.png){kind=icon}
.png){kind=icon}
Return
Multichannel Orthogonal Transform-Based Perceptron Layers for Efficient ResNets
.png){kind=icon}
PRE
AI
DOI：10.1109/TNNLS.2024.3384316.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In this article, we propose a set of transform-based neural network layers as an alternative to the 3 x 3 Conv2D layers in convolutional neural networks (CNNs). The proposed layers can be implemented based on orthogonal transforms, such as the discrete cosine transform (DCT), Hadamard transform (HT), and biorthogonal block wavelet transform (BWT). Furthermore, by taking advantage of the convolution theorems, convolutional filtering operations are performed in the transform domain using elementwise multiplications. Trainable soft-thresholding layers, that remove noise in the transform domain, bring nonlinearity to the transform domain layers. Compared with the Conv2D layer, which is spatial-agnostic and channel-specific, the proposed layers are location-specific and channel-specific. Moreover, these proposed layers reduce the number of parameters and multiplications significantly while improving the accuracy results of regular ResNets on the ImageNet-1K classification task. Furthermore, they can be inserted with a batch normalization (BN) layer before the global average pooling layer in the conventional ResNets as an additional layer to improve classification accuracy.
Keywords:
Transforms
Convolution
Discrete cosine transforms
Feature extraction
Discrete Fourier transforms
Filters
Wavelet transforms
Convolution theorem
image classification
soft thresholding
transform-based convolutional layer
Journal
IF:
8.9
Papers:
7.5K
Citations:
7.2W
