For shallow ReLU implicit neural representations with Fourier features, O(K^d) low-pass samples suffice for width-1 recovery and O(s^2) samples for width-s recovery in 2D, using generalized weight decay.
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
ACCEPT 1representative citing papers
citing papers explorer
-
Sampling Theory for Super-Resolution with Implicit Neural Representations
For shallow ReLU implicit neural representations with Fourier features, O(K^d) low-pass samples suffice for width-1 recovery and O(s^2) samples for width-s recovery in 2D, using generalized weight decay.