Research map / 3D vision / Face and shape
MINER: Multiscale Implicit Neural Representation
Vishwanath Saragadam, Jasper Tan, Guha Balakrishnan, Richard G. Baraniuk, Ashok Veeraraghavan
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 finest scale (scale 0) instead of coarsest scale 3 | MINER · Thai Statue 3D point cloud | IoU: 0.95 -> 0.99 | Improved |
| use coarsest scale instead of finest scale | MINER · Thai Statue 3D point cloud | GPU memory: 6.6GB -> 1.8GB | Improved |
| MINER instead of ACORN for 3D fitting | MINER vs ACORN · Thai Statue 3D point cloud | training time to IoU 0.99: 53min -> 6min | Improved |
| MINER instead of ACORN for parameter count | MINER vs ACORN · Thai Statue 3D point cloud | # parameters: 17 million -> 9.9 million | Improved |
| MINER instead of ACORN on gigapixel image | MINER vs ACORN · Gigapixel image | PSNR: 32.6dB -> 38.5dB | Improved |
| MINER instead of ACORN, time to 40dB on 2MP image | MINER vs ACORN · 2 megapixel image | time to 40dB: 7 minutes -> less than 10s | Improved |
| use MSE loss instead of logistic loss (ablation) | MINER · 3D occupancy fitting | convergence speed | Improved |
| use Laplacian pyramid instead of Gaussian pyramid representation (ablation) | MINER · Pluto 4MP image | training error stability | Improved |
Closest papers by meaning
- Grids Often Outperform Implicit Neural Representation at Compressing Dense Signals (NeurIPS 2025 · similarity 0.85)
- M-plicits: Neural Implicit Surfaces via Nested Multiscale Residuals (NeurIPS 2026 · similarity 0.85)
- MIRE: Matched Implicit Neural Representations (CVPR 2025 · similarity 0.84)
- FINER: Flexible Spectral-bias Tuning in Implicit NEural Representation by Variable-periodic Activation Functions (CVPR 2024 · similarity 0.84)
- The Multiscale Single-Index Model: A Toy Model for Hierarchical Feature Learning (NeurIPS 2026 · similarity 0.84)
- FreSh: Frequency Shifting for Accelerated Neural Representation Learning (ICLR 2025 · 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.
