RARE: Image Reconstruction using Deep Priors Learned without Ground Truth
Regularization by denoising (RED) is an image reconstruction framework that uses an image denoiser as a prior. Recent work has shown the state-of-the-art performance of RED with learned denoisers corresponding to pre-trained convolutional neural nets (CNNs). In this work, we propose to broaden the current denoiser-centric view of RED by considering priors corresponding to networks trained for more general artifact-removal. The key benefit of the proposed family of algorithms, called regularization by artifact-removal (RARE), is that it can leverage priors learned on datasets containing only undersampled measurements. This makes RARE applicable to problems where it is practically impossible to have fully-sampled groundtruth data for training. We validate RARE on both simulated and experimentally collected data by reconstructing a free-breathing whole-body 3D MRIs into ten respiratory phases from heavily undersampled k-space measurements. Our results corroborate the potential of learning regularizers for iterative inversion directly on undersampled and noisy measurements.
Code (1)
Tasks
DenoisingImage ReconstructionSimilar Papers 제목 키워드 기반
Image Reconstruction without Explicit Priors
We consider solving ill-posed imaging inverse problems without access to an explicit image prior or ground-truth examples. An overarching challenge in inverse problems is that there are many undesired images that fit to …
Image ReconstructionDual-Integrated Low-Latency Single-Lens Infrared Computational Imaging for Object Detection
Computational imaging enables compact infrared systems, but deep-learning pipelines that combine image reconstruction and object detection often introduce substantial inference latency. Most existing acceleration strateg…
Real-Time Object DetectionImage ReconstructionHigh-Fidelity Accelerated MRI Reconstruction by Scan-Specific Fine-Tuning of Physics-Based Neural Networks
Long scan duration remains a challenge for high-resolution MRI. Deep learning has emerged as a powerful means for accelerated MRI reconstruction by providing data-driven regularizers that are directly learned from data. …
Deep LearningMRI ReconstructionTransfer LearningInvertible generative models for inverse problems: mitigating representation error and dataset bias
Trained generative models have shown remarkable performance as priors for inverse problems in imaging -- for example, Generative Adversarial Network priors permit recovery of test images from 5-10x fewer measurements tha…
Compressive SensingDecoderDenoisingGenerative Adversarial NetworkEnhancing Low-dose CT Image Reconstruction by Integrating Supervised and Unsupervised Learning
Traditional model-based image reconstruction (MBIR) methods combine forward and noise models with simple object priors. Recent application of deep learning methods for image reconstruction provides a successful data-driv…
Computed Tomography (CT)CT ReconstructionImage Reconstruction