Corpus ID: 236428889

LAConv: Local Adaptive Convolution for Image Fusion

@article{Jin2021LAConvLA,
  title={LAConv: Local Adaptive Convolution for Image Fusion},
  author={Zi-Rong Jin and Liang-Jian Deng and Tai-Xiang Jiang and Tian-Jing Zhang},
  journal={ArXiv},
  year={2021},
  volume={abs/2107.11617}
}
  • Zi-Rong Jin, Liang-Jian Deng, +1 author Tian-Jing Zhang
  • Published 2021
  • Computer Science, Engineering
  • ArXiv
The convolution operation is a powerful tool for feature extraction and plays a prominent role in the field of computer vision. However, when targeting the pixel-wise tasks like image fusion, it would not fully perceive the particularity of each pixel in the image if the uniform convolution kernel is used on different patches. In this paper, we propose a local adaptive convolution (LAConv), which is dynamically adjusted to different spatial locations. LAConv enables the network to pay attention… Expand

Figures and Tables from this paper

References

SHOWING 1-10 OF 30 REFERENCES
Detail Injection-Based Deep Convolutional Neural Networks for Pansharpening
TLDR
This article developed a DCNN that outperforms both the other detail injection-based proposals and several state-of-the-art pansharpening methods and is inspired by the direct difference between the PAN image and the upsampled low-resolution MS image. Expand
Pansharpening via Detail Injection Based Convolutional Neural Networks
  • Lin He, Yizhou Rao, +4 authors B. Li
  • Engineering, Computer Science
  • IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
  • 2019
TLDR
This paper designs a new detail injection based convolutional neural network (DiCNN) framework for pansharpening with the MS details being directly formulated in end-to-end manners, where the first detail injections based CNN mines MS details through the PAN image and the MS image, and the second one utilizes only the PAN picture. Expand
MTF-tailored Multiscale Fusion of High-resolution MS and Pan Imagery
TLDR
This work presents a multiresolution framework for merging a multispectral image having an arbitrary number of bands with a higher-resolution panchromatic observation, and demonstrates that a superior spatial enhancement is achieved by means of the MTF-adjusted fusion. Expand
Remote Sensing Image Fusion Based on Two-Stream Fusion Network
TLDR
Experiments on Quickbird and GaoFen-1 satellite images demonstrate that the proposed TFNet can fuse PAN and MS images, effectively, and produce pan-sharpened images competitive with even superior to state of the arts. Expand
Optimal MMSE Pan Sharpening of Very High Resolution Multispectral Images
TLDR
It is shown that the proposed method is computationally practical, even in the case of local optimization, and it outperforms the best state-of-the-art Pan-sharpening algorithms, as resulted from the IEEE Data Fusion Contest 2006, on true Ikonos and QuickBird data and on simulated Pleiades data. Expand
Deep Multiscale Detail Networks for Multiband Spectral Image Sharpening
TLDR
A new deep detail network architecture with grouped multiscale dilated convolutions to sharpen images contain multiband spectral information to improve generalization to new satellites, and is able to generalize different satellite images without the need for parameter tuning. Expand
Regularizing Hyperspectral and Multispectral Image Fusion by CNN Denoiser
TLDR
A novel HSI and MSI fusion method based on the subspace representation and convolutional neural network (CNN) denoiser, i.e., a well-trained CNN for gray image denoising, which has superior performance over the state-of-the-art fusion methods. Expand
Context-driven fusion of high spatial and spectral resolution images based on oversampled multiresolution analysis
TLDR
Two general and formal solutions to the problem of fusion of multispectral images with high-resolution panchromatic observations by exploiting the undecimated discrete wavelet transform and the generalized Laplacian pyramid are compared. Expand
DyNet: Dynamic Convolution for Accelerating Convolutional Neural Networks
TLDR
A novel dynamic convolution method to adaptively generate convolution kernels based on image contents that can reduce the computation cost remarkably while maintaining the performance and verify the scalability. Expand
Generalized Assorted Pixel Camera: Postcapture Control of Resolution, Dynamic Range, and Spectrum
TLDR
A comprehensive optimization method to arrive at the spatial and spectral layout of the color filter array of a GAP camera is presented and a novel algorithm for reconstructing the under-sampled channels of the image while minimizing aliasing artifacts is developed. Expand
...
1
2
3
...