SAPE: Spatially-Adaptive Progressive Encoding for Neural Optimization

Implementing implicit neural representations with common neural structures, e.g., multilayer perecep-trons with ReLU activations (ReLU MLPs), proves to be challenging in the presence of signals with