Research map / 3D vision / Face and shape
Neural Poisson Solver: A Universal and Continuous Framework for Natural Signal Blending
Delong Wu, Hao Zhu, Qi Zhang, You Li, Xun Cao, Zhan Ma
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 |
|---|---|---|---|
| solve Poisson blending with a neural INR and PDE loss instead of discretized PIE solver | 2D image blending · 2D blending (5 scene pairs) | gradient and color error per pixel: 10x to 1000x lower error | Improved |
| solve Poisson blending with a neural INR on board/slogan scene | 2D image blending · Board/Slogan (max-gradient guide) | gradient error per pixel: 32.495 -> 0.1744 | Improved |
| compute color loss over the entire background instead of only the blend boundary (ablation) | Neural Poisson Solver · 2D blending ablation | qualitative blending quality | Improved |
| remove Dirichlet boundary condition and use global color loss | Neural Poisson Solver · 2D blending with varied mask shapes | robustness to mask shape | Improved |
| increase lambda weight on color loss (ablation) | Neural Poisson Solver · 2D blending | boundary smoothness vs feature clarity | Mixed |
| apply Neural Poisson Solver to radiance fields instead of direct ROI replacement | NeRF blending · 5D radiance field blending | qualitative coherence | Improved |
Closest papers by meaning
- Grids Often Outperform Implicit Neural Representation at Compressing Dense Signals (NeurIPS 2025 · similarity 0.87)
- Implicit Representations via Operator Learning (ICML 2024 · similarity 0.86)
- MIRE: Matched Implicit Neural Representations (CVPR 2025 · similarity 0.85)
- VI^3NR: Variance Informed Initialization for Implicit Neural Representations (CVPR 2025 · similarity 0.85)
- Neural Experts: Mixture of Experts for Implicit Neural Representations (NeurIPS 2024 · similarity 0.85)
- Learning Transferable Features for Implicit Neural Representations (NeurIPS 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.
