AI Paper Atlas

Research map / Point clouds / Shape and surface completion

Fast Learning of Signed Distance Functions From Noisy Point Clouds via Noise to Noise Mapping

Junsheng Zhou, Baorui Ma, Yu-Shen Liu, Zhizhong Han

TPAMI 2024 · Point clouds › Shape and surface completion · approach: Point cloud upsampling and normals · 19 citations (OpenAlex)

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 changedWhereMeasuredEffect
use EMD instead of Chamfer distance as the noise-to-noise distance (ablation)Noise2NoiseMapping · PU denoisingL2CD x1e4: 73.786 -> 4.221Improved
add geometric consistency regularization (lambda 0.1) to EMD for reconstruction (ablation)Noise2NoiseMapping · PU reconstructionL2CD x1e4: 80.917 -> 4.277Improved
increase regularization weight lambda from 0.1 to 0.2 (ablation)Noise2NoiseMapping · PU reconstructionL2CD x1e4: 4.277 -> 4.993Hurt
add regularization to denoising (lambda 0 vs 0.2) (ablation)Noise2NoiseMapping · PU denoisingL2CD x1e4: 4.221 -> 4.832Hurt
increase number of noisy observations N from 1 to 200 (ablation)Noise2NoiseMapping · PU denoisingL2CD x1e4: 4.976 -> 4.221Improved
increase batch size B from 250 to 10000 (ablation)Noise2NoiseMapping · PU denoisingL2CD x1e4: 4.221 -> 6.217Hurt
train for more iterations (40e4 to 80e4) (ablation)Noise2NoiseMapping · PU denoisingL2CD x1e4: 4.887 -> 4.221Improved
mix multiple noisy observations into one (ablation)Noise2NoiseMapping · PU denoisingL2CD x1e4: 4.221 -> 4.244No clear effect
add pull loss to hash-grid fast learning (ablation)Noise2NoiseMapping (+Fast) · PU denoising 10KL2CD x1e4: 8.016 -> 4.537Improved
add Eikonal regularization to hash-grid fast learning (ablation)Noise2NoiseMapping (+Fast) · PU denoising 10KL2CD x1e4: 5.255 -> 4.537Improved
use multi-resolution hash encoding (Fast)Noise2NoiseMapping · 3D Scenetraining time: 913 s -> 63 sImproved
use learned SDF prior from noisy SfM points to guide NeuSNeuS · DTU (15 scenes)Chamfer distance: 0.87 -> 0.72Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.

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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy