AI Paper Atlas

Research map / 3D vision / Face and shape

Multiresolution Deep Implicit Functions for 3D Shape Representation

Zhang Chen, Yinda Zhang, Kyle Genova, Sean Fanello, Sofien Bouaziz, Christian Häne, Ruofei Du, Cem Keskin, Thomas Funkhouser, Danhang Tang

ICCV 2021 · 3D vision › Face and shape · approach: Primitive-based shape abstraction · 43 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 multiresolution hierarchy instead of global-only latent (ablation)MDIF · ShapeNet chair auto-encodingChamfer L2: 0.228 -> 0.009Improved
use multiresolution hierarchy instead of local-only latent grid (ablation)MDIF · ShapeNet chair shape completionChamfer L2: 5.47 -> 1.34Improved
add global consistency loss (ablation)MDIF · ShapeNet chair shape completionChamfer L2: 1.43 -> 1.34Improved
decode residuals instead of SDF directly (ablation)MDIF · ShapeNet chair shape completionChamfer L2: 3.00 -> 1.34Improved
add latent grid dropout (ablation)MDIF · ShapeNet chair shape completionChamfer L2: 8.38 -> 1.34Improved
add global connection from level 0 to higher levels (ablation)MDIF · ShapeNet chair shape completionChamfer L2: 19.9 -> 1.34Improved
add global connection (ablation)MDIF · ShapeNet chair auto-encodingChamfer L2: 0.086 -> 0.009Improved
add latent grid dropout (ablation)MDIF · ShapeNet chair auto-encodingChamfer L2: 0.026 -> 0.009Improved
use decoder-only latent optimization instead of encoder-decoder inferenceMDIF · 3D-R2N2 test auto-encodingChamfer L2: 0.19 -> 0.10Improved
increase from 5 to 8 levels (ablation)MDIF · 3D-R2N2 test auto-encodingChamfer L2: 0.19 -> 0.12Improved
use MDIF instead of IF-NetMDIF · point cloud completionChamfer L2: 1.61 -> 0.39Improved
use MDIF instead of IF-NetMDIF · voxel super-resolutionChamfer L2: 1.82 -> 0.96Improved

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