An improved neural network infers spatially adaptive sparsity level maps for convolutional dictionaries, enabling filter-permutation invariance and dictionary changes at inference while showing reduced degradation on out-of-distribution low-field MRI data.
This can be explained by the reduced reliance on training data due to its underlying model-based reconstruc- tion component
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.IV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Learning spatially adaptive sparsity level maps for arbitrary convolutional dictionaries
An improved neural network infers spatially adaptive sparsity level maps for convolutional dictionaries, enabling filter-permutation invariance and dictionary changes at inference while showing reduced degradation on out-of-distribution low-field MRI data.