Research map / 3D vision / Face and shape
PET-NeuS: Positional Encoding Tri-Planes for Neural Surfaces
Yiqun Wang, Ivan Skorokhodov, Peter Wonka
This paper tried a common change: Add positional encoding
Add positional encoding to tri-plane SDF features · DTU · Chamfer distance: 1.23 -> 0.94 Improved
What 646 accepted papers report for this change
- 525 improved
- 58 no clear effect
- 7 mixed
- 56 hurt
- Learning-Augmented Scalable Linear Assignment Problem Optimization via Neural Dual Warm-Starts ICML 2026 · add positional encodings to features (d=13 -> d=21) · Block-structured, N=4096 · speedupImproved
- Mesh Based Simulations with Spatial and Temporal awareness ICML 2026 · use learned positional embedding, learnable relative bias or weighted adjacency instead of RoPE · Aneurysm · RMSEHurt
- Sparse ActionGen: Accelerating Diffusion Policy with Real-time Pruning ICML 2026 · add sinusoidal positional encoding to the pruner · Robomimic Tool Hang · success rate (%): 28 -> 54Improved
See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.
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 |
|---|---|---|---|
| Replace NeuS with PET-NeuS | NeuS · NeRF-synthetic | Chamfer distance: 1.97 -> 0.84 | Improved |
| Replace NeuS with PET-NeuS | NeuS · DTU | Chamfer distance: 0.84 -> 0.71 | Improved |
| Add positional encoding to tri-plane SDF features (ablation) | TP-NeuS · DTU | Chamfer distance: 1.23 -> 0.94 | Improved |
| Add multi-scale tri-planes (ablation) | TP-NeuS + PE · DTU | Chamfer distance: 0.94 -> 0.88 | Improved |
| Add self-attention convolution (ablation) | TP-NeuS + PE · DTU | Chamfer distance: 0.94 -> 0.82 | Improved |
| Add modulated positional encoding (ablation) | TP-NeuS + PE · DTU | Chamfer distance: 0.94 -> 0.80 | Improved |
| Add both MPE and SAC (ablation) | TP-NeuS + PE · DTU | Chamfer distance: 0.94 -> 0.71 | Improved |
| Initialize tri-planes from a geometrically initialized MLP (ablation) | PET-NeuS · DTU initialization ablation | Chamfer distance: 1.86 -> 1.05 | Improved |
| Use standard MLP geometric initialization and random tri-plane initialization instead of converted initialization (ablation) | PET-NeuS · DTU initialization ablation | Chamfer distance: 1.05 -> 1.34 | Hurt |
| Use 4 frequency bands instead of 2 (ablation) | SAC + MPE · DTU | Chamfer distance: 0.73 -> 0.71 | Improved |
| Use 6 frequency bands instead of 4 (ablation) | SAC + MPE · DTU | Chamfer distance: 0.71 -> 0.74 | Hurt |
| Use SAC + MPE instead of MSA + SAC (ablation) | Tri-plane surface reconstruction · DTU | Chamfer distance: 0.84 -> 0.71 | Improved |
Later papers that compared with this method
1 later accepted paper reports PET-NeuS as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- CRAYM: Neural Field Optimization via Camera RAY Matching (NeurIPS 2024 · better on 3 of 3 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- HF-NeuS — HF-NeuS: Improved Surface Reconstruction Using High-Frequency Details (NeurIPS 2022) · better on 3 of 3 shared results
- NeuS — NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstruction (NeurIPS 2021) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- SC-NeuS: Consistent Neural Surface Reconstruction from Sparse and Noisy Views (AAAI 2024 · similarity 0.88)
- NeuRodin: A Two-stage Framework for High-Fidelity Neural Surface Reconstruction (NeurIPS 2024 · similarity 0.87)
- Neural Processing of Tri-Plane Hybrid Neural Fields (ICLR 2024 · similarity 0.87)
- Neural Signed Distance Function Inference through Splatting 3D Gaussians Pulled on Zero-Level Set (NeurIPS 2024 · similarity 0.87)
- From Transparent to Opaque: Rethinking Neural Implicit Surfaces with $\alpha$-NeuS (NeurIPS 2024 · similarity 0.87)
- NeuSurf: On-Surface Priors for Neural Surface Reconstruction from Sparse Input Views (AAAI 2024 · similarity 0.86)
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.
