implicit neural representations
Implicit neural representations encode data (e.g., images or shapes) within continuous neural networks rather than discrete pixel values. This approach allows for high-resolution output and enables interpolation or extrapolation tasks without requiring explicit grids.
- A Few Moments Please: Scalable Graphon Learning via Moment Matching
- FLOWING: Implicit Neural Flows for Structure-Preserving Morphing
- Grids Often Outperform Implicit Neural Representation at Compressing Dense Signals
- Looking Into the Water by Unsupervised Learning of the Surface Shape
- MoRIC: A Modular Region-based Implicit Codec for Image Compression
- PDEfuncta: Spectrally-Aware Neural Representation for PDE Solution Modeling
- Ultra-high Resolution Watermarking Framework Resistant to Extreme Cropping and Scaling
- Understanding Bias Terms in Neural Representations