检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Yuan, Zheng Zhang, Jie Wang, Yude Shan, Shiguang Chen, Xilin Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing100190 China University of Chinese Academy of Sciences Beijing100049 China

The attention mechanism has been proven effective on various visual tasks in recent years. In the semantic segmentation task, the attention mechanism is applied in various methods, including the case of both Convolution Neural Networks (CNN) and Vision Transformer (ViT) as backbones. However, we observe that the attention mechanism is vulnerable to patch-based adversarial attacks. Through the analysis of the effective receptive field, we attribute it to the fact that the wide receptive field brought by global attention may lead to the spread of the adversarial patch. To address this issue, in this paper, we propose a Robust Attention Mechanism (RAM) to improve the robustness of the semantic segmentation model, which can notably relieve the vulnerability against patch-based attacks. Compared to the vallina attention mechanism, RAM introduces two novel modules called Max Attention Suppression and Random Attention Dropout, both of which aim to refine the attention matrix and limit the influence of a single adversarial patch on the semantic segmentation results of other positions. Extensive experiments demonstrate the effectiveness of our RAM to improve the robustness of semantic segmentation models against various patch-based attack methods under different attack settings. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Disentangled latent factors for muti-cause treatment effect estimation 6

Disentangled latent factors for muti-cause treatment effect ...

引用

2024 6th International Conference on Artificial Intelligence Technologies and Applications, ICAITA 2024

作者： Zhao, Zhuoyue Wang, Haotian Xu, Liyang Wang, Qian Tang, Yuhua Department of Intelligent Data Science National University of Defense Technology Changsha410073 China Intelligent Game and Decision Lab. Academy of Military Science Beijing100071 China Institute for Quantum Information State Key Laboratory of High-Performance Computing College of Computer Science and Technology National University of Defense Technology Changsha410073 China

Existing methods estimate treatment effects from observational data and assume that covariates are all confounders. However, observed covariates may not directly represent confounding variables that influence both treatment and outcome. They always include variables that only affect the treatment or the outcome. In addition, for multi-dimensional binary treatments, disentangled methods are mainly designed for binary variables and ignore the impact of multi-cause treatment variables on the inference of latent factors. To address these two issues, based on variable decomposition and proxy inference, we propose the Disentangled Latent Factors for Multi-cause Treatment Estimation (DEMTE) algorithm. It utilizes an identifiable autoencoder to infer and disentangle latent factors based on the joint distribution of variables in observational data. DEMTE evaluates the treatment effect on the disentangled factors. Synthetic experiments and semi-synthetic experiments demonstrate the effectiveness of the inference and disentanglement techniques and our method achieves more accurate treatment effect estimation. © 2024 institute of Physics Publishing. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Adaptive Spatial Modeling and Multi-Scale Attention Aggregation for Semantic Segmentation of Remote Sensing Images

Adaptive Spatial Modeling and Multi-Scale Attention Aggregat...

引用

Signal, information and Data processing (ICSIDP), IEEE International Conference on

作者： Wenying Yang Yuchuan Zhang Yuxuan Jin Yupei Wang Liang Chen National Key Laboratory of Science and Technology on Space-Born Intelligent Information Processing Beijing Institute of Technology Chongqing Innovation Center Chongqing China School of Artificial Intelligence Jianghan University Wuhan China National Key Laboratory of Science and Technology on Space-Born Intelligent Information Processing Beijing Institute of Technology Beijing China National Key Laboratory of Science and Technology on Space-Born Intelligent Information Processing Beijing Institute of Technology Chongqing Innovation Center Beijing Institute of Technology Beijing China

ISBN: (数字)9798331515669

ISBN: (纸本)9798331515676

As remote sensing technology advances and high-resolution sensors are deployed, the analysis of high-resolution remote sensing images encounters challenging issues, such as intra-class variability, and inter-class similarity, due to the unobvious appearance and complicated background. The high-resolution remote sensing images pose critical issues for precise semantic segmentation due to these fundamental issues. To this end, we propose an adaptive hierarchical aggregation network. Different from the previous approach of modeling semantic interdependencies solely within a single feature map in the channel dimension, we introduce multi-scale channel-wise feature aggregation to capture global contextual dependencies. Meanwhile, we design an adaptive detail-aware module to model long-range spatial dependencies, adaptively extracting effective details from different levels as compensation. Additionally, this module utilizes depth-wise separable convolution to selectively filter out irrelevant spatial details from lower-level feature maps. The outputs of the two modules are fused layer by layer to achieve refined feature representation, thereby generating precise pixel-level segmentation results. Our experiments are conducted on the Vaihingen dataset and the Potsdam dataset. The experimental results demonstrate that the proposed algorithm surpasses several existing state-of-the-art approaches.

关键词： Image sensors Adaptation models Adaptive systems Semantic segmentation Semantics Filtering algorithms Feature extraction Data processing Sensors Remote sensing

来源：评论

学校读者我要写书评

暂无评论

Architecture Design of Equipment Warehouse Scheduling System Based on Software Definition 5th

Architecture Design of Equipment Warehouse Scheduling System...

引用

5th International Conference on Smart Vehicular technology, Transportation, Communication and Applications, VTCA 2022

作者： Zhang, Xue Ting Pan, Yan Peng Yang, Li Jie Xue, Chen Chen Zhang, Fu Quan China Academy of Launch Vehicle Technology Beijing100076 China School of Automation Beijing Institute of Technology Beijing100081 China Fujian Provincial Key Laboratory of Information Processing and Intelligent Control Minjiang University Fuzhou350121 China

ISBN: (纸本)9789819908479

The technology of software-defined changing all areas of life, but the equipment storage system in the field of application is still in its infancy. At present, the traditional equipment warehousing scheduling system is difficult to meet the new needs of rapid scheduling and flexible maintenance of large quantities of equipment. Based on the development and application characteristics of software definition technology, this paper integrates the traditional equipment storage scheduling system and presents the research on the architecture design technology of equipment storage scheduling system under software definition technology. This paper clarifies the key application technologies of the system, realizes the design of new open system architecture based on software-defined technology and the optimization design of system center service platform architecture, and improves the architecture design ability of automation, configuration ability, and easy maintenance of the existing equipment storage scheduling system. It solves the new needs of rapid deployment, intelligent upgrade, flexible configuration, and convenient service of large quantities of equipment warehouse scheduling system. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Application programs

来源：评论

学校读者我要写书评

暂无评论

Text-Driven Generative Domain Adaptation with Spectral Consistency Regularization

Text-Driven Generative Domain Adaptation with Spectral Consi...

引用

International Conference on Computer Vision (ICCV)

作者： Zhenhuan Liu Liang Li Jiayu Xiao Zheng-Jun Zha Qingming Huang Institute of Computing Technology Key Laboratory of Intelligent Information Processing CAS University of Chinese Academy of Sciences University of Science and Technology of China Peng Cheng Laboratory

Combined with the generative prior of pre-trained models and the flexibility of text, text-driven generative domain adaptation can generate images from a wide range of target domains. However, current methods still suffer from overfitting and the mode collapse problem. In this paper, we analyze the mode collapse from the geometric point of view and reveal its relationship to the Hessian matrix of generator. To alleviate it, we propose the spectral consistency regularization to preserve the diversity of source domain without restricting the semantic adaptation to target domain. We also design granularity adaptive regularization to flexibly control the balance between diversity and stylization for target model. We conduct experiments for broad target domains compared with state-of-the-art methods and extensive ablation studies. The experiments demonstrate the effectiveness of our method to preserve the diversity of source domain and generate high fidelity target images. Source code has been released in https://***/Victarry/Adaptation-SCR.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Unified segment-to-segment framework for simultaneous sequence generation 23

Unified segment-to-segment framework for simultaneous sequen...

引用

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

作者： Shaolei Zhang Yang Feng Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences (ICT/CAS) and University of Chinese Academy of Sciences

Simultaneous sequence generation is a pivotal task for real-time scenarios, such as streaming speech recognition, simultaneous machine translation and simultaneous speech translation, where the target sequence is generated while receiving the source sequence. The crux of achieving high-quality generation with low latency lies in identifying the optimal moments for generating, accomplished by learning a mapping between the source and target sequences. However, existing methods often rely on task-specific heuristics for different sequence types, limiting the model's capacity to adaptively learn the source-target mapping and hindering the exploration of multi-task learning for various simultaneous tasks. In this paper, we propose a unified segment-to-segment framework (Seg2Seg) for simultaneous sequence generation, which learns the mapping in an adaptive and unified manner. During the process of simultaneous generation, the model alternates between waiting for a source segment and generating a target segment, making the segment serve as the natural bridge between the source and target. To accomplish this, Seg2Seg introduces a latent segment as the pivot between source to target and explores all potential source-target mappings via the proposed expectation training, thereby learning the optimal moments for generating. Experiments on multiple simultaneous generation tasks demonstrate that Seg2Seg achieves state-of-the-art performance and exhibits better generality across various tasks. Code is available at: https://***/ictnlp/Seg2Seg.

关键词：

来源：评论

学校读者我要写书评

暂无评论

SKT-MOT and DyTracker: A Multiobject Tracking Dataset and a Dynamic Tracker for Speed Skating Video

引用

Scientific Programming 2023年第1期2023卷

作者： Wang, Junwu Li, Zongmin Li, Yachuan Yang, Shaobo Wang, Ben Li, Hua Qingdao266580 China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing100190 China

Speed skating serves as a significant application domain for multiobject tracking (MOT), presenting unique challenges such as frequent occlusion, highly similar appearances, and motion blur. To address these challenges, this paper constructs an MOT dataset called SKT-MOT for speed skating and analyzes the shortcomings of existing datasets and methods. Accordingly, we propose a dynamic MOT method called DyTracker. The method builds upon the DeepSORT baseline and enhances three key modules. At the global level, we design the track dynamic management (TDM) algorithm. In the motion branch, a novel metric is proposed to evaluate occlusion and Kalman filter dynamic update (KFDU) is implemented. In the appearance branch, we account for the difference in human posture and propose the feature dynamic selection and updating (FDSU) strategy. This makes our DyTracker flexible and efficient to achieve a multiobject tracking accuracy (MOTA) of 93.70% and identification F1 (IDF1) score of 92.39% on SKT-MOT, which is a significant advantage over existing SOTA methods. To validate the generalization of our proposed module, two dynamic update modules are inserted into other methods and validated on the public dataset MOT17, and the accuracy is generally improved by 0.2%-0.6%. © 2023 Junwu Wang et al.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Long-Short Term Policy for Visual Object Navigation

Long-Short Term Policy for Visual Object Navigation

引用

IEEE/RSJ International Conference on intelligent Robots and Systems (IROS)

作者： Yubing Bai Xinhang Song Weijie Li Sixian Zhang Shuqiang Jiang Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences (CAS) Beijing China University of Chinese Academy of Sciences Beijing China Institute of Intelligent Computing Technology CAS Suzhou

The goal of visual object navigation for an agent is to find the target objects accurately. Recent works mainly focus on the feature of embedding, attempting to learn better features with different variants, such as object distribution and graph representations. However, some typical navigation problems in complex environments, such as partially known and obstacle problems, may not be effectively addressed by previous feature embedding methods. In this paper, we propose a framework with a long-short objective policy, where the hidden states are classified according to the navigation objectives at that moment and separately rewarded. Specifically, we consider two objectives: the long-term objective is to go closer to the target, and the short-term objective is for obstacle avoidance and exploration. To alleviate the effect of long-term and short-term alternation, we build a state memory and propose an adjustment gate to update the state memory. Finally, all past hidden states are reweighted and combined for action prediction with an action-boosting gate. Experimental results on RoboTHOR show that the proposed method can significantly outperform the state-of-the-art.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Predicting Calibrated Conversion Rate of Online Advertising Using a Multi-task Mixture-of-Experts Calibration Model 12th

Predicting Calibrated Conversion Rate of Online Advertising...

引用

12th CCF Conference on BigData, BigData 2024

作者： Zhang, Xinyue Guo, Yuyao Ao, Xiang School of Information Engineering China University of Geosciences Beijing China Beijing China Key Lab of Intelligent Information Processing Institute of Computing Technology CAS Beijing China

ISBN: (纸本)9789819610235

Accurately predicting conversion rate (CVR) is paramount in online advertising. However, traditional models may face problems such as delayed feedback, where there is a delay of an indeterminate amount of time between click and conversion. Calibration is an effective way to optimize conversion rate estimates in online advertising. Unlike conversion delays, post-click user behaviors occur rapidly and are informative to conversion rate prediction. Our proposed solution, the Multi-Task Mixture-of-Experts Calibration (MTMEC) framework, integrates multi-task learning and mixture-of-experts models. It modifies CVR prediction using post-click user behavior data, utilizing streaming learning for real-time data access. Each task is weighted by a gating network, enabling adaptive loss functions through multi-task learning. Parametric scaling further minimizes calibration errors, enhancing prediction accuracy without excessive parameters. Experiments on real-world datasets validate the effectiveness of MTMEC. It improves model prediction and reduces calibration errors. This framework offers a robust solution for online advertising, bridging the gap between calibration accuracy and system responsiveness. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Multi-task learning

来源：评论

学校读者我要写书评

暂无评论

Face Forgery Detection Algorithm Based on Improved MobileViT Network 8

Face Forgery Detection Algorithm Based on Improved MobileViT...

引用

8th International Conference on intelligent computing and Signal processing, ICSP 2023

作者： Wang, Tiantian Lu, Xiaoqi Inner Mongolia University of Science and Technology School of Information Engineering Baotou014010 China Inner Mongolia Key Laboratory of Pattern Recognition and Intelligent Image Processing Baotou014010 China Inner Mongolia University of Technology Institute of Information Engineering Hohhot010051 China

ISBN: (纸本)9798350302455

DeepFakes blur the boundaries between reality and forgery, resulting in the collapse of exiting credit system, causing immeasurable consequences for national security and social order. Through analysis of existing face forgery techniques, it is found that most generation techniques rely on random noise distribution, and global information will be lost after up sampling. Therefore, we propose a deepfake detection algorithm based on improved MobileViT, which uses CNN local space biasing and the global space representation of the Transformer network to learn the local features and global representation of forged faces, respectively. Coordinate attention is introduced to obtain directional perception and position sensitive information, making the model locate synthetic traces of fake faces better and fusion local and global representation more effectively. For the improved generalization of the model, with the GELU activation function to solve the problem of neuron death. Our model achieved 96.2% on FF++(C23) datasets, and 93.7%,94.1%,96.3%,87.9% on DF, F2F, FS, and NT datasets, respectively. Comparing with previous methods, our model has shown detection robustness and better generalization. © 2023 IEEE.

关键词： National security

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：