Corpus ID: 236428985

Log-Polar Space Convolution for Convolutional Neural Networks

@article{Su2021LogPolarSC,
  title={Log-Polar Space Convolution for Convolutional Neural Networks},
  author={Bing Su and Ji-Rong Wen},
  journal={ArXiv},
  year={2021},
  volume={abs/2107.11943}
}
  • Bing Su, Ji-Rong Wen
  • Published 2021
  • Computer Science
  • ArXiv
Convolutional neural networks use regular quadrilateral convolution kernels to extract features. Since the number of parameters increases quadratically with the size of the convolution kernel, many popular models use small convolution kernels, resulting in small local receptive fields in lower layers. This paper proposes a novel log-polar space convolution (LPSC) method, where the convolution kernel is elliptical and adaptively divides its local receptive field into different regions according… Expand

Figures and Tables from this paper

References

SHOWING 1-10 OF 51 REFERENCES
Active Convolution: Learning the Shape of Convolution for Image Classification
  • Yunho Jeon, Junmo Kim
  • Computer Science, Mathematics
  • 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2017
TLDR
The proposed ACU is a generalization of convolution, it can define not only all conventional convolutions, but also convolutions with fractional pixel coordinates, which provides greater freedom to form CNN structures. Expand
Scale-Adaptive Convolutions for Scene Parsing
TLDR
The proposed scale-adaptive convolutions are not only differentiable to learn the convolutional parameters and scale coefficients in an end-to-end way, but also of high parallelizability for the convenience of GPU implementation. Expand
Perspective-Adaptive Convolutions for Scene Parsing
TLDR
This work proposes perspective-adaptive convolutions to acquire receptive fields of flexible sizes and shapes during scene parsing through adding a new perspective regression layer, which can dynamically infer the position- Adaptive perspective coefficient vectors utilized to reshape the convolutional patches. Expand
Drop an Octave: Reducing Spatial Redundancy in Convolutional Neural Networks With Octave Convolution
TLDR
This work proposes to factorize the mixed feature maps by their frequencies, and design a novel Octave Convolution (OctConv) operation to store and process feature maps that vary spatially “slower” at a lower spatial resolution reducing both memory and computation cost. Expand
Understanding the Effective Receptive Field in Deep Convolutional Neural Networks
TLDR
The notion of an effective receptive fieldsize is introduced, and it is shown that it both has a Gaussian distribution and only occupies a fraction of the full theoretical receptive field size. Expand
DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs
TLDR
This work addresses the task of semantic image segmentation with Deep Learning and proposes atrous spatial pyramid pooling (ASPP), which is proposed to robustly segment objects at multiple scales, and improves the localization of object boundaries by combining methods from DCNNs and probabilistic graphical models. Expand
Xception: Deep Learning with Depthwise Separable Convolutions
  • François Chollet
  • Computer Science, Mathematics
  • 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2017
TLDR
This work proposes a novel deep convolutional neural network architecture inspired by Inception, where Inception modules have been replaced with depthwise separable convolutions, and shows that this architecture, dubbed Xception, slightly outperforms Inception V3 on the ImageNet dataset, and significantly outperforms it on a larger image classification dataset. Expand
PSConv: Squeezing Feature Pyramid into One Compact Poly-Scale Convolutional Layer
TLDR
The proposed convolution operation, named Poly-Scale Convolution (PSConv), mixes up a spectrum of dilation rates and tactfully allocate them in the individual convolutional kernels of each filter regarding a single convolutionAL layer. Expand
Very Deep Convolutional Networks for Large-Scale Image Recognition
TLDR
This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers. Expand
Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation
TLDR
This work extends DeepLabv3 by adding a simple yet effective decoder module to refine the segmentation results especially along object boundaries and applies the depthwise separable convolution to both Atrous Spatial Pyramid Pooling and decoder modules, resulting in a faster and stronger encoder-decoder network. Expand
...
1
2
3
4
5
...