Filters








5,276 Hits in 3.1 sec

Generative networks as inverse problems with fractional wavelet scattering networks [article]

Jiasong Wu, Jing Zhang, Fuzhi Wu, Youyong Kong, Guanyu Yang, Lotfi Senhadji, Huazhong Shu
2020 arXiv   pre-print
In order to solve or alleviate the synchronous training difficult problems of GANs and VAEs, recently, researchers propose Generative Scattering Networks (GSNs), which use wavelet scattering networks (  ...  In order to further improve the quality of generated images while keep the advantages of GSNs, this paper proposes Generative Fractional Scattering Networks (GFRSNs), which use more expressive fractional  ...  Fractional wavelet Scattering Networks (FrScatNets) In this subsection, the Fractional wavelet Scattering Networks (FrScatNets) [70] will be briefly introduced.  ... 
arXiv:2007.14177v1 fatcat:3djok4gbajdt7a2ak3g2a4segq

Fractional Wavelet Scattering Network and Applications

Li Liu, Jiasong Wu, Dengwang Li, Lotfi Senhadji, Huazhong Shu
2018 IEEE Transactions on Biomedical Engineering  
Objective: The present study introduces a fractional wavelet scattering network (FrScatNet), which is a generalized translation invariant version of the classical wavelet scattering network (ScatNet).  ...  Methods: In our approach, the FrScatNet is constructed based on the fractional wavelet transform (FRWT).  ...  The obtained FrScatNet generalizes the classical wavelet scattering network (ScatNet) from the scattering domain to the fractional scattering domain.  ... 
doi:10.1109/tbme.2018.2850356 pmid:29993504 fatcat:27si3nj5qne3zk3mrht26az3vy

Learnable Wavelet Scattering Networks: Applications to Fault Diagnosis of Analog Circuits and Rotating Machinery

Varun Khemani, Michael H. Azarian, Michael G. Pecht
2022 Electronics  
The learnable wavelet scattering networks are developed using the genetic algorithm-based optimization of second-generation wavelet transform operators.  ...  Wavelet scattering networks, which are fixed time–frequency representations based on existing wavelets, are modified to be learnable so that they can learn features that are optimal for fault diagnosis  ...  Learnable wavelet scattering networks are developed using the genetic-algorithm-based optimization of second-generation wavelet transform operators.  ... 
doi:10.3390/electronics11030451 fatcat:zyn4ab4cyrd3tg7h4dexamwhke

Features Reduction using Wavelet and Discriminative Common Vector and Recognizing Faces using RBF

T. Kathirvalavakumar, J. Jebakumari Beulah Vasanthi
2013 International Journal of Computer Applications  
The discriminative common vectors are extracted using the within-class scatter matrix method from the wavelet coefficients.  ...  General Terms Face Recognition  ...  [15] have presented a frame work based on a combination of Gabor wavelets and General Discriminant Analysis for face identification and verification. Guang Dai et al.  ... 
doi:10.5120/12884-9796 fatcat:kbo6pa7ferbrbdrxr6znx3qil4

Multi-Resolution Dual-Tree Wavelet Scattering Network for Signal Classification [article]

Amarjot Singh, Nick Kingsbury
2017 arXiv   pre-print
This paper introduces a Deep Scattering network that utilizes Dual-Tree complex wavelets to extract translation invariant representations from an input signal.  ...  The computationally efficient Dual-Tree wavelets decompose the input signal into densely spaced representations over scales.  ...  The test set generalization error of the proposed Deep DTCWT multi-resolution scattering network is reported on each Dataset (Table. 1) and compared with the scattering network proposed by Mallat et  ... 
arXiv:1702.03345v1 fatcat:j7r37ezywbcgngl6tllu74jrni

Molecular Graph Generation via Geometric Scattering [article]

Dhananjay Bhaskar, Jackson D. Grady, Michael A. Perlmutter, Smita Krishnaswamy
2021 arXiv   pre-print
Graph neural networks (GNNs) have been used extensively for addressing problems in drug design and discovery.  ...  We guide the latent representation of an autoencoder by capturing graph structure information with the geometric scattering transform and apply penalties that structure the representation also by molecular  ...  Geometric Scattering The scattering transform, originally introduced in Mallat [2012] for Euclidean data is a wavelet-based, feed-forward network which produces a latent-space representation of an input  ... 
arXiv:2110.06241v1 fatcat:r3lzthkivzbtfeddaffefdtnt4

Target Classification in Synthetic Aperture Radar Images Using Quantized Wavelet Scattering Networks

Raghu G. Raj, Maxine R. Fox, Ram M. Narayanan
2021 Sensors  
However, large CNNs and wavelet scattering networks (WSNs), which share similar properties, have extensive memory requirements and are not readily extendable to other datasets and architectures—and especially  ...  Thus, the WSN-based quantization studies performed in this investigation provide a good benchmark and important guidance for the design of quantized neural networks architectures for target classification  ...  and statically generated levels of the PDF-based quantile scales.  ... 
doi:10.3390/s21154981 fatcat:2nbhyctkpbavjgnsv52x2fitem

Deep Adaptive Wavelet Network

Maria Ximena Bastidas Rodriguez, Adrien Gruson, Luisa F. Polania, Shin Fujieda, Flavio Prieto Ortiz, Kohei Takayama, Toshiya Hachisuka
2020 2020 IEEE Winter Conference on Applications of Computer Vision (WACV)  
By using the lifting scheme, it is possible to generate a wavelet representation and design a network capable of learning wavelet coefficients in an endto-end form.  ...  the deep neural network design.  ...  [23] proposed a hybrid network which replaces the first layers of ResNet by a wavelet scattering network.  ... 
doi:10.1109/wacv45572.2020.9093580 dblp:conf/wacv/Bastidas-Rodriguez20 fatcat:yhnvxecspfapzeut6pouk7hd5i

Deep Adaptive Wavelet Network [article]

Maria Ximena Bastidas Rodriguez, Adrien Gruson, Luisa F. Polania, Shin Fujieda, Flavio Prieto Ortiz, Kohei Takayama, Toshiya Hachisuka
2019 arXiv   pre-print
By using the lifting scheme, it is possible to generate a wavelet representation and design a network capable of learning wavelet coefficients in an end-to-end form.  ...  the deep neural network design.  ...  [22] proposed a hybrid network which replaces the first layers of ResNet by a wavelet scattering network.  ... 
arXiv:1912.05035v1 fatcat:mhrhzts3bzct3kfatwrtxmbaxu

Glioma Grade Predictions using Scattering Wavelet Transform-Based Radiomics [article]

Qijian Chen, Lihui Wang, Li Wang, Zeyu Deng, Jian Zhang, Yuemin Zhu
2019 arXiv   pre-print
In this paper, we present a novel scattering wavelet-based radiomics method to predict noninvasively and accurately the glioma grades.  ...  The wavelet scattering-based features and traditional radiomics features were firstly extracted from both intratumoral and peritumoral regions respectively.  ...  IV.DISCUSSION The proposed glioma grade prediction method is based on local invariant features extracted from wavelet scattering network instead of traditional features.  ... 
arXiv:1905.09589v1 fatcat:tvgalhs5c5em3dynul5l5taqe4

CNN-Based Invertible Wavelet Scattering for the Investigation of Diffusion Properties of the In Vivo Human Heart in Diffusion Tensor Imaging [article]

Zeyu Deng, Lihui Wang, Zixiang Kuai, Qijian Chen, Xinyu Cheng, Feng Yang, Jie Yang, Yuemin Zhu
2019 arXiv   pre-print
The method is based on an invertible Wavelet Scattering achieved by means of Convolutional Neural Network (WSCNN).  ...  spatial DW images by performing an inverse wavelet scattering transform achieved using CNN.  ...  Since wavelet scattering transform is not exactly invertible [27] , we propose to use a network based on CNN layers and residual block to achieve the inverse wavelet scattering transform by mapping multiple  ... 
arXiv:1912.07776v1 fatcat:j7i7yl65dfbjvdswfdwyvexwli

Steerable Wavelet Scattering for 3D Atomic Systems with Application to Li-Si Energy Prediction [article]

Xavier Brumwell and Paul Sinz and Kwang Jin Kim and Yue Qi and Matthew Hirn
2019 arXiv   pre-print
A general machine learning architecture is introduced that uses wavelet scattering coefficients of an inputted three dimensional signal as features.  ...  The scattering coefficients inherit from the wavelets invariance to translations and rotations.  ...  Section 6 contains some concluding remarks. 2 Equivarient scattering networks with steerable wavelet filters 2.1 Periodic steerable wavelets We introduce a general class of 3-dimensional steerable  ... 
arXiv:1812.02320v2 fatcat:zejpbpo7bzbq7ftd2uq6vvxbkm

Glioma Grade Prediction Using Wavelet Scattering-Based Radiomics

Qijian Chen, Lihui Wang, Li Wang, Zeyu Deng, Jian Zhang, Yuemin Zhu
2020 IEEE Access  
We present a novel wavelet scattering-based radiomic method to predict noninvasively and accurately the glioma grades.  ...  The method consists of wavelet scattering feature extraction, dimensionality reduction, and glioma grade prediction.  ...  FIGURE 2 . 2 Overall workflow of glioma grade prediction based on traditional features and wavelet scattering features. FIGURE 3 . 3 Scheme of the second-order wavelet scattering network.  ... 
doi:10.1109/access.2020.3000895 fatcat:6h5iqosgizdcvdm7kppkg2c5yq

Taking ROCKET on an Efficiency Mission: Multivariate Time Series Classification with LightWaveS [article]

Leonardos Pantiskas, Kees Verstoep, Mark Hoogendoorn, Henri Bal
2022 arXiv   pre-print
Specifically, utilizing wavelet scattering transformation and distributed feature selection, we manage to create a solution that employs just 2.5% of the ROCKET features, while achieving accuracy comparable  ...  An MTSC model that has attracted attention recently is ROCKET, based on random convolutional kernels, both because of its very fast training process and its state-of-the-art accuracy.  ...  Algorithm Based on the above observations, we improve the approach and reach the crux of LightWaveS: Lightweight Wavelet Scattering based on random wavelets.  ... 
arXiv:2204.01379v3 fatcat:ri7dbz6pk5g4tex2lcyog5bdjy

Spatio-Temporal Graph Scattering Transform [article]

Chao Pan, Siheng Chen, Antonio Ortega
2021 arXiv   pre-print
It performs iterative applications of spatio-temporal graph wavelets and nonlinear activation functions, which can be viewed as a forward pass of spatio-temporal graph convolutional networks without training  ...  Our proposed spatio-temporal graph scattering transform (ST-GST) extends traditional scattering transforms to the spatio-temporal domain.  ...  Geometric scattering wavelets. Geometric scattering wavelet filter bank (Gao et al., 2019) contains a set of filters based on lazy random walk matrix.  ... 
arXiv:2012.03363v3 fatcat:cb7wn5xv5fhfli6bumvu3lant4
« Previous Showing results 1 — 15 out of 5,276 results