Research map / 3D vision / Face and shape
SAPE: Spatially-Adaptive Progressive Encoding for Neural Optimization
Amir Hertz, Or Perel, Raja Giryes, Olga Sorkine-hornung, Daniel Cohen-or
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 |
|---|---|---|---|
| Add SAPE to Fourier features | FFN · Natural Images | PSNR: 25.57 \pm 4.19 -> 28.09 \pm 4.04 | Improved |
| Add SAPE to Fourier features | FFN · Text Images | PSNR: 30.47 \pm 2.11 -> 31.84 \pm 2.15 | Improved |
| Add SAPE to Fourier features | FFN · Thingi10K | IoU: 0.945 \pm 0.03 -> 0.981 \pm 0.008 | Improved |
| Add SAPE to Fourier features | FFN · Turbosquid | IoU: 0.964 \pm 0.017 -> 0.969 \pm 0.013 | Improved |
| Add SAPE to Fourier features | FFN · MPEG7 | IoU: 0.873 \pm 0.142 -> 0.928 \pm 0.095 | Improved |
| Add SAPE to Fourier features | FFN · Mesh transfer | Chamfer: 1.27 -> 0.39 | Improved |
| Add SAPE to Fourier features | FFN · Mesh transfer | Hausdorff: 0.94 -> 0.49 | Improved |
| Add SAPE to Fourier features | FFN · Mesh transfer | Dirichlet: 2.22 -> 1.82 | Hurt |
| Add progressive frequency exposure and spatial adaptation (ablation) | Fourier feature network · 2D image regression | PSNR | Improved |
| Use spatially adaptive progression instead of global progression (ablation) | SAPE · 1D regression and 2D image regression | reconstruction quality | Improved |
| Use a spatial grid resolution approximating the training-coordinate sampling ratio (ablation) | SAPE · 2D image regression | reconstruction quality | Improved |
Closest papers by meaning
- How Learnable Grids Recover Fine Detail in Low Dimensions: A Neural Tangent Kernel Analysis of Multigrid Parametric Encodings (ICLR 2025 · similarity 0.83)
- Coordinate-Aware Modulation for Neural Fields (ICLR 2024 · similarity 0.83)
- Nonparametric Teaching of Implicit Neural Representations (ICML 2024 · similarity 0.83)
- MIRE: Matched Implicit Neural Representations (CVPR 2025 · similarity 0.82)
- Forward Learning with Top-Down Feedback: Empirical and Analytical Characterization (ICLR 2024 · similarity 0.82)
- Tuning the Frequencies: Robust Training for Sinusoidal Neural Networks (CVPR 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.
