检索结果-内蒙古大学图书馆

6th IEEE International Conference on computer and Communications, ICCC 2020

作者： Liu, Guangqi Ding, Qinghai Ju, Moran Luo, Haibo Jin, Tianming He, Miao The Key Laboratory of Image Understanding and Computer Vision Shenyang110016 China Space Star Technology Company Ltd. Beijing100086 China

ISBN: (纸本)9781728186351

A novel filtering algorithm is proposed based on level set method (LSM) and linear time Euclidean distance transform (LET) algorithm in this paper, which has the property of shape retention and thus is suitable for post-processing of the initial contours for contacting instances in digital Pap image. As one of our contributions, we propose two new metrics based on the pixel-level average false positive rate and false negative rate that used by baseline method. A significant decrease in pixel-level average false positive rate (FP) by 62% can obtain by our proposed method. The result of quantitative and qualitative evaluation shows that our proposed shape retentive filtering algorithm (SRFA) can effectively filter out the false positive fragments of the initial instance contour of cervical cells from the ISBI-2014 dataset. © 2020 IEEE.

关键词： Numerical methods

来源：评论

学校读者我要写书评

暂无评论

SAR image despeckling based on edge detection and nonsubsampled second generation bandelets

引用

Journal of Systems Engineering and Electronics 2009年第3期20卷 519-526页

作者： Zhang Wenge~(1,2),Liu Fang~(1,2),Jiao Licheng~(2,3)& Gao Xinbo~(2,3) 1.School of computer Science and Technology,Xidian Univ.,Xi’an 710071,P.R.China 2.key lab.of Intelligent Perception and image understanding of Ministry of Education of China,Xi’an 710071,P.R.China 3.Inst,of Intelligent Information Processing,Xidian Univ.,Xi’an 710071,P.R.China School of Computer Science and Technology Xidian Univ. Xi'an 710071 P. R. China Key Lab. of Intelligent Perception and Image Understanding of Ministry of Education of China Xi'an 710071 P. R. China Key Lab. of Intelligent Perception and Image Understanding of Ministry of Education of China Xi'an 710071 P. R. China Inst. of Intelligent Information Processing Xidian Univ. Xi'an 710071 P. R. China

To preserve the sharp features and details of the synthetic aperture radar （SAR） image effectively when despeckling, a despeckling algorithm with edge detection in nonsubsampled second generation bandelet transform （NSBT） domain is proposed. First, the Canny operator is utilized to detect and remove edges from the SAR image. Then the NSBT which has an optimal approximation to the edges of images and a hard thresholding rule are used to approximate the details while despeckling the edge-removed image. Finally, the removed edges are added to the reconstructed image. As the edges axe detected and protected, and the NSBT is used, the proposed algorithm reaches the state-of-the-art effect which realizes both despeckling and preserving edges and details simultaneously. Experimental results show that both the subjective visual effect and the mainly objective performance indexes of the proposed algorithm outperform that of both Bayesian wavelet shrinkage with edge detection and Bayesian least square-Gaussian scale mixture （BLS-GSM）.

关键词： computer image processing synthetic aperture radar speckle edge detection nonsubsampled second generation bandelet transform Canny operator threshold shrinkage.

来源：评论

学校读者我要写书评

暂无评论

Uncertainty Quantification in Medical image Segmentation

Uncertainty Quantification in Medical Image Segmentation

引用

International Conference on computer and Communications (ICCC)

作者： Haixing Li Haibo Luo The Key Lab of Image Understanding and Computer Vision Shenyang City Liaoning Province China

ISBN: (数字)9781728186351

ISBN: (纸本)9781728186368

In medical images, the observer's manual description of different structures is very different, and it spans a wide range of various structures and pathologies. This variability (which is a characteristic of biological issues, imaging modality and expert annotators) has not been fully considered in the design of computer algorithms for medical image quantification. So far, few people predict the uncertainty of medical image segmentation. In this paper, we designed a V-shaped network to quantify the uncertainty in prostate MRI image segmentation. We have embedded a feature pyramid attention module in the backbone network, which can extract high-level semantic context information at different scales and provide a pixel-level attention to the decoder. At the same time, the module will not bring a large computational burden. In our experiments, we tested the performance of the proposed method on 55 clinical subjects.

关键词： image segmentation Uncertainty Magnetic resonance imaging Semantics Feature extraction Prediction algorithms Biomedical imaging

来源：评论

学校读者我要写书评

暂无评论

End-to-end, single-stream temporal action detection in untrimmed videos 28

End-to-end, single-stream temporal action detection in untri...

引用

28th British Machine vision Conference, BMVC 2017

作者： Buch, Shyamal Escorcia, Victor Ghanem, Bernard Fei-Fei, Li Niebles, Juan Carlos Stanford Vision and Learning Lab. Dept. of Computer Science Stanford University United States Image and Video Understanding Lab. Visual Computing Center KAUST Saudi Arabia

ISBN: (纸本)190172560X

In this work, we present a new intuitive, end-to-end approach for temporal action detection in untrimmed videos. We introduce our new architecture for Single-Stream Temporal Action Detection (SS-TAD), which effectively integrates joint action detection with its semantic sub-tasks in a single unifying end-to-end framework. We develop a method for training our deep recurrent architecture based on enforcing semantic constraints on intermediate modules that are gradually relaxed as learning progresses. We find that such a dynamic learning scheme enables SS-TAD to achieve higher overall detection performance, with fewer training epochs. By design, our single-pass network is very efficient and can operate at 701 frames per second, while simultaneously outperforming the state-of-the-art methods for temporal action detection on THUMOS’14. © 2017. The copyright of this document resides with its authors.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Dynamic thresholding networks for schizophrenia diagnosis

引用

Artificial Intelligence in Medicine 2019年 96卷 25-32页

作者： Zou, Hongliang Yang, Jian PCA Lab Key Lab of Intelligent Perception and systems for High-Dimensional Information of Ministry of Education Jiangsu Key Lab of Image and Video Understanding for Social Security School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing 210094 China

Background and objective: Functional connectivity (FC) based on resting-state functional magnetic resonance imaging (rs-fMRI) is an effective approach to describe the neural interaction between distributed brain regions. Recent progress in neuroimaging study reported that the connection between regions is time-varying, which may enhance understanding of normal cognition and alterations that result from brain disorders. However, conventional sliding window based dynamic FC (DFC) analysis has several drawbacks, including arbitrary choice of window length, inaccurate descriptor of FC, and the fact that many spurious connections were included in the fully-connected networks due to noise. This study aims to develop an effective dynamic thresholding brain networks method to diagnose schizophrenia. Methods: In this study, we proposed a time-varying window length DFC method based on dynamic time warping to construct brain functional networks. To further eliminate the influence of spurious connections caused by noise, orthogonal minimum spanning tree was applied in these networks to generate time-varying window length dynamic thresholding FC (TVWDTFC) networks. To validate the effectiveness of our proposed method, experiments were conducted on a dataset, which including 56 individuals with schizophrenia and 74 healthy controls. Results: We achieved a classification accuracy of 0.8077 (p < 0.001, permutation test) using support vector machine. Experimental results demonstrated that the proposed method outperforms several state-of-the-art approaches, which verified the effectiveness of our proposed TVWDTFC method in schizophrenia diagnosis. Additionally, we also found that the selected discriminative features were mostly distributed in frontal, parietal, and limbic area. Conclusions: The results suggest that our approach may be a promising tool for computer-aided diagnosis of schizophrenia. © 2019 Elsevier B.V.

关键词： Dynamic time warping Orthogonal minimum spanning tree rs-fMRI Schizophrenia Time-varying window length DFC

来源：评论

学校读者我要写书评

暂无评论

Cross-view Action Recognition understanding From Exocentric to Egocentric Perspective

arXiv

引用

arXiv 2023年

作者： Truong, Thanh-Dat Luu, Khoa Computer Vision and Image Understanding Lab University of Arkansas Fayetteville72701 United States

understanding action recognition in egocentric videos has emerged as a vital research topic with numerous practical applications. With the limitation in the scale of egocentric data collection, learning robust deep learning-based action recognition models remains difficult. Transferring knowledge learned from the large-scale exocentric data to the egocentric data is challenging due to the difference in videos across views. Our work introduces a novel cross-view learning approach to action recognition (CVAR) that effectively transfers knowledge from the exocentric to the selfish view. First, we present a novel geometric-based constraint into the self-attention mechanism in Transformer based on analyzing the camera positions between two views. Then, we propose a new cross-view self-attention loss learned on unpaired cross-view data to enforce the self-attention mechanism learning to transfer knowledge across views. Finally, to further improve the performance of our cross-view learning approach, we present the metrics to measure the correlations in videos and attention maps effectively. Experimental results on standard egocentric action recognition benchmarks, i.e., Charades-Ego, EPIC-Kitchens-55, and EPIC-Kitchens-100, have shown our approach’s effectiveness and state-of-the-art performance. Copyright © 2023, The Authors. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Cross-View Action Recognition understanding from Exocentric to Egocentric Perspective

SSRN

引用

SSRN 2024年

作者： Truong, Thanh-Dat Luu, Khoa Computer Vision and Image Understanding Lab University of Arkansas Fayetteville72701 United States

understanding action recognition in egocentric videos has emerged as a vital research topic with numerous practical applications. With the limitation in the scale of egocentric data collection, learning robust deep learning-based action recognition models remains difficult. Transferring knowledge learned from the large-scale exocentric data to the egocentric data is challenging due to the difference in videos across views. Our work introduces a novel cross-view learning approach to action recognition (CVAR) that effectively transfers knowledge from the exocentric to the selfish view. First, we present a novel geometric-based constraint into the self-attention mechanism in Transformer based on analyzing the camera positions between two views. Then, we propose a new cross-view self-attention loss learned on unpaired cross-view data to enforce the self-attention mechanism learning to transfer knowledge across views. Finally, to further improve the performance of our cross-view learning approach, we present the metrics to measure the correlations in videos and attention maps effectively. Experimental results on standard egocentric action recognition benchmarks, i.e., Charades-Ego, EPIC-Kitchens-55, and EPIC-Kitchens-100, have shown our approach's effectiveness and state-of-the-art performance. © 2024, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Planning with Learned Dynamic Model for Unsupervised Point Cloud Registration 30

Planning with Learned Dynamic Model for Unsupervised Point C...

引用

30th International Joint Conference on Artificial Intelligence, IJCAI 2021

作者： Jiang, Haobo Qian, Jianjun Xie, Jin Yang, Jian PCA Lab Key Lab of Intelligent Perception and Systems for High-Dimensional Information Ministry of Education Jiangsu Key Lab of Image and Video Understanding for Social Security School of Computer Science and Engineering Nanjing University of Science and Technology China

ISBN: (纸本)9780999241196

Point cloud registration is a fundamental problem in 3D computer vision. In this paper, we cast point cloud registration into a planning problem in reinforcement learning, which can seek the transformation between the source and target point clouds through trial and error. By modeling the point cloud registration process as a Markov decision process (MDP), we develop a latent dynamic model of point clouds, consisting of a transformation network and evaluation network. The transformation network aims to predict the new transformed feature of the point cloud after performing a rigid transformation (i.e., action) on it while the evaluation network aims to predict the alignment precision between the transformed source point cloud and target point cloud as the reward signal. Once the dynamic model of the point cloud is trained, we employ the cross-entropy method (CEM) to iteratively update the planning policy by maximizing the rewards in the point cloud registration process. Thus, the optimal policy, i.e., the transformation between the source and target point clouds, can be obtained via gradually narrowing the search space of the transformation. Experimental results on ModelNet40 and 7Scene benchmark datasets demonstrate that our method can yield good registration performance in an unsupervised manner. © 2021 International Joint Conferences on Artificial Intelligence. All rights reserved.

关键词： Dynamic models

来源：评论

学校读者我要写书评

暂无评论

Unified Density-Aware image Dehazing and Object Detection in Real-World Hazy Scenes 15th

Unified Density-Aware Image Dehazing and Object Detection in...

引用

15th Asian Conference on computer vision, ACCV 2020

作者： Zhang, Zhengxi Zhao, Liang Liu, Yunan Zhang, Shanshan Yang, Jian PCA Lab Key Lab of Intelligent Perception and Systems for High -Dimensional Information of Ministry of Education Jiangsu Key Lab of Image and Video Understanding for Social Security School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China

ISBN: (纸本)9783030695378

It is an important yet challenging task to detect objects on hazy images in real-world applications. The major challenge comes from low visual quality and large haze density variations. In this work, we aim to jointly solve the image dehazing and the object detection tasks in real hazy scenarios by using haze density as prior knowledge. Our proposed Unified Dehazing and Detection (UDnD) framework consists of three parts: a residual-aware haze density classifier, a density-aware dehazing network, and a density-aware object detector. First, the classifier exploits the residuals of hazy images to accurately predict density levels, which provide rich domain knowledge for the subsequent two tasks. Then, we design respectively a High-Resolution Dehazing Network (HRDN) and a Faster R-CNN-based multi-domain object detector to leverage the extracted density information and tackle hazy object detection. Experiments demonstrate that UDnD performs favorably against other methods for object detection in real-world hazy scenes. Also, HRDN achieves better results than state-of-the-art dehazing methods in terms of PSNR and SSIM. Hence, HRDN can conduct haze removal effectively, based on which UDnD is able to provide high-quality detection results. © 2021, Springer Nature Switzerland AG.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Multi-view Action Recognition via Directed Gromov-Wasserstein Discrepancy

arXiv

引用

arXiv 2024年

作者： Nguyen, Hoang-Quan Truong, Thanh-Dat Luu, Khoa Computer Vision and Image Understanding Lab University of Arkansas FayettevilleAR72701 United States

Action recognition has become one of the popular research topics in computer vision. There are various methods based on Convolutional Networks and self-attention mechanisms as Transformers to solve both spatial and temporal dimensions problems of action recognition tasks that achieve competitive performances. However, these methods lack a guarantee of the correctness of the action subject that the models give attention to, i.e., how to ensure an action recognition model focuses on the proper action subject to make a reasonable action prediction. In this paper, we propose a multi-view attention consistency method that computes the similarity between two attentions from two different views of the action videos using Directed Gromov-Wasserstein Discrepancy. Furthermore, our approach applies the idea of Neural Radiance Field to implicitly render the features from novel views when training on single-view datasets. Therefore, the contributions in this work are three-fold. Firstly, we introduce the multi-view attention consistency to solve the problem of reasonable prediction in action recognition. Secondly, we define a new metric for multi-view consistent attention using Directed Gromov-Wasserstein Discrepancy. Thirdly, we built an action recognition model based on Video Transformers and Neural Radiance Fields. Compared to the recent action recognition methods, the proposed approach achieves state-of-the-art results on three large-scale datasets, i.e., Jester, Something-Something V2, and Kinetics-400. © 2024, CC BY.

关键词： Large datasets

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：