检索结果-内蒙古大学图书馆

FERMixNet: An Occlusion Robust Facial Expression Recognition Model with Facial Mixing Augmentation and Mid-Level Representation Learning

引用

IEEE Transactions on Affective Computing 2024年第2期16卷 639-654页

作者： Huang, Yansong Peng, Junjie Zhang, Wenqiang Zhao, Tong Chen, Gan Tan, Shuhua Yi, Fen Wang, Lu Shanghai University School of Computer Engineering and Science Shanghai China Shanghai University School of Computer Engineering and Science The Shanghai Institute for Advanced Communication and Data Science Shanghai China Fudan University Academy for Engineering and Technology School of Computer Science and Technology Shanghai China YTO Express Co. Ltd. The National Logistics Engineering Laboratory Shanghai China

Facial expressions can provide a better understanding of people's mental status and attitudes towards specific things. However, facial occlusion in real world is an unfavorable phenomenon that greatly affects the performance of facial expression recognition models. Recent works addressing the occlusion problem have primarily relied on attention mechanisms or occlusion discarding methods that focus on non-occluded regions of the face. However, these methods have not achieved a good balance between occlusion robustness and model efficiency. In this paper, we propose a simple and efficient model, called FERMixNet, for occluded facial expression recognition. The model incorporates a novel facial mixing augmentation strategy (FERMix) that generates new training samples by simulating real-world facial occlusion and preserving high expression-related semantic information. By co-training the original and newly generated samples, the model's occlusion robustness is improved without increasing its complexity during inference. Additionally, to further enhance the model's occlusion robustness, we include mid-level representation learning in the network to learn the discriminative non-occluded local features of the samples with low computational cost. Extensive experiments on four public facial occlusion datasets: Occlusion-RAF-DB, Occlusion-FERPlus and FED-RO show that the proposed model achieves state-of-the-art results which demonstrates the good robustness of our method for occluded facial expression recognition. Meanwhile, the proposed model also achieves state-of-the-art results on the in-the-wild facial expression datasets RAF-DB, AffectNet-8, and AffectNet-7. It proves that the proposed model has good application prospects in real world. © 2010-2012 IEEE.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Selective sampling and temporal positional encoding for monocular video based 3D human pose and shape estimation 5

Selective sampling and temporal positional encoding for mono...

引用

5th International Conference on Signal Processing and computer science, SPCS 2024

作者： Hou, Yupeng Wen, Han Zeng, Guangping School of Computer & Communication Engineering University of Science and Technology Beijing Beijing100083 China

ISBN: (数字)9781510686731

ISBN: (纸本)9781510686724

This study introduces a novel framework for video-based 3D human pose and shape estimation, termed Selective sampling and Temporal Positional Encoding (STPE). Our method leverages selective sampling and advanced positional encoding to tackle the temporal complexities of video data and the high cost and scarcity of annotated datasets. Inspired by the Masked Autoencoder (MAE), our approach adopts a selective sampling strategy that efficiently captures the essential dynamics of human motion from partial views, significantly reducing reliance on continuous frames. The framework incorporates Rotary Position Embedding (RoPE), using rotational angles to simplify positional encoding. This innovation decreases model complexity and boosts learning effectiveness. We also introduce randomized index positions during training, introducing variability and enhancing generalization across various datasets and motion patterns. Our model, validated on standard datasets like 3DPW, MPI-INF-3DHP, and Human3.6M, shows enhanced performance in accurate and robust 3D pose and shape capture compared to existing methods. Our results demonstrate that strategic frame sampling and sophisticated positional encoding can significantly improve accuracy and robustness of video-based pose estimation systems. © 2025 SPIE.

关键词： Video recording

来源：评论

学校读者我要写书评

暂无评论

USRD: User-Searchable and Revocable Data Sharing Scheme in Multi-Fog-IoVs 6

USRD: User-Searchable and Revocable Data Sharing Scheme in M...

引用

6th International Conference on Next Generation Data-Driven Networks, NGDN 2024

作者： Jiang, LuYao Li, Yongkang Chen, DongDong Gu, Ke Zuo, YiMing School of Computer and Communication Engineering Changsha University of Science & Technology ChangSha China

ISBN: (纸本)9798350388374

Aiming at the security of data sharing in the Internet of vehicles environment, we propose a user-searchable and revocable data-sharing scheme in the Multi-Fog-IoVs. The scheme utilizes a fog node to alleviate the computing burden on users and enhance the computing efficiency of the system. The Garbled Bloom filter and inverted index are used to create a bidirectional index for efficient multi-keyword search and document retrieval. Furthermore, a flexible and efficient user revocation mechanism is constructed while realizing the dynamic updates of data security. Finally, the safety analysis and performance evaluation of the proposed scheme are conducted, and the results indicate that the scheme meets both forward and backward safety requirements while demonstrating outstanding retrieval efficiency. © 2024 IEEE.

关键词： Fog computing

来源：评论

学校读者我要写书评

暂无评论

Emotion-Aware Synthetic Speech Detection with Multi-level Fusion Features 2nd

Emotion-Aware Synthetic Speech Detection with Multi-level Fu...

引用

2nd International Conference on the Frontiers of Robotics and Software Engineering, FRSE 2024

作者： Xiang, Lingyun Zhou, Zhiguo Liu, Yangfan School of Computer and Communication Engineering Changsha University of Science and Technology Changsha410114 China

ISBN: (纸本)9789819631605

Current synthetic speech detection methods often overlook the emotional distinctions between synthetic and real speech. To comprehensively leverage emotional information and deep features to further enhance the accuracy of distinguishing synthetic speeches, this paper proposes an emotion-aware synthetic speech detection method by fusing multi-level different features. Initially, a multi-scale emotional feature extraction model is devised to extract emotional features from 3-D log-Mels. Considering that the emotional features learned by a single model may not be comprehensive enough, the emotion2vec model is utilized to automatically learn additional emotional features from the temporal domain of the original speech. Subsequently, by employing a co-attention network, the extracted two kinds of emotional features are both incorporated into the deep features learned by the end-to-end model Rawnet2, thereby obtaining highly sensitive and weighted deep features. Finally, the extracted emotional features and the weighted deep features are fused to classify synthetic speech from real speech. Experimental results on several datasets demonstrate that the proposed method outperforms baseline methods, with a detection accuracy exceeding 0.97. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Speech recognition

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Multi-view Pedestrian Detection 24

Unsupervised Multi-view Pedestrian Detection

引用

32nd ACM International Conference on Multimedia, MM 2024

作者： Liu, Mengyin Zhu, Chao Ren, Shiqi Yin, Xu-Cheng School of Computer and Communication Engineering University of Science and Technology Beijing Beijing China

ISBN: (纸本)9798400706868

With the prosperity of the intelligent surveillance, multiple cameras have been applied to localize pedestrians more accurately. However, previous methods rely on laborious annotations of pedestrians in every frame and camera view. Therefore, we propose in this paper an Unsupervised Multi-view Pedestrian Detection approach (UMPD) to learn an annotation-free detector via vision-language models and 2D-3D cross-modal mapping: 1) Firstly, Semantic-aware Iterative Segmentation (SIS) is proposed to extract unsupervised representations of multi-view images, which are converted into 2D masks as pseudo labels, via our proposed iterative PCA and zero-shot semantic classes from vision-language models;2) Secondly, we propose Geometry-aware Volume-based Detector (GVD) to end-to-end encode multi-view 2D images into a 3D volume to predict voxel-wise density and color via 2D-to-3D geometric projection, trained by 3D-to-2D rendering losses with SIS pseudo labels;3) Thirdly, for better detection results, i.e., the 3D density projected on Birds-Eye-View, we propose Vertical-aware BEV Regularization (VBR) to constrain pedestrians to be vertical like the natural poses. Extensive experiments on popular multi-view pedestrian detection benchmarks Wildtrack, Terrace, and MultiviewX, show that our proposed UMPD, as the first fully-unsupervised method to our best knowledge, performs competitively to the previous state-of-the-art supervised methods. Code is available at https://***/lmy98129/UMPD. © 2024 ACM.

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

AIO: Automating I/O Optimization Pipeline for Data-Intensive Applications in HPC 22

AIO: Automating I/O Optimization Pipeline for Data-Intensive...

引用

22nd IEEE International Symposium on Parallel and Distributed Processing with Applications, ISPA 2024

作者： Wang, Wanxin Wu, Huijun Yang, Lihua Liu, Zhangyu Zheng, Bin Changsha University of Science and Technology School of Computer and Communication Engineering China National University of Defense Technology College of Computer Science and Technology China

ISBN: (纸本)9798331509712

High-performance computing (HPC) systems has entered the exascale era, but I/O performance has lagged behind due to storage hardware limitations, creating a "storage wall effect"that hinders HPC systems full potential. Modern HPC storage systems employ a layered storage architecture with local node storage, burst buffers, and global parallel file systems to improve I/O performance. However, application workloads vary, and fast storage layers may not always outperform global parallel file systems, while a static storage software stack configuration may not achieve optimal performance across different applications. Traditional I/O optimization is time-consuming, laborintensive, and prone to human error. To address this issue, this paper proposes AIO, a data-driven I/O optimization method. AIO automatically selects the optimal storage hierarchy and software stack based on the task's I/O patterns, and further enhances performance by identifying and applying the most efficient configuration settings. We evaluated AIO with eight applications in a cluster that features shared burst buffers, local burst buffers, and a global shared file system. The experiments demonstrate that AIO effectively optimizes I/O performance, achieving an overall speedup of 2.74. © 2024 IEEE.

关键词： Buffer storage

来源：评论

学校读者我要写书评

暂无评论

Channel Estimation for Double Intelligent Reflecting Surface Assisted MISO System in Sub-6G 3

Channel Estimation for Double Intelligent Reflecting Surface...

引用

3rd International Conference on Computing, communication, Perception and Quantum technology, CCPQT 2024

作者： Wang, Liang Zhang, Jun Shi, Wenzheng Wang, Shiguo School of Computer and Communication Engineering Changsha University of Science and Technology Changsha China

ISBN: (纸本)9798331528386

In contrast to the single intelligent reflecting surface (S-IRS) systems, improved performance is attainable by employing double-IRS (D-IRS) in a wireless-communication system. Nevertheless, precise channel estimation is crucial in D-IRS assisted wireless communication systems for achieving significant performance enhancements. In some specialized communication scenarios, obstacles may block the communication links between IRS1 and IRS2, thereby diminishing the potential gains of the D-IRS system. To tackle these challenges, this article introduces a novel D-IRS method with maximal a posterior probability rule (DMAP) for a multi-input single-output (MISO) uplink communication where the line-of-sight (LOS) link between the base station (BS) and the users are ignored. Furthermore, the BS-IRS links are harnessed as a common-link for decomposing the complex cascaded channels and IRSs are kept always "ON" in the channel estimation protocol. Finally, optimization for the phase configuration of IRSs is constructed to ensure better estimation accuracy. Experimental results show that DMAP significantly reduces pilot overhead and exceeds baseline performance levels. © 2024 IEEE.

关键词： channel estimation DMAP double-IRS (D-IRS) multiple-input single-output (MISO) single intelligent reflecting surface (S-IRS)

来源：评论

学校读者我要写书评

暂无评论

Exploration of Quantum Computing in Multi-Die FPGA Routing Problems 9

Exploration of Quantum Computing in Multi-Die FPGA Routing P...

引用

9th International Conference on Integrated Circuits and Microsystems, ICICM 2024

作者： Tian, Haochang Zhang, Xu Wang, Chengkai Ji, Weiqing Kou, Mingyang Yao, Hailong School of Computer & Communication Engineering University of Science and Technology Beijing Beijing China

ISBN: (纸本)9798331509453

This paper delves into the innovative application of quantum computing to solve complex multi-die FPGA (FieldProgrammable Gate Array) routing challenges. Leveraging the Quantum Approximate Optimization Algorithm (QAOA), we propose a novel approach that integrates classical routing techniques with quantum computing methodologies. This integration specifically utilizes QAOA to enhance the efficiency and effectiveness of the routing process. This hybrid approach partitions the routing task into intra-FPGA and inter-FPGA components, optimizing each segment through the appropriate computational techniques. Extensive experiments conducted on quantum computing simulators, alongside trials on actual quantum hardware, demonstrate the theoretical potential of quantum computing in enhancing the efficiency of multi-die FPGA routing solutions. Our findings suggest that despite the current limitations of quantum hardware, such as noise and error rates, quantum computing holds promise for advancing EDA methodologies. This paper contributes to the field by pioneering the integration of quantum algorithms with EDA techniques, offering a novel perspective on future applications of quantum computing in complex routing problems. © 2024 IEEE.

关键词： Quantum computers

来源：评论

学校读者我要写书评

暂无评论

Analyzing Public Disputation of Autonomous Driving technology: A Study Based on Data Mining 4

Analyzing Public Disputation of Autonomous Driving Technolog...

引用

4th IEEE International Conference on Digital Twins and Parallel Intelligence, DTPI 2024

作者： Lv, Lehao Chen, Yuan Ma, Fanshu Li, Yongwei University of Science and Technology Beijing School of Computer and Communication Engineering Beijing China

ISBN: (纸本)9798350349252

With the rapid development of autonomous driving technology, public's opinion of this technology shows diversification and complexity. The discussion about autonomous driving online has a large quantity and diverse perspectives, which causes the traditional data analysis methods hard to catch and understand the whole spectrum of these data. This paper explores how to use text mining techniques effectively, especially the Latent Dirichlet Allocation (LDA) model, to analyze and understand people's attitude and viewpoint towards autonomous driving technology. By calculating the perplexity, coherence, Jensen-Shannon (JS)divergence and other related parameter indicators of the LDA model, striving to obtain the text mining model with best performance, in order to get valuable information faster and more efficiently. © 2024 IEEE.

关键词： Autonomous driving data mining Latent Dirichlet Allocation

来源：评论

学校读者我要写书评

暂无评论

Community Topic Propagation Prediction-Based Malicious User Detection Scheme in Overlapping Community Networks 6

Community Topic Propagation Prediction-Based Malicious User ...

引用

6th International Conference on Next Generation Data-Driven Networks, NGDN 2024

作者： Deng, Yang Liu, Jian Gu, Ke Chen, Dongdong Li, Yongkang Jiang, LuYao Changsha University of Science & Technology School of Computer and Communication Engineering ChangSha China

ISBN: (纸本)9798350388374

In this paper, we propose a malicious user detection scheme for overlapping communities based on community topic propagation prediction. First, we construct a malicious topic classification model based on topic text feature vectors. At the same time, based on the community attributes of users obtained from overlapping community detection, we construct user attribute vectors and calculate the fit between users and topics, based on which we construct a topic propagation prediction model. Secondly, we combine the malicious topic classification model and topic propagation prediction model to construct the overlapping community malicious user detection model. Finally, experiments verify that our scheme can effectively identify malicious users. © 2024 IEEE.

关键词： Prediction models

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：