AI Paper Atlas

Research map / 3D vision / Face and shape

Neural Body: Implicit Neural Representations With Structured Latent Codes for Novel View Synthesis of Dynamic Humans

CVPR 2021 · 3D vision › Face and shape · approach: Full-body avatar reconstruction · 633 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 Neural Body rather than NHRZJU-MoCap novel-view synthesis · ZJU-MoCap averagePSNR: 23.90 -> 28.10Helped
Use Neural Body rather than NHRZJU-MoCap novel-view synthesis · ZJU-MoCap averageSSIM: 0.897 -> 0.944Helped
Add per-frame latent embeddings for temporally varying appearance (ablation)Neural Body · TwirlPSNR: 30.03 -> 30.56; 0.53 PSNR improvementHelped
Increase training camera views from 1 to 6 (ablation)Neural Body · Twirl, six-view ablation testPSNR: 25.08 -> 32.73Helped
Increase training camera views from 1 to 6 (ablation)Neural Body · Twirl, six-view ablation testSSIM: 0.912 -> 0.979Helped
Train with more video frames, increasing from 1 to 300 (ablation)Neural Body · Twirl, first framePSNR: 25.64 -> 30.66Helped
Train with more video frames, increasing from 300 to 1200 (ablation)Neural Body · Twirl, first framePSNR: 30.66 -> 29.97Hurt

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