Wavelet Convolutional Neural Networks

Fujieda, Shin; Takayama, Kohei; Hachisuka, Toshiya

doi:10.48550/arxiv.1805.08620

Cited by 34 publications

(35 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The conventional 2D CNN can be considered a limited version of a multi-resolution CNN that can consider both spectral and spatial information [26]. Previous works have been successful in establishing the convolution and pooling function in a 2D CNN as filtering and downsampling [27].…”

Section: Spectralnetmentioning

confidence: 99%

SpectralNET: Exploring Spatial-Spectral WaveletCNN for Hyperspectral Image Classification

Chakraborty,

Trehan

2021

Preprint

View full text Add to dashboard Cite

Hyperspectral Image (HSI) classification using Convolutional Neural Networks (CNN) is widely found in the current literature. Approaches vary from using SVMs to 2D CNNs, 3D CNNs, 3D-2D CNNs. Besides 3D-2D CNNs and FuSENet, the other approaches do not consider both the spectral and spatial features together for HSI classification task, thereby resulting in poor performances. 3D CNNs are computationally heavy and are not widely used, while 2D CNNs do not consider multiresolution processing of images, and only limits itself to the spatial features. Even though 3D-2D CNNs try to model the spectral and spatial features their performance seems limited when applied over multiple dataset. In this article, we propose SpectralNET, a wavelet CNN, which is a variation of 2D CNN for multi-resolution HSI classification. A wavelet CNN uses layers of wavelet transform to bring out spectral features. Computing a wavelet transform is lighter than computing 3D CNN. The spectral features extracted are then connected to the 2D CNN which bring out the spatial features, thereby creating a spatialspectral feature vector for classification. Overall a better model is achieved that can classify multi-resolution HSI data with high accuracy. Experiments performed with SpectralNET on benchmark dataset, i.e. Indian Pines, University of Pavia, and Salinas Scenes confirm the superiority of proposed SpectralNET with respect to the state-of-the-art methods. The code is publicly available in https://github.com/tanmay-ty/SpectralNET.

show abstract

Section: Spectralnetmentioning

confidence: 99%

SpectralNET: Exploring Spatial-Spectral WaveletCNN for Hyperspectral Image Classification

Chakraborty,

Trehan

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Ejbali et al [12] convolve with optimally selected wavelets from a pre-defined filter bank. Wavelets are also applied to inputs of standard convolutional layers in a preprocessing fashion [13]. In [41] authors present a framework for convolutional weight modulation, achieving enhanced filters with binarised weights.…”

Section: Previous Workmentioning

confidence: 99%

Learnable Gabor modulated complex-valued networks for orientation robustness

Richards,

Paiement,

Xie

et al. 2020

Preprint

View full text Add to dashboard Cite

Robustness to transformation is desirable in many computer vision tasks, given that input data often exhibits pose variance within classes. While translation invariance and equivariance is a documented phenomenon of CNNs, sensitivity to other transformations is typically encouraged through data augmentation. We investigate the modulation of complex valued convolutional weights with learned Gabor filters to enable orientation robustness. With Gabor modulation, the designed network is able to generate orientation dependent features free of interpolation with a single set of rotation-governing parameters. Moreover, by learning rotation parameters alongside traditional convolutional weights, the representation space is not constrained and may adapt to the exact input transformation. We present Learnable Convolutional Gabor Networks (LCGNs), that are parameter-efficient and offer increased model complexity while keeping backpropagation simple. We demonstrate that learned Gabor modulation utilising an end-to-end complex architecture enables rotation invariance and equivariance on MNIST and a new dataset of simulated images of galactic cirri.

show abstract

“…This has been the starting point for a number of wavelet-inspired CNNs; see e.g. (Wiatowski & Bölcskei, 2017;Fujieda et al, 2018;Williams & Li, 2018) and the references therein. Usually they exploit the spectral information or the multiscale nature of wavelets.…”

Section: Related Workmentioning

confidence: 99%

Translating Diffusion, Wavelets, and Regularisation into Residual Networks

Alt¹,

Weickert²,

Peter³

2020

Preprint

View full text Add to dashboard Cite

Convolutional neural networks (CNNs) often perform well, but their stability is poorly understood. To address this problem, we consider the simple prototypical problem of signal denoising, where classical approaches such as nonlinear diffusion, wavelet-based methods and regularisation offer provable stability guarantees. To transfer such guarantees to CNNs, we interpret numerical approximations of these classical methods as a specific residual network (ResNet) architecture. This leads to a dictionary which allows to translate diffusivities, shrinkage functions, and regularisers into activation functions, and enables a direct communication between the four research communities. On the CNN side, it does not only inspire new families of nonmonotone activation functions, but also introduces intrinsically stable architectures for an arbitrary number of layers.

show abstract

Wavelet Convolutional Neural Networks

Cited by 34 publications

References 33 publications

SpectralNET: Exploring Spatial-Spectral WaveletCNN for Hyperspectral Image Classification

SpectralNET: Exploring Spatial-Spectral WaveletCNN for Hyperspectral Image Classification

Learnable Gabor modulated complex-valued networks for orientation robustness

Translating Diffusion, Wavelets, and Regularisation into Residual Networks

Contact Info

Product

Resources

About