the object interaction. Prior work on InterNav learns shortterm interaction through extensive trials with reinforcement learning. However, interaction does not guarantee efficient navigation, that is, plan-ning obstacle interactions that make shorter paths and con-sume less effort is also crucial. In this paper, we introduce an effect-oriented affordance map to enable longterm interactive navigation, extending the existing map-based nav-igation framework to the domain of dynamic environment. We train a set of affordance functions predicting available interactions and the time cost of removing obstacles, which informatively support an interactive modular system to ad-dress interaction and longterm planning. Experiments on the ProcTHOR simulator demonstrate the capability of our affordance-driven system in longterm navigation in complex dynamic environments.

关键词： Visualization Computer vision Uncertainty Costs Navigation Affordances Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Learning Open-vocabulary Semantic Segmentation Models From Natural Language Supervision

arXiv

引用

arXiv 2023年

作者： Xu, Jilan Hou, Junlin Zhang, Yuejie Feng, Rui Wang, Yi Qiao, Yu Xie, Weidi School of Computer Science Shanghai Key Lab of Intelligent Information Processing Shanghai Collaborative Innovation Center of Intelligent Visual Computing Fudan University China Shanghai AI Laboratory China Shanghai Jiaotong University China

In this paper, we consider the problem of open-vocabulary semantic segmentation (OVS), which aims to segment objects of arbitrary classes instead of pre-defined, closed-set categories. The main contributions are as follows: First, we propose a transformer-based model for OVS, termed as OVSegmentor, which only exploits web-crawled image-text pairs for pre-training without using any mask annotations. OVSegmentor assembles the image pixels into a set of learnable group tokens via a slot-attention based binding module, and aligns the group tokens to the corresponding caption embedding. Second, we propose two proxy tasks for training, namely masked entity completion and cross-image mask consistency. The former aims to infer all masked entities in the caption given the group tokens, that enables the model to learn fine-grained alignment between visual groups and text entities. The latter enforces consistent mask predictions between images that contain shared entities, which encourages the model to learn visual invariance. Third, we construct CC4M dataset for pre-training by filtering CC12M with frequently appeared entities, which significantly improves training efficiency. Fourth, we perform zero-shot transfer on three benchmark datasets, PASCAL VOC 2012, PASCAL Context, and COCO Object. Our model achieves superior segmentation results over the state-of-the-art method by using only 3% data (4M vs 134M) for pre-training. Code and pre-trained models will be released for future research. © 2023, CC BY.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Hybrid Near- and Far-Field Source Localization Using Symmetric Sparse Array

Hybrid Near- and Far-Field Source Localization Using Symmetr...

引用

IEEE Asia-Pacific Conference on Antennas and Propagation (APCAP)

作者： Jiawen Zhang Junwei Qi Yingsong Li College of Information and Communication Engineering Harbin Engineering University Harbin China Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Anhui University China

Recently, symmetric linear nonuniform arrays (SLNAs) have been promoted for hybrid near-field (NF) and far-field (FF) source localization. A new SLNA based on symmetric nested array (SNA) is presented in the paper. The proposed array and its difference coarray (DCA) configuration can be expressed in closed-form. Comparing to the existing SLNAs with same sensors number, the proposed array can locate more sources and provide longer consecutive lags, which optimizes the detection results. Simulation results prove that the designed array outperforms traditional compressed symmetric nested array (CSNA).

关键词： Location awareness Simulation Sensors Noise measurement Sensor arrays Antenna arrays

来源：评论

学校读者我要写书评

暂无评论

(Image Present) ForgerySleuth: Empowering Multimodal Large Language Models for Image Manipulation Detection

arXiv

引用

arXiv 2024年

作者： Sun, Zhihao Jiang, Haoran Chen, Haoran Cao, Yixin Qiu, Xipeng Wu, Zuxuan Jiang, Yu-Gang Shanghai Key Lab of Intell. Info. Processing School of CS Fudan University China Shanghai Collaborative Innovation Center of Intelligent Visual Computing China School of Computer Science Fudan University China

(Image Present) Multimodal large language models have unlocked new possibilities for various multimodal tasks. However, their potential in image manipulation detection remains unexplored. When directly applied to the IMD task, M-LLMs often produce reasoning texts that suffer from hallucinations and overthinking. To address this, in this work, we propose ForgerySleuth, which leverages M-LLMs to perform comprehensive clue fusion and generate segmentation outputs indicating specific regions that are tampered with. Moreover, we construct the ForgeryAnalysis dataset through the Chain-of-Clues prompt, which includes analysis and reasoning text to upgrade the image manipulation detection task. A data engine is also introduced to build a larger-scale dataset for the pre-training phase. Our extensive experiments demonstrate the effectiveness of ForgeryAnalysis and show that ForgerySleuth significantly outperforms existing methods in generalization, robustness, and explainability. The resource is available at https://***/sunzhihao18/ForgerySleuth. Copyright © 2024, The Authors. All rights reserved.

关键词： Large datasets

来源：评论

学校读者我要写书评

暂无评论

The Application of the ADI-FDTD method in the 3D hybrid subgrid technique

The Application of the ADI-FDTD method in the 3D hybrid subg...

引用

2022 International Applied Computational Electromagnetics Society Symposium, ACES-China 2022

作者： Song, Ziheng Xie, Guoda Huang, Zhixiang Hou, Guilin Pan, Pan Anhui University School of Electronic and Information Engineering Hefei China Education of Anhui Province Anhui Province Key Laboratory of Target Recognition and Feature Extraction Anhui Luan237000 China Ministry of Education The Key Laboratory of Intelligent Computing and Signal Processing Hefei China Anhui Province Anhui University Key Laboratory of Electromagnetic Environmental Sensing Hefei230601 China Anhui University Hefei China

ISBN: (纸本)9781665452366

A hybrid subgrid scheme based on the conventional finite-difference time-domain (FDTD) schemes are proposed. The alternating-direction-implicit FDTD (ADI-FDTD) is used to calculate electromagnetic fields in fine grid regions subgrids and the FDTD method is applied to the coarse grid regions. Numerical results demonstrate that the drastically reduced computation time and memory can be achieved by using the proposed 3D hybrid algorithm. © 2022 IEEE.

关键词： Finite difference time domain method

来源：评论

学校读者我要写书评

暂无评论

Non-autoregressive Machine Translation with Probabilistic Context-free Grammar

arXiv

引用

arXiv 2023年

作者： Gui, Shangtong Shao, Chenze Ma, Zhengrui Zhang, Xishan Chen, Yunji Feng, Yang State Key Lab of Processors Institute of Computing Technology Chinese Academy of Sciences China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences China University of Chinese Academy of Sciences China Cambricon Technologies China

Non-autoregressive Transformer(NAT) significantly accelerates the inference of neural machine translation. However, conventional NAT models suffer from limited expression power and performance degradation compared to autoregressive (AT) models due to the assumption of conditional independence among target tokens. To address these limitations, we propose a novel approach called PCFG-NAT, which leverages a specially designed Probabilistic Context-Free Grammar (PCFG) to enhance the ability of NAT models to capture complex dependencies among output tokens. Experimental results on major machine translation benchmarks demonstrate that PCFG-NAT further narrows the gap in translation quality between NAT and AT models. Moreover, PCFG-NAT facilitates a deeper understanding of the generated sentences, addressing the lack of satisfactory explainability in neural machine translation.2 Copyright © 2023, The Authors. All rights reserved.

关键词： Computer aided language translation

来源：评论

学校读者我要写书评

暂无评论

2023 7th International Conference on Communication and Information Systems An Improved IOTA Protocol Based on Optimal Midway Walk

2023 7th International Conference on Communication and Infor...

引用

IEEE International Conference on Communication and Information Systems (ICCIS)

作者： Hongcheng Wu Hui Zhi Yangyang Liu Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education Anhui University School of Electronic and Information Engineering Anhui University Hefei China Hefei Zhilian Technology Industry Investment Co. Ltd Hefei China

IOTA (Internet of Things Applications) protocol aims to provide a secure and trustworthy protocol for realizing data-value interactions between IoT devices, and it has become the next possible direction for the development of Blockchain applications. IOTA protocol has the problem that the time-consuming cumulative weight update grows too fast with the increase of the number of transactions, this paper proposed an improved IOTA protocol based on optimal midway walk, the improved protocol uses cutting-edge transaction selection algorithm of weighted Monte Carlo Markov random walk to reduce cumulative weight update delay. Firstly, this improved IOTA protocol mainly ensures the efficiency of transaction confirmation without losing the security of consensus by introducing the optimal midway walk method, i.e., the starting point of each walk is chosen to depart at an optimal location. Secondly, this paper designs the indirect verification identity record algorithm and the layer degree of confirmation principle are designed to realize the improved IOTA protocol's resilience and resistance to Parasite chain attack. The experimental results show that the improved IOTA using random function to select the midway walk starting point from i-50λ, to i-75λsubstantially solves the system computational burden problem and greatly reduces the cumulative weight update delay.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Low-profile Quadrifilar Helix Antenna for Beidou Navigation System

A Low-profile Quadrifilar Helix Antenna for Beidou Navigatio...

引用

IEEE Asia-Pacific Conference on Antennas and Propagation (APCAP)

作者： Jiahong Wang Junwei Qi Yingsong Li College of Information and Communication Engineering Harbin Engineering University Harbin China Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Anhui University China

Ahstract-A new low-profile quadrifilar helical antenna (QHA) is proposed and investigated. Four spiral arms of the QHA are divided and connected to for a sleeve structure to reduce the height of the antenna. In this case, four spiral arms on the inner cylindrical sleeve are connected with the ground, while the upper ends of the four spiral arms on the outer cylindrical sleeve are short-circuited, and inner and outer spiral arms are connected in the middle. The QHA has a beam width reaches to above $120 ^{\circ}$ and a cardioid pattern at 1.575 GHz. As a result, the QHA has a very low profile of only 22 mm. The simulation and optimization from the CST show that the center frequency of the QHA is 1.575 GHz, which is used for the satellite navigation system, with axis ratio in the operating band less than 3 dB.

关键词： Satellite antennas Helical antennas Spirals Satellite navigation systems Optimization Antenna radiation patterns

来源：评论

学校读者我要写书评

暂无评论

Sparse Linear Array Designed via ULA Fitting

Sparse Linear Array Designed via ULA Fitting

引用

IEEE Asia-Pacific Conference on Antennas and Propagation (APCAP)

作者： Wanlu Shi Yingsong Li Zhixiang Huang College of Information and Communication Engineering Harbin Engineering University Harbin China Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Anhui University China

A sparse array (SA), termed as ATLF3-2BL, is proposed based on the uniform linear array (ULA) fitting (UF) principle. UF is a general SA devising scheme, based on which designers can devise SAs using a series of ULAs. SAs devised upon UF enjoy closed-form expressions for both array configuration and uniform degrees of freedom. Comprised of 7 ULAs, the proposed ATLF3-2BL can provide a long virtual difference coarray than most existing SAs. Besides, the mutual coupling in ATLF3-2BL remains a low level. Therefore, the ATLF3-2BL gives a considerable compromise between mutual coupling and uniform degrees of freedom. Simulations attest the good performance of the proposed ATLF3-2BL.

关键词： Couplings Mutual coupling Closed-form solutions Fitting Sensors Antenna arrays

来源：评论

学校读者我要写书评

暂无评论

High order SF-SFDTD method for solving the time-dependent Schrödinger equation

High order SF-SFDTD method for solving the time-dependent Sc...

引用

2022 International Applied Computational Electromagnetics Society Symposium, ACES-China 2022

作者： Pan, Pan Xie, Guoda Huang, Zhixiang Hou, Guilin Song, Ziheng Anhui University School of Electronic and Information Engineering Hefei China Education of Anhui Province Anhui Province Key Laboratory of Target Recognition and Feature Extraction Anhui Luan237000 China Ministry of Education The Key Laboratory of Intelligent Computing and Signal Processing Hefei China Anhui Province Anhui University Key Laboratory of Electromagnetic Environmental Sensing Hefei230601 China Anhui University Hefei China

ISBN: (纸本)9781665452366

A high-order spatial filtering-symplectic finite difference time domain (SF-SFDTD) scheme with controllable stability condition is proposed to solve the time-dependent Schrödinger equation (TDSE). Firstly, the higher-order symplectic framework for the discretization of Schrödinger equation is described, and then the spatial filtering method is carried out to extend the stability of the standard SFDTD method. Finally, a numerical example is given to evaluate the correctness of the SF-SFDTD method in numerical solving Schrödinger equation. © 2022 IEEE.

关键词： Finite difference time domain method

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：