no code implementations • 14 Jan 2024 • Chengli Tan, Jiangshe Zhang, Junmin Liu, Yicheng Wang, Yunda Hao
Recently, sharpness-aware minimization (SAM) has attracted a lot of attention because of its surprising effectiveness in improving generalization performance. However, training neural networks with SAM can be highly unstable since the loss does not decrease along the direction of the exact gradient at the current point, but instead follows the direction of a surrogate gradient evaluated at another point nearby.
1 code implementation • 9 Jun 2022 • Chengli Tan, Jiangshe Zhang, Junmin Liu
In this study, we argue that the hypothesis set SGD explores is trajectory-dependent and thus may provide a tighter bound over its Rademacher complexity.
1 code implementation • 5 May 2021 • Chengli Tan, Jiangshe Zhang, Junmin Liu
Instead, inspired by the short-range correlation emerging in the SGN series, we propose that SGD can be viewed as a discretization of an SDE driven by fractional Brownian motion (FBM).
1 code implementation • 10 Mar 2021 • Shuang Xu, Jiangshe Zhang, Kai Sun, Zixiang Zhao, Lu Huang, Junmin Liu, Chunxia Zhang
Pansharpening is a fundamental issue in remote sensing field.
1 code implementation • CVPR 2021 • Shuang Xu, Jiangshe Zhang, Zixiang Zhao, Kai Sun, Junmin Liu, Chunxia Zhang
Specifically, two optimization problems regularized by the deep prior are formulated, and they are separately responsible for the generative models for panchromatic images and low resolution multispectral images.
no code implementations • 31 Dec 2020 • Zixiang Zhao, Jiangshe Zhang, Shuang Xu, Kai Sun, Lu Huang, Junmin Liu, Chunxia Zhang
In addition, the latent information of features can be preserved effectively through adversarial training.
no code implementations • 16 Dec 2020 • Chengyang Liang, Zixiang Zhao, Junmin Liu, Jiangshe Zhang
Notably, scale-space filtering is exploited to implement adaptive searching for regions to be aligned, and instance-level features in each region are refined to reduce redundancy and noise mentioned in the second issue.
no code implementations • 21 Sep 2020 • Yicheng Wang, Shuang Xu, Junmin Liu, Zixiang Zhao, Chun-Xia Zhang, Jiangshe Zhang
Multi-Focus Image Fusion (MFIF) is a promising image enhancement technique to obtain all-in-focus images meeting visual needs and it is a precondition of other computer vision tasks.
no code implementations • 2 Sep 2020 • Zixiang Zhao, Jiangshe Zhang, Shuang Xu, Kai Sun, Chunxia Zhang, Junmin Liu
The core idea is that the encoder decomposes an image into base and detail feature maps with low- and high-frequency information, respectively, and that the decoder is responsible for the original image reconstruction.
2 code implementations • 18 May 2020 • Shuang Xu, Zixiang Zhao, Yicheng Wang, Chun-Xia Zhang, Junmin Liu, Jiangshe Zhang
Image fusion is a significant problem in many fields including digital photography, computational imaging and remote sensing, to name but a few.
Infrared And Visible Image Fusion Multi-Exposure Image Fusion
2 code implementations • 12 May 2020 • Zixiang Zhao, Shuang Xu, Chun-Xia Zhang, Junmin Liu, Jiangshe Zhang
In this paper, a novel Bayesian fusion model is established for infrared and visible images.
no code implementations • 12 May 2020 • Zixiang Zhao, Shuang Xu, Jiangshe Zhang, Chengyang Liang, Chunxia Zhang, Junmin Liu
The proposed AUIF model starts with the iterative formulas of two traditional optimization models, which are established to accomplish two-scale decomposition, i. e., separating low-frequency base information and high-frequency detail information from source images.
Infrared And Visible Image Fusion Rolling Shutter Correction
2 code implementations • 20 Mar 2020 • Zixiang Zhao, Shuang Xu, Chun-Xia Zhang, Junmin Liu, Pengfei Li, Jiangshe Zhang
Infrared and visible image fusion, a hot topic in the field of image processing, aims at obtaining fused images keeping the advantages of source images.
Ranked #5 on Semantic Segmentation on FMB Dataset
no code implementations • 12 Feb 2020 • Shuang Xu, Xiaoli Wei, Chunxia Zhang, Junmin Liu, Jiangshe Zhang
It is found that current methods are evaluated on simulated image sets or Lytro dataset.