Research map / 3D vision / NeRF and radiance fields
BACON: Band-Limited Coordinate Networks for Multiscale Scene Representation
Check 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 BACON instead of NeRF | Neural radiance field rendering · Multiscale Blender dataset, 1× and 1/8 | PSNR: 26.734 -> 27.430; 26.464 -> 28.475 | Helped |
| Use BACON instead of NeRF | Neural radiance field rendering · Multiscale Blender dataset, 1/2 and 1/4 | PSNR: 28.941 -> 28.066; 29.297 -> 28.520 | Hurt |
| Use BACON instead of NeRF | Neural radiance field rendering · Multiscale Blender dataset, average | PSNR: 27.859 -> 28.123 | Helped |
| Use BACON instead of Mip-NeRF | Neural radiance field rendering · Multiscale Blender dataset, average | PSNR: 31.526 -> 28.123 | Hurt |
| Use BACON instead of Mip-NeRF | Neural radiance field rendering · Multiscale Blender dataset, 1×, 1/2, 1/4, and 1/8 | PSNR: 29.874 -> 27.430; 31.307 -> 28.066; 32.093 -> 28.520; 32.832 -> 28.475 | Hurt |
| Represent multiple scales with BACON rather than Fourier Features | 3D shape fitting · Stanford 3D scanning, 5 test scenes | Chamfer: 2.166e-6 -> 2.198e-6 | Hurt |
| Represent multiple scales with BACON rather than Fourier Features | 3D shape fitting · Stanford 3D scanning, 5 test scenes | IOU: 9.841e-1 -> 9.833e-1 | Hurt |
| Represent multiple scales with BACON rather than SIREN | 3D shape fitting · Stanford 3D scanning, 5 test scenes | Chamfer: 2.780e-6 -> 2.198e-6 | Helped |
| Represent multiple scales with BACON rather than SIREN | 3D shape fitting · Stanford 3D scanning, 5 test scenes | IOU: 9.751e-1 -> 9.833e-1 | Helped |
| Combine adaptive-frequency evaluation with multiscale grid subdivision | BACON SDF mesh extraction · SDF evaluation, 5 test scenes | Time (s): 17.91 -> 0.222 | Helped |
| Use adaptive-frequency evaluation rather than dense-grid evaluation | BACON SDF mesh extraction · SDF evaluation, 5 test scenes | Time (s): 17.91 -> 5.50 | Helped |
Closest papers by meaning
- Synergistic Integration of Coordinate Network and Tensorial Feature for Improving Neural Radiance Fields from Sparse Inputs (ICML 2024 · similarity 0.86)
- Light Field Networks: Neural Scene Representations with Single-Evaluation Rendering (NeurIPS 2021 · similarity 0.85)
- Residual Multiplicative Filter Networks for Multiscale Reconstruction (NeurIPS 2022 · similarity 0.85)
- BLiRF: Bandlimited Radiance Fields for Dynamic Scene Modeling (AAAI 2024 · similarity 0.84)
- ResFields: Residual Neural Fields for Spatiotemporal Signals (ICLR 2024 · similarity 0.83)
- Batch Normalization Alleviates the Spectral Bias in Coordinate Networks (CVPR 2024 · similarity 0.83)
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.