Research map / Image generation / Image restoration
WIRE: Wavelet Implicit Neural Representations
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use a complex Gabor wavelet activation instead of other tested INR nonlinearities | INR · Image representation | PSNR: 43.2dB | Helped |
| Use a complex Gabor wavelet activation instead of other tested INR nonlinearities | INR · Thai statue occupancy volume | IOU: 0.99 | Helped |
| Use WIRE activation instead of the next-highest nonlinearity | NeRF · Drums, 25 training images | PSNR: +0.2dB | Helped |
| Use WIRE activation instead of SIREN | NeRF · Drums, 100 training images | PSNR: 0.4dB higher | Helped |
| Use WIRE instead of other tested nonlinearities | INR · Multi-image super-resolution | PSNR: 1dB or better | Helped |
| Use WIRE instead of other tested nonlinearities | INR · Multi-image super-resolution | SSIM: 0.04 higher | Helped |
| Add multidimensional localization using additional Gaussian windows (ablation) | WIRE · Denoising, CT reconstruction, and super-resolution | PSNR and SSIM | Helped |
| Add multidimensional localization using additional Gaussian windows (ablation) | WIRE · Sparse image representation | representation accuracy | Helped |
| Use SIREN-like weight initialization instead of standard initialization (ablation) | WIRE · Image representation and denoising | PSNR: up to 1dB higher | Helped |
| Set the Gaussian width parameter to zero, reducing WIRE to a sine activation (ablation) | WIRE · Image representation and denoising | accuracy | Hurt |
| Set the frequency parameter to zero, reducing WIRE to a Gaussian activation (ablation) | WIRE · Image representation and denoising | accuracy | Hurt |
Closest papers by meaning
- Implicit Neural Representations and the Algebra of Complex Wavelets (ICLR 2024 · similarity 0.90)
- MIRE: Matched Implicit Neural Representations (CVPR 2025 · similarity 0.86)
- Grids Often Outperform Implicit Neural Representation at Compressing Dense Signals (NeurIPS 2025 · similarity 0.86)
- A sampling theory perspective on activations for implicit neural representations (ICML 2024 · similarity 0.86)
- VI^3NR: Variance Informed Initialization for Implicit Neural Representations (CVPR 2025 · similarity 0.85)
- Implicit Representations via Operator Learning (ICML 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.