NC-GTV introduces a graph Huber function and Gershgorin-circle-based convexity parameter selection, then unrolls the ADMM solver into a lightweight feed-forward denoising network that matches larger CNNs with far fewer parameters.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.IV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Unrolling Nonconvex Graph Total Variation for Image Denoising
NC-GTV introduces a graph Huber function and Gershgorin-circle-based convexity parameter selection, then unrolls the ADMM solver into a lightweight feed-forward denoising network that matches larger CNNs with far fewer parameters.