Joint Optimization of Hadamard Sensing and Reconstruction in Compressed Sensing Fluorescence Microscopy

Wang, Alan Q.; LaViolette, Aaron K.; Moon, Leo; Xu, Chris; Sabuncu, Mert R.

doi:10.1007/978-3-030-87231-1_13

Cited by 1 publication

(1 citation statement)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since M θ receives noisy images as input, this network can be viewed as a post-processing network preceded by an optional model-based inversion step (e.g. applying the inverse Fourier transform to under-sampled measurements), which is a common reconstruction pipeline in the literature (Jin et al, 2017;Kang et al, 2017;Wang et al, 2021b). We expect the conclusions drawn in this work to hold for any deep learning-based main network, particularly state-of-the-art unrolled networks (Monga et al, 2019).…”

Section: Model Detailsmentioning

confidence: 77%

Computing Multiple Image Reconstructions with a Single Hypernetwork

Wang

Dalca

Sabuncu

2022

Melba

View full text Add to dashboard Cite

Deep learning based techniques achieve state-of-the-art results in a wide range of image reconstruction tasks like compressed sensing. These methods almost always have hyperparameters, such as the weight coefficients that balance the different terms in the optimized loss function. The typical approach is to train the model for a hyperparameter setting determined with some empirical or theoretical justification. Thus, at inference time, the model can only compute reconstructions corresponding to the pre-determined hyperparameter values. In this work, we present a hypernetwork-based approach, called HyperRecon, to train reconstruction models that are agnostic to hyperparameter settings. At inference time, HyperRecon can efficiently produce diverse reconstructions, which would each correspond to different hyperparameter values. In this framework, the user is empowered to select the most useful output(s) based on their own judgement. We demonstrate our method in compressed sensing, super-resolution and denoising tasks, using two large-scale and publicly-available MRI datasets. Our code is available at <a href='https://github.com/alanqrwang/hyperrecon'>https://github.com/alanqrwang/hyperrecon</a>.

show abstract