Research map / Image generation / GANs
Neural Space-Filling Curves
Hanyu Wang, Kamal Gupta, Larry Davis, Abhinav Shrivastava
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 Neural SFC order instead of Hilbert curve | LZW compression · MNIST | LZW size in bytes: 182.7 (+7.3) -> 171.1 (-4.3) | Improved |
| use Neural SFC order instead of raster scan | LZW compression · Fashion-MNIST | LZW size in bytes: 425.8 -> 412.4 (-13.4) | Improved |
| use Neural SFC order instead of raster scan | LZW compression · FFHQ (32x32) | LZW size in bytes: 688.0 -> 678.3 (-9.7) | Improved |
| use Neural SFC order instead of raster scan | LZW compression · TGIF | LZW size in bytes per frame: 563.9 -> 556.9 (-7.0) | Improved |
| use Hilbert curve instead of raster scan | LZW compression · MNIST | LZW size in bytes: 175.4 -> 182.7 (+7.3) | Hurt |
| use Neural SFC instead of Dafner SFC averaged over image set | autocorrelation · MNIST | lag-6 autocorrelation: 0.401 -> 0.558 | Improved |
| use Neural SFC instead of Hilbert curve | autocorrelation · Fashion-MNIST | lag-10 autocorrelation: 0.647 -> 0.705 | Improved |
| use Neural SFC instead of Hilbert curve | autocorrelation · FFHQ (32x32) | lag-6 autocorrelation: 0.924 -> 0.943 | Improved |
| average Dafner SFC weights over the image set instead of using Hilbert curve | autocorrelation · MNIST | lag-6 autocorrelation: 0.475 -> 0.401 | Hurt |
| train with lag k=4 and 6 together instead of k=6 only (ablation) | Neural SFC · MNIST | lag-k autocorrelation (k=4 to 6): even better autocorrelations from k=4 to 6 | Improved |
| train with k=4, 6, 8 instead of k=6 only (ablation) | Neural SFC · MNIST | lag-k autocorrelation: worse performance | Hurt |
| train for k=6 instead of other single k values (ablation) | Neural SFC · MNIST | lag-k autocorrelation: k=6 generally best single k | Improved |
Closest papers by meaning
- Efficient Deep Space Filling Curve (ICCV 2023 · similarity 0.89)
- REOrdering Patches Improves Vision Models (NeurIPS 2025 · similarity 0.85)
- MoTIF: Learning Motion Trajectories with Local Implicit Neural Functions for Continuous Space-Time Video Super-Resolution (ICCV 2023 · similarity 0.83)
- Coordinate-Aware Modulation for Neural Fields (ICLR 2024 · similarity 0.83)
- Explaining the Implicit Neural Canvas: Connecting Pixels to Neurons by Tracing their Contributions (CVPR 2024 · similarity 0.82)
- Grids Often Outperform Implicit Neural Representation at Compressing Dense Signals (NeurIPS 2025 · similarity 0.82)
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.
