4.6 Article

Compressed Sensing via Measurement-Conditional Generative Models

Journal

IEEE ACCESS
Volume 9, Issue -, Pages 155335-155352

Publisher

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/ACCESS.2021.3128721

Keywords

Generators; Training; Image reconstruction; Generative adversarial networks; Magnetic resonance imaging; Phase measurement; Artificial neural networks; Compressed sensing; artificial neural networks; image reconstruction; image enhancement; signal reconstruction and prediction; measurement-conditional generative models; mitigation of signal presence condition; magnetic resonance imaging

Funding

  1. National Research Foundation of Korea (NRF) - Korea Government, Ministry of Science and ICT (MSIT) [2018R1A5A1059921, 2019R1C1C1009192, 2021R1F1A106153511]
  2. Institute of Information & Communications Technology Planning & Evaluation (IITP) - Korea Government (MSIT) [2019-0-00075]
  3. Korea Medical Device Development Fund - Korea Government (the MSIT)
  4. Articial Intelligence Graduate School Program [Korea Advanced Institute of Science and Technology (KAIST)]
  5. Korea Medical Device Development Fund - Korea Government (MSIT) [202011B08-02, KMDF_PR_20200901_0014-2021-02]
  6. Korea Medical Device Development Fund - Korea Government (MSIT, the Ministry of Trade, Industry and Energy) [202011B08-02, KMDF_PR_20200901_0014-2021-02]
  7. Korea Medical Device Development Fund - Korea Government (Ministry of Health Welfare) [202011B08-02, KMDF_PR_20200901_0014-2021-02]
  8. Korea Medical Device Development Fund - Korea Government (Ministry of Food and Drug Safety) [202011B08-02, KMDF_PR_20200901_0014-2021-02]
  9. Future Medicine 20*30 Project of the Samsung Medical Center [SMX1210791]
  10. Korea Evaluation Institute of Industrial Technology (KEIT) [202011B08-02] Funding Source: Korea Institute of Science & Technology Information (KISTI), National Science & Technology Information Service (NTIS)

Ask authors/readers for more resources

The study introduces a framework that combines pre-trained generators and measurement information from training samples to achieve more accurate signal reconstruction. Experimental results show that this framework consistently outperforms existing methods across various applications and significantly reduces reconstruction errors.
Pre-trained generators have been frequently adopted in compressed sensing (CS) owing to their ability to effectively estimate signals with the prior of NNs. To further refine the NN-based prior, we propose a framework that allows the generator to utilize additional information from given measurements of training samples for prior learning, thereby yielding more accurate reconstruction for signals. As our framework has a simple form, it can be easily applied to existing CS methods using pre-trained generators. Through extensive experiments, we demonstrate that our framework consistently outperforms these works by a large margin and can reduce the reconstruction error up to an order of magnitude for the presented target applications. We also explain the experimental success theoretically by showing that our framework can slightly relax the stringent signal presence condition, which is required to guarantee the success of signal recovery.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.6
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available