AI Paper Atlas

Research map / 3D vision / Face and shape

ICON: Implicit Clothed Humans Obtained From Normals

CVPR 2022 · 3D vision › Face and shape · approach: Clothed human modeling · 313 citations (OpenAlex)

Check your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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 local point features rather than global image/normal encoder features (ablation)ICON · CAPE-NFPChamfer: 1.096 vs 1.254; 1.096 vs 1.180Helped
Add SMPL-X body-normal conditioning to normal prediction (ablation)ICON · CAPEChamfer: 1.156 -> 1.142Helped
Add the body-normal feature to local implicit features (ablation)ICON · CAPEChamfer: 1.235 -> 1.142Helped
Add the body-refinement module when SMPL-X is perturbed (ablation)ICON · CAPEChamfer: 1.417 -> 1.339Helped
Use 50 feedback-loop refinement iterations rather than none (ablation)ICON · CAPEChamfer: 1.417 -> 1.339Helped
Use 50 feedback-loop refinement iterations rather than 10 (ablation)ICON · CAPEP2S: 1.515 -> 1.378Helped
Use ICON instead of PaMIRclothed-human reconstruction · CAPE-NFPChamfer: 2.216 -> 1.096Helped
Use body prior for back-side normal prediction (ablation)normal prediction on in-the-wild images · In-the-wild images, back normalspreference: 52.9% vs 47.1%; P-value 6.66e-2No 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