Adaptive Selection of Sampling-Reconstruction in Fourier Compressed Sensing
Not on the map. Why?
Abstract (arXiv)
Compressed sensing (CS) has emerged to overcome the inefficiency of Nyquist sampling. However, traditional optimization-based reconstruction is slow and can not yield an exact image in practice. Deep learning-based reconstruction has been a promising alternative to optimization-based reconstruction, outperforming it in accuracy and computation speed. Finding an efficient sampling method with deep learning-based reconstruction, especially for Fourier CS remains a challenge. Existing joint optimization of sampling-reconstruction works (H1) optimize the sampling mask but have low potential as it is not adaptive to each data point. Adaptive sampling (H2) has also disadvantages of difficult optimization and Pareto sub-optimality. Here, we propose a novel adaptive selection of sampling-reconstruction (H1.5) framework that selects the best sampling mask and reconstruction network for each input data. We provide theorems that our method has a higher potential than H1 and effectively solves the Pareto sub-optimality problem in sampling-reconstruction by using separate reconstruction networks for different sampling masks. To select the best sampling mask, we propose to quantify the high-frequency Bayesian uncertainty of the input, using a super-resolution space generation model. Our method outperforms joint optimization of sampling-reconstruction (H1) and adaptive sampling (H2) by achieving significant improvements on several Fourier CS problems.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 23 of its 41 references that are in the map. First posted on arXiv 18 Sep 2024.
Builds on
23 of 41 references.
References · 23
- 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
- 2014Deep Learning Face Attributes in the Wildcited by 3,679
- 2013Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computationcited by 2,031
- 2016Density estimation using Real NVPcited by 2,029
- 2016Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoisingcited by 1,352
- 2018fastMRI: An Open Dataset and Benchmarks for Accelerated MRIcited by 464
- 2021SRDiff: Single Image Super-Resolution with Diffusion Probabilistic Modelscited by 415
- 2020SRFlow: Learning the Super-Resolution Space with Normalizing Flowcited by 209
- 2020End-to-End Variational Networks for Accelerated MRI Reconstructioncited by 201
- 2019Reducing Uncertainty in Undersampled MRI Reconstruction with Active Acquisitioncited by 55
- 2020Active MR k-space Sampling with Reinforcement Learningcited by 45
- 2016Learning Diverse Image Colorizationcited by 43
11 more
- 2020Experimental design for MRI by greedy policy searchcited by 40
- 2021B-spline Parameterized Joint Optimization of Reconstruction and K-space Trajectories (BJORK) for Accelerated 2D MRIcited by 29
- 2020Extending LOUPE for K-space Under-sampling Pattern Optimization in Multi-coil MRIcited by 22
- 2022On learning adaptive acquisition policies for undersampled multi-coil MRI reconstructioncited by 14
- 2021End-to-End Sequential Sampling and Reconstruction for MRIcited by 13
- 2021Image Restoration by Deep Projected GSUREcited by 8
- 2022FS-NCSR: Increasing Diversity of the Super-Resolution Space via Frequency Separation and Noise-Conditioned Normalizing Flowcited by 8
- 2022On the Robustness of Normalizing Flows for Inverse Problems in Imagingcited by 8
- 2019Fast Bayesian Uncertainty Estimation and Reduction of Batch Normalized Single Image Super-Resolution Networkcited by 6
- 2021Learning the sampling density in 2D SPARKLING MRI acquisition for optimized image reconstructioncited by 3
- 2023Learning to sample in Cartesian MRIcited by 1
Built on by
Citing papers per year: 2025: 1; 2026: 1.
2 papers.