287 Hits in 1.5 sec

Cascaded Subpatch Networks for Effective CNNs [article]

Xiaoheng Jiang, Yanwei Pang, Manli Sun, Xuelong Li
2016 arXiv   pre-print
Taking the output of one subpatch network as input, we further repeat constructing subpatch networks until the output contains only one neuron in spatial domain.  ...  These subpatch networks form a new network called Cascaded Subpatch Network (CSNet). The feature layer generated by CSNet is called csconv layer.  ...  A deep neural network can be obtained by stacking multiple csconv layers. For clarity, in the rest of this paper, the overall deep network containing multiple csconv layers is called a CSNet.  ... 
arXiv:1603.00128v1 fatcat:o5flelcrsrbxve4jr52gmkv72a

Multi-Scale Deep Compressive Imaging [article]

Thuong Nguyen Canh, Byeungwoo Jeon
2020 arXiv   pre-print
From this perspective, it would be easier for the network to learn multi-scale features with a multi-scale sampling architecture.  ...  While multi-scale has shown superior performance over single-scale, research in DCI has been limited to single-scale sampling.  ...  ] , a signal is decomposed into multiple layers using a set of Gaussian smoothing filters.  ... 
arXiv:2008.00802v1 fatcat:h55v64hzkfgd5byqyrlzvfff2q

Discriminative Feature Learning for Unsupervised Video Summarization

Yunjae Jung, Donghyeon Cho, Dahun Kim, Sanghyun Woo, In So Kweon
For the second problem, we design a novel two-stream network named Chunk and Stride Network (CSNet) that utilizes local (chunk) and global (stride) temporal view on the video features.  ...  Our CSNet gives better summarization results for long-length videos compared to the existing methods. In addition, we introduce an attention mechanism to handle the dynamic information in videos.  ...  The weights of the network are randomly initialized. M in CSNet is experimentally picked as 4. We implement our method using Pytorch.  ... 
doi:10.1609/aaai.v33i01.33018537 fatcat:vjs2ssy47fgzdmouwnud3kb7lm

Multi-Scale Deep Compressive Sensing Network [article]

Thuong Nguyen Canh, Byeungwoo Jeon
2018 arXiv   pre-print
In this paper, we propose a multi-scale DCS convolutional neural network (MS-DCSNet) in which we convert image signal using multiple scale-based wavelet transform, then capture it through convolution block  ...  With joint learning of sampling and recovery, the deep learning-based compressive sensing (DCS) has shown significant improvement in performance and running time reduction.  ...  We review related works in multi-scale sampling and deep learning CS in Section 2. Section 3 proposes our multi-scale DCS network (MS-DCSNet) with multiple phases of training.  ... 
arXiv:1809.05717v2 fatcat:nbll4vfjofhfrjuk4hivy32b7e

High-speed Millimeter-wave 5G/6G Image Transmission via Artificial Intelligence [article]

Shaolin Liao, Lu Ou
2020 arXiv   pre-print
Specifically, we have developed a Dictionary Learning Compressed Sensing neural Network (DL-CSNet) to realize three key functionalities: 1) to learn the dictionary basis of the images for transmission;  ...  Artificial Intelligence (AI) has been used to jointly optimize a mmWave Compressed Sensing (CS) for high-speed 5G/6G image transmission.  ...  In this paper, we *Shaolin Liao is the corresponding author. apply Artificial Intelligence (AI) and develop the Dictionary Learning Compressed Sensing neural Network (DL-CSNet) for the optimized mmWave  ... 
arXiv:2007.03153v1 fatcat:ujko6tcabnanrkyyl5faquyek4

Research computer networks and their interconnection

L. Landweber, D. Jennings, I. Fuchs
1986 IEEE Communications Magazine  
A rich computing and communications environment, providing access to a wide variety of computational resources and communications services, can significantly improve the productivity of engineers  ...  The DOD RFC 822 [13] electronic mail format standard is also used, so no translation is required in relaying messages between Phonenet and the other CSNET component networks. 9 Cypress-a New CSNET Network  ...  EARN-European Academic Research Network EARN is a clone of the U.S. BITNET in that i t uses the IBM RSCS protocol and is constructed using leased lines.  ... 
doi:10.1109/mcom.1986.1093103 fatcat:idu552pgzvdwdfahuik2wzzbau

Discriminative Feature Learning for Unsupervised Video Summarization [article]

Yunjae Jung, Donghyeon Cho, Dahun Kim, Sanghyun Woo, In So Kweon
2018 arXiv   pre-print
For the second problem, we design a novel two-stream network named Chunk and Stride Network (CSNet) that utilizes local (chunk) and global (stride) temporal view on the video features.  ...  Our CSNet gives better summarization results for long-length videos compared to the existing methods. In addition, we introduce an attention mechanism to handle the dynamic information in videos.  ...  However, if a test set is randomly selected, there may be video that is not used in the test set or is used multiple times in duplicate, making it difficult to evaluate fairly.  ... 
arXiv:1811.09791v1 fatcat:d3vwgetc3bh55lojouijfk43mi

Global Sensing and Measurements Reuse for Image Compressed Sensing [article]

Zi-En Fan, Feng Lian, Jia-Ni Quan
2022 arXiv   pre-print
They ignore there are low, mid, and high-level features in the network and all of them are essential for high-quality reconstruction.  ...  However, existing methods obtain measurements only from partial features in the network and use them only once for image reconstruction.  ...  Acknowledgement: This research was supported by the National Natural Science Foundation of China (62173266).  ... 
arXiv:2206.11629v1 fatcat:zjemnmf6nvadvfibbfe32b462a

Highly Efficient Salient Object Detection with 100K Parameters [article]

Shang-Hua Gao, Yong-Qiang Tan, Ming-Ming Cheng, Chengze Lu, Yunpeng Chen, Shuicheng Yan
2020 arXiv   pre-print
In this paper, we aim to relieve the contradiction between computation cost and model performance by improving the network efficiency to a higher degree.  ...  The effective dynamic weight decay scheme stably boosts the sparsity of parameters during training, supports learnable number of channels for each scale in gOctConv, allowing 80% of parameters reduce with  ...  When utilizing the recently proposed Res2Net as the backbone network, the performance is further boosted. Performance of CSNet with learnable channels in gOctConv.  ... 
arXiv:2003.05643v2 fatcat:3uudr4ike5dgjghim4tglrnblq

Hierarchical distillation for image compressive sensing reconstruction

Bokyeung Lee, Bonhwa Ku, Wanjin Kim, Hanseok Ko
2021 Electronics Letters  
CS models combining traditional optimisation-based CS methods and deep learning have been used to improve image reconstruction performance.  ...  In this letter, a deep learning-based CS model incorporating hierarchical knowledge distillation to improve image reconstruction even at varied sample ratios.  ...  Acknowledgement: This work was funded by the Agency for Defense Development of Korea under Grant 190005DD.  ... 
doi:10.1049/ell2.12284 fatcat:42xxjxgpevdfxiuizsezce2n7q

Flame Wars on Worldnet: Early Constructions of the International User [chapter]

Christopher Leslie
2016 IFIP Advances in Information and Communication Technology  
Moving forward to CSNET, one can also see a strong insistence that the network provide connectivity beyond the United States.  ...  Contrary to those who might tell the history of the Internet as a story of a technology that was first perfected by the military, adapted by U.S. academics and then brought to the rest of the world in  ...  Grateful acknowledgement is made to Aye Muang who, as an undergraduate research student, conducted the evaluation of Human-Nets.  ... 
doi:10.1007/978-3-319-49463-0_9 fatcat:bchnvsskpzbf7hcpqhesogdixm

CSNet: Cascade stereo matching network using multi‐information cost volume

XiaoTao Shao, Wen Zhang, MingKun Guo, SiQi Guo, ManYi Qian
2021 IET Intelligent Transport Systems  
For most stereo matching networks, cost volume plays a crucial role in the accuracy of disparity maps.  ...  To refine the disparity map and enhance the semantics of small objects, a cascade stereo network called CSNet is proposed, with a dilation feature fusion unit (DFFU) to calculate and integrate disparity  ...  The first stage of the network, composed of Dispnet, was used to generate the initial disparity, and the second stage of the network was used to learn the correction. Liang et al.  ... 
doi:10.1049/itr2.12050 fatcat:jkeftlzupfhyxdeaqhb4goulxe

Optimal Combination of Image Denoisers [article]

Joon Hee Choi, Omar Elgendy, Stanley H. Chan
2019 arXiv   pre-print
; (2) A deep neural network to estimate the mean squared error (MSE) of denoised images without needing the ground truths; (3) An image boosting procedure using a deep neural network to improve contrast  ...  The proposed framework, called the Consensus Neural Network (CsNet), introduces three new concepts in image denoising: (1) A provably optimal procedure to combine the denoised outputs via convex optimization  ...  The advantage of CsNet relative to other class-aware neural network denoisers is that we allow combination of multiple denoisers.  ... 
arXiv:1711.06712v4 fatcat:oftvkqovivalvpvknm4jurd6ua

MD-Recon-Net: A Parallel Dual-Domain Convolutional Neural Network for Compressed Sensing MRI [article]

Maosong Ran, Wenjun Xia, Yongqiang Huang, Zexin Lu, Peng Bao, Yan Liu, Huaiqiang Sun, Jiliu Zhou, Yi Zhang
2020 arXiv   pre-print
The simulated experimental results show that the proposed method not only achieves competitive visual effects to several state-of-the-art methods, but also outperforms other DL-based methods in terms of  ...  In this paper, inspired by deep learning's (DL's) fast inference and excellent end-to-end performance, we propose a novel cascaded convolutional neural network called MD-Recon-Net to facilitate fast and  ...  For examples, coil compression algorithms can be used at first to combine the multiple coil k-space data or a coil data fusion layer can be added to current network.  ... 
arXiv:1910.10392v2 fatcat:duu7ibfj5zfmdg5qk5a6zzk7km

Knee Cartilage Defect Assessment by Graph Representation and Surface Convolution [article]

Zixu Zhuang, Liping Si, Sheng Wang, Kai Xuan, Xi Ouyang, Yiqiang Zhan, Zhong Xue, Lichi Zhang, Dinggang Shen, Weiwu Yao, Qian Wang
2022 arXiv   pre-print
In this way, many attempts have been made on knee cartilage defect assessment by applying convolutional neural networks (CNNs) to knee MRI.  ...  Then, guided by the cartilage graph representation, we design a non-Euclidean deep learning network with the self-attention mechanism, to extract cartilage features in the local and global, and to derive  ...  The continuity of the cartilage across multiple slices should also be considered.  ... 
arXiv:2201.04318v1 fatcat:jyy2vozx4zfctkv756hamzohfq
« Previous Showing results 1 — 15 out of 287 results