检索结果-内蒙古大学图书馆

49th IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Huang, Xianhan Zhu, Qingmeng Yu, Zhipeng Liu, Kun He, Hao Chinese Acad Sci Inst Software Beijing Peoples R China Beijing Univ Technol Beijing Peoples R China

ISBN: (纸本)9798350344868;9798350344851

Recent advances in Learned Image Compression (LIC) utilize the powerful masked Image Modeling (MIM) framework (e.g., MAE). However, the random masking treats all patches equally, which could lead (i) missing critical information and (ii) poor robustness, since different patches may have different contributions to image reconstruction. The principle experiment shows that the patch-level inherent feature evaluation is highly correlated with the quality of the reconstructed image. Based on this key insight, we propose a simple mask selection approach based on the patch's inherent feature. Specifically, we design plug-and-play frequency and spatial inherent feature modules to select masks, enhancing various MIMs based on the MAE model and achieving high-quality image reconstruction and strong robustness at high mask rates for data-intensive compression tasks.

关键词： Image Compression masked Image Modeling masked autoencoder

来源：评论

学校读者我要写书评

暂无评论

TrustHeartbeat: A Trustworthy Multimodal Interactive System with Heartbeat Recognition and User Identification

TrustHeartbeat: A Trustworthy Multimodal Interactive System ...

引用

IEEE Conference on Communications and Network Security (CNS)

作者： Zhang, Yan Univ Akron Dept Elect & Comp Engn Akron OH 44325 USA

ISBN: (纸本)9798350375978;9798350375961

The growing popularity of cardiac sensors has led to an increasing demand for smarter interactive systems in disease diagnosis. Various cardiac sensors have emerged as alternatives for heartbeat monitoring and cardiac disease diagnosis, but existing single -modal systems often exhibit biases in real -world applications. In this paper, we propose TrustHeartbeat, a multi modal interactive system that fuses multiple cardiac signals to recognize and identify heartbeat patterns, offering more robust and convenient interaction. TrustHeartbeat is built on a Multi modal masked autoencoder (Mu1tiMAE), capable of reconstructing multi-modal heartbeat signals even when some modalities are unavailable. We further enhance the Mu1tiMAE model by replacing the decoder with a global classifier for classification tasks. TrustHeartbeat fuses embeddings from multiple modalities for downstream tasks, such as heartbeat classification and user identification. A prototype of TrustHeartbeat was implemented and evaluated, with extensive experiments demonstrating its effectiveness. The system achieved Fl -scores of 85% for user identification and 92% for heartbeat recognition, maintaining robustness across different experimental settings.

关键词： masked autoencoder Multimodal Trustworthy

来源：评论

学校读者我要写书评

暂无评论

Self Supervised Temporal Ultrasound Reconstruction for Muscle Atrophy Evaluation 6th

Self Supervised Temporal Ultrasound Reconstruction for Muscl...

引用

6th Chinese Conference on Pattern Recognition and Computer Vision (PRCV)

作者： Zhang, Yue Du, Getao Zhan, Yonghua Guo, Kaitai Zheng, Yang Guo, Jianzhong Chen, Xiaoping Liang, Jimin Xidian Univ Sch Elect Engn Xian 710071 Shaanxi Peoples R China Xidian Univ Sch Life Sci & Technol Xian 710071 Shaanxi Peoples R China Shanxi Normal Univ Sch Phys & Informat Technol Xian 710071 Shaanxi Peoples R China China Astronaut Res & Training Ctr Natl Key Lab Human Factors Engn Beijing Peoples R China

ISBN: (纸本)9789819985456;9789819985463

Muscle atrophy is a widespread disease that can reduce quality of life and increase morbidity and mortality. The development of noninvasive method to evaluate muscle atrophy is of great practical value. However, obtaining accurate criteria for the evaluation of muscle atrophy under non-invasive conditions is extremely difficult. This paper proposes a self-supervised temporal ultrasound reconstruction method based on masked autoencoder to explore the dynamic process of muscle atrophy. A score-position embedding is designed to realize the quantitative evaluation of muscle atrophy. Ultrasound images of the hind limb muscle of six macaque monkeys were acquired consecutively during 38 days of head-down bed rest experiments. Given an ultrasound image sequence, an asymmetric encoder-decoder structure is used to reconstruct the randomly masked images for the purpose of modelling the dynamic muscle atrophy process. We demonstrate the feasibility of using the position indicator as muscle atrophy score, which can be used to predict the degree of muscle atrophy. This study achieves the quantitative evaluation of muscle atrophy in the absence of accurate evaluation criteria for muscle atrophy.

关键词： masked autoencoder Muscle atrophy quantization Position embedding Ultrasound image

来源：评论

学校读者我要写书评

暂无评论

masked Scale-Recurrent Network for Incomplete Blurred Image Restoration 1

引用

32nd International Conference on Artificial Neural Networks (ICANN)

作者： Zhu, Jingzhou Chao, Wentao Yang, Dong Beijing Normal Univ Sch Artificial Intelligence Beijing Peoples R China

ISBN: (数字)9783031442230

ISBN: (纸本)9783031442223;9783031442230

Muilti-scale learning has been demonstrated to be an excellent deblurring approach in image restoration according to recent studies. It makes the optimization of the function easier to achieve the global optimum. In order to restore an image that is both incomplete and blurry, we propose a masked Scale-Recurrent Network (MSRN) in this paper, a restoration method based on multi-scale learning and an asymmetric autoencoder. It implements restoration in an end-to-end manner without any prior knowledge or other given conditions. Firstly, we process the GoPro dataset and obtain a dataset of incomplete images. And then, we perform a self-supervised reconstruction pre-training on the autoencoder, with a series of resblocks that increase the quality of the input image and improve the representation learning in the latent space. Finally, on the processed data, we train the model and finish the adjustment of the entire network. Compared with classical multi-scale learning, we introduce masks to help the model train more efficiently by focusing on essential regions of the image. It is also shown that MSRN has successful image restoration capability as well as robustness, as demonstrated in our experiments.

关键词： image restoration masked autoencoder residual network scale-recurrent network multi-scale learning

来源：评论

学校读者我要写书评

暂无评论

Self-Supervised Learning of 3D Point Clouds: A Survey 36

Self-Supervised Learning of 3D Point Clouds: A Survey

引用

36th Chinese Control and Decision Conference (CCDC)

作者： Guo, Qi Liu, Wei Zhao, Qixi Ning, Zuotao Cheng, Shuai Hu, Jun Northeastern Univ Coll Informat Sci & Engn Shenyang Peoples R China Neusoft Reach Automot Technol Ltd Shenyang Peoples R China

ISBN: (纸本)9798350387780;9798350387797

The use of 3D point cloud is widespread in robotics and autonomous driving systems. With the development of deep learning, an increasing number of models are being proposed to address tasks in 3D point cloud processing, including shape classification and 3D object detection. However, training these models requires large amounts of labeled data, which is expensive to obtain. Therefore, self-supervised learning methods for 3D point clouds have recently gained significant attention, which train the model with unlabeled data by designing pretext tasks. This paper aims to review these methods. Based on the pretext tasks designed, we divide those more than 30 existing methods into three categories: reconstruction-based, contrastive-based and MAE(masked autoencoder)-based methods. Then, we introduce the research motivations, implementations and characteristics of these methods one by one. Finally, two performance evaluation criteria are introduced and the performance of each selfsupervised learning method is assessed and analyzed based on these criteria.

关键词： Self-supervised learning Pre-train 3D Point clouds Contrastive learning masked autoencoder

来源：评论

学校读者我要写书评

暂无评论

Community-Guided Contrastive Learning with Anomaly-Aware Reconstruction for Anomaly Detection on Attributed Networks 29th

Community-Guided Contrastive Learning with Anomaly-Aware Rec...

引用

29th International Conference on Database Systems for Advanced Applications (DASFAA)

作者： Wang, Yang Wang, Xinye He, Chengxin Chen, Xiaocong Luo, Zhaohang Duan, Lei Zuo, Jie Sichuan Univ Sch Comp Sci Chengdu Peoples R China Nucl Power Inst China Chengdu Peoples R China

ISBN: (纸本)9789819755745;9789819755752

Anomaly detection on attributed networks is of wide practical application in many domains, such as business and cybersecurity. Typically, existing methods mainly focus on utilizing the graph neural networks (GNNs) that aggregate information from neighbors to learn node representations for detecting anomalies. However, it may ignore the information beyond the neighbors like community associations. Furthermore, roughly stacking multiple GNNs layers may lead to the over-smoothing problem, making nodes representations more similar and anomalies undistinguishable. In this paper, we propose a novel method, named CARD, to tackle these issues. Specifically, we propose different augmentation strategies to offer diverse scale information for CARD. Then, to better capture community associations, we establish a community-guided contrastive learning module that can capture different scale of structure information as well. To capture multiple attribute information and aid in anomaly detection, we design an anomaly-aware masked autoencoder, effectively making anomalies more distinguished. Extensive experiments on nine datasets show the superiority of CARD. Our code are available at https://***/scu-kdde/OAM-CARD-2024.

关键词： Anomaly detection Attributed networks Community structure masked autoencoder Unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

MSTEM: masked Spatiotemporal Event Series Modeling for Urban Undisciplined Events Forecasting 24

MSTEM: Masked Spatiotemporal Event Series Modeling for Urban...

引用

33rd ACM International Conference on Information and Knowledge Management (CIKM)

作者： Gu, Zehao Zhou, Shiyang Xiong, Yun Luo, Yang Ren, Hongrun Wang, Qiang Gao, Xiaofeng Yu, Philip Fudan Univ Sch Comp Sci Shanghai Key Lab Data Sci Shanghai Peoples R China Shanghai Jiao Tong Univ Dept Comp Sci & Engn MoE Key Lab Artificial Intelligence Shanghai Peoples R China Shanghai Ctr Meteorol Disaster Prevent Technol Shanghai Peoples R China Univ Illinois Chicago IL USA

ISBN: (纸本)9798400704369

Urban undisciplined events (UUE) are of increasing concern to urban officials because they reduce the quality of life and cause societal disorder. How to accurately predict future occurrences is a key point in preventing these events. However, existing supervised methods struggle to perform well on sparse UUEs while self-supervised MAE-based methods adopt a traditional random masking strategy which leads to limited performance on UUE forecasting. Fortunately, we have designed an innovative spatiotemporal masking strategy and its corresponding pre-training task called masked Spatio-Temporal Event Series Modeling (MSTEM). Through Cluster-assisted region masking, MSTEM efficiently distributes masked regions evenly among different clusters, enhancing the model's ability to capture spatial correlation and heterogeneity while addressing sparse region distribution of UUEs. Frequency-enhanced patch masking helps the model to sufficiently extract the temporal features of UUEs by reconstructing multiple views. Additionally, we propose future merge and cluster label modeling to enhance the extraction of spatiotemporal dependencies, thereby improving the performance of MSTEM on downstream prediction tasks. Experimental evaluations on four real-world datasets including crimes and disorderly conduct show that our masked autoencoder with MSTEM outperforms most of the state-of-the-art baselines.

关键词： Spatiotemporal Prediction masked autoencoder Self-supervised learning Smart City

来源：评论

学校读者我要写书评

暂无评论

DART: DEFORMABLE ANATOMY-AWARE REGISTRATION TOOLKIT FOR LUNG CT REGISTRATION WITH KEYPOINTS SUPERVISION 21

DART: DEFORMABLE ANATOMY-AWARE REGISTRATION TOOLKIT FOR LUNG...

引用

21st IEEE International Symposium on Biomedical Imaging (ISBI)

作者： Zhu, Yunzheng Zhuang, Luoting Lin, Yannan Zhang, Tengyue Tabatabaei, Hossein Aberle, Denise R. Prosper, Ashley E. Chien, Aichi Hsu, William UCLA David Geffen Sch Med Dept Radiol Sci Med & Imaging Informat Los Angeles CA 90095 USA UCLA Dept Elect & Comp Engn Samueli Sch Engn Los Angeles CA 90095 USA UCLA Med Informat Home Area Grad Programs Biosci Los Angeles CA USA UCLA Dept Comp Sci Samueli Sch Engn Los Angeles CA USA UCLA David Geffen Sch Med Dept Radiol Sci Los Angeles CA USA UCLA Dept Bioengn Samueli Sch Engn Los Angeles CA USA

ISBN: (纸本)9798350313345;9798350313338

Spatially aligning two computed tomography (CT) scans of the lung using automated image registration techniques is a challenging task due to the deformable nature of the lung. However, existing deep-learning-based lung CT registration models are not trained with explicit anatomical knowledge. We propose the deformable anatomy-aware registration toolkit (DART), a masked autoencoder (MAE)-based approach, to improve the keypoint-supervised registration of lung CTs. Our method incorporates features from multiple decoders of networks trained to segment anatomical structures, including the lung, ribs, vertebrae, lobes, vessels, and airways, to ensure that the MAE learns relevant features corresponding to the anatomy of the lung. The pretrained weights of the transformer encoder and patch embeddings are then used as the initialization for the training of downstream registration. We compare DART to existing state-of-the-art registration models. Our experiments show that DART outperforms the baseline models (Voxelmorph, ViT-V-Net, and MAE-TransRNet) in terms of target registration error of both corrField-generated keypoints with 17%, 13%, and 9% relative improvement, respectively, and bounding box centers of nodules with 27%, 10%, and 4% relative improvement, respectively. Our implementation is available at https://***/yunzhengzhu/DART.

关键词： Lung image registration masked autoencoder Anatomy-Aware Pretraining

来源：评论

学校读者我要写书评

暂无评论

SPEECH GUIDEDmasked IMAGEMODELING FOR VISUALLY GROUNDED SPEECH 49

SPEECH GUIDEDMASKED IMAGEMODELING FOR VISUALLY GROUNDED SPEE...

引用

49th IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Woo, Jongbhin Ryu, Hyeonggon Senocak, Arda Chung, Joon Son Korea Adv Inst Sci & Technol Daejeon South Korea

ISBN: (纸本)9798350344868;9798350344851

The objective of this study is to investigate the learning process of Visually Grounded Speech (VGS) models through joint learning that combines contrastive learning and masked image modeling. Typically, VGS models aim to establish audio-visual alignment between images and their spoken captions within a contrastive learning framework. Building upon this seminal concept, in this work, we explore whether visual reconstruction with the help of cross-modality can enhance alignment, given that spoken captions describe visual appearances. To achieve this, we extend the contrastive learning-based VGS models by incorporating a masked autoencoder that utilizes cross-attention in the decoder. Through this cross-modal interaction in the decoder, spoken caption features guide the model to reconstruct the masked patches and capture correspondence between the two modalities. Our findings suggest that integrating cross-modal reconstruction within the contrastive learning framework enhances audio-visual feature alignment. Consequently, our proposed method gives comparable performance to existing models that utilize prior knowledge or other modalities, such as object region proposals or Contrastive Language-Image Pretraining (CLIP).

关键词： Visually Grounded Speech Self-supervised Learning masked autoencoder Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

MAEDAY: MAE for few- and zero-shot AnomalY-Detection

引用

COMPUTER VISION AND IMAGE UNDERSTANDING 2024年 241卷

作者： Schwartz, Eli Arbelle, Assaf Karlinsky, Leonid Harary, Sivan Scheidegger, Florian Doveh, Sivan Giryes, Raja IBM Res Haifa Israel Tel Aviv Univ Tel Aviv Israel MIT IBM Watson AI Lab Cambridge MA USA

We propose using masked Auto -Encoder (MAE), a transformer model self-supervisedly trained on image inpainting, for anomaly detection (AD). Assuming anomalous regions are harder to reconstruct compared with normal regions. MAEDAY is the first image-reconstruction-based anomaly detection method that utilizes a pre-trained model, enabling its use for Few-Shot Anomaly Detection (FSAD). We also show the same method works surprisingly well for the novel tasks of Zero-Shot AD (ZSAD) and Zero-Shot Foreign Object Detection (ZSFOD), where no normal samples are available.

关键词： Anomaly-detection masked autoencoder Foreign object detection

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：