Research map / Point clouds / Shape and surface completion
AdaFit: Rethinking Learning-Based Normal Estimation on Point Clouds
Runsong Zhu, Yuan Liu, Zhen Dong, Yuan Wang, Tengping Jiang, Wenping Wang, Bisheng Yang
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 offset prediction (ablation) | weighted surface fitting (n=3) · PCPNet average | normal angle RMSE (degrees): 12.33 -> 11.22 | Improved |
| add offset prediction (ablation) | weighted surface fitting (n=1) · PCPNet average | normal angle RMSE (degrees): 12.54 -> 11.24 | Improved |
| add offset prediction (ablation) | weighted surface fitting (n=2) · PCPNet average | normal angle RMSE (degrees): 13.73 -> 11.26 | Improved |
| add offset prediction (ablation) | weighted surface fitting (n=3) · PCPNet no noise | normal angle RMSE (degrees): 7.91 -> 6.00 | Improved |
| add offset prediction (ablation) | weighted surface fitting (n=3) · PCPNet high noise | normal angle RMSE (degrees): 21.48 -> 21.76 | No clear effect |
| add offset prediction (ablation) | weighted surface fitting (n=2) · PCPNet varying density (strips) | normal angle RMSE (degrees): 12.00 -> 7.18 | Improved |
| add offset prediction makes performance insensitive to polynomial order (ablation) | weighted surface fitting · PCPNet average | normal angle RMSE (degrees): 13.73 vs 11.26 at n=2 | Improved |
| add CSA layers (scale 700) (ablation) | AdaFit with offsets · PCPNet average | normal angle RMSE (degrees): 11.22 -> 10.76 | Improved |
| use smaller neighborhood scale 256 instead of 700 without CSA (ablation) | AdaFit with offsets · PCPNet average | normal angle RMSE (degrees): 11.22 -> 11.02 | Mixed |
| truncate low-weight points (threshold 0.05) instead of offset prediction (ablation) | weighted fitting · PCPNet average | normal angle RMSE (degrees): 11.95 vs offset 11.22 | Hurt |
| truncate points with threshold 0.50 (ablation) | weighted fitting · PCPNet average | normal angle RMSE (degrees): 12.33 -> 13.18 | Hurt |
| use AdaFit instead of DeepFit | DeepFit · SceneNN | normal angle RMSE (degrees): 18.33 -> 16.25 | Improved |
Later papers that compared with this method
5 later accepted papers report AdaFit as a baseline in their results tables; 5 of them report a better number on most of the results they share with it.
- Learning Signed Hyper Surfaces for Oriented Point Cloud Normal Estimation (TPAMI 2024 · better on 1 of 1 shared result)
- NeAF: Learning Neural Angle Fields for Point Normal Estimation (AAAI 2023 · better on 7 of 8 shared results)
- Rethinking the Approximation Error in 3D Surface Fitting for Point Cloud Normal Estimation (CVPR 2023 · better on 2 of 2 shared results)
- GraphFit: Learning Multi-Scale Graph-Convolutional Representation for Point Cloud Normal Estimation (ECCV 2022 · better on 9 of 9 shared results)
- HSurf-Net: Normal Estimation for 3D Point Clouds by Learning Hyper Surfaces (NeurIPS 2022 · better on 5 of 5 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.
- DeepFit — DeepFit: 3D Surface Fitting via Neural Network Weighted Least Squares (ECCV 2020) · better on 9 of 9 shared results
- Nesti-Net — Nesti-Net: Normal Estimation for Unstructured 3D Point Clouds Using Convolutional Neural Networks (CVPR 2019) · 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
- Rethinking the Approximation Error in 3D Surface Fitting for Point Cloud Normal Estimation (CVPR 2023 · similarity 0.90)
- HSurf-Net: Normal Estimation for 3D Point Clouds by Learning Hyper Surfaces (NeurIPS 2022 · similarity 0.90)
- CMG-Net: Robust Normal Estimation for Point Clouds via Chamfer Normal Distance and Multi-Scale Geometry (AAAI 2024 · similarity 0.88)
- Inferring Neural Signed Distance Functions by Overfitting on Single Noisy Point Clouds through Finetuning Data-Driven based Priors (NeurIPS 2024 · similarity 0.86)
- LiSu: A Dataset and Method for LiDAR Surface Normal Estimation (CVPR 2025 · similarity 0.86)
- PLATYPUS: Progressive Local Surface Estimator for Arbitrary-Scale Point Cloud Upsampling (AAAI 2025 · 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.
