AI Paper Atlas

Research map / 3D vision / Face and shape

Intrinsic Neural Fields: Learning Functions on Manifolds

Lukas Koestler, Daniel Grittner, Michael Moeller, Daniel Cremers, Zorah Lähner

ECCV 2022 · 3D vision › Face and shape · approach: Neural implicit surface reconstruction · 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
add more eigenfunctions (1023 instead of 64) (ablation)Intrinsic neural field · Texture reconstruction catPSNR: 31.18 -> 34.82Improved
add more eigenfunctions (1023 instead of 4) (ablation)Intrinsic neural field · Texture reconstruction catPSNR: 22.84 -> 34.82Improved
use the larger MLP (329k instead of 140k params) (ablation)Intrinsic neural field · Texture reconstruction catPSNR: 34.57 -> 34.82Improved
use the larger MLP (329k instead of 140k params) (ablation)Intrinsic neural field · Texture reconstruction humanLPIPS: 0.379 -> 0.306Improved
use intrinsic LBO embedding instead of random Fourier features (sigma=8)Texture Fields · Texture reconstruction catPSNR: 34.39 -> 34.82Improved
use intrinsic LBO embedding instead of random Fourier features (sigma=8), 200 epochsTexture Fields · Texture reconstruction humanPSNR: 31.85 -> 32.37Improved
use intrinsic LBO embedding instead of mapping to a sphereNeuTex · Texture reconstruction catPSNR: 31.96 -> 34.82Improved
increase NeuTex embedding sizeNeuTex · Texture reconstruction humanPSNR: 29.49 -> 29.22Mixed
change RFF sigma from 8 to 16TF+RFF · Texture reconstruction humanPSNR: 32.26 -> 31.53Hurt
train for 1000 instead of 200 epochsIntrinsic neural field · Texture reconstruction catPSNR: 34.79 -> 34.82No clear effect
sample point cloud with fewer points than vertices instead of meshIntrinsic neural field · Discretization robustness catPSNR: 34.82 -> 33.17Hurt
remesh with uniform isotropic remeshing instead of original meshIntrinsic neural field · Discretization robustness humanPSNR: 32.48 -> 32.63No clear effect

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