AI Paper Atlas

Research map / 3D vision / NeRF and radiance fields

Real-Time Neural Radiance Talking Portrait Synthesis via Audio-Spatial Decomposition

Jiaxiang Tang, Kaisiyuan Wang, Hang Zhou, Xiaokang Chen, Dongliang He, Tianshu Hu, Jingtuo Liu, Ziwei Liu, Gang Zeng, Jingdong Wang

IJCV 2025 · 3D vision › NeRF and radiance fields · approach: Single-view and streaming radiance fields · 140 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 decomposed audio-spatial modulation instead of deformation backbone (ablation)Ours (head model) · Self-driven (Obama)LMD: 3.225 -> 2.696Improved
use decomposed audio-spatial modulation instead of deformation backbone (ablation)Ours (head model) · Self-driven (Obama)PSNR: 33.43 -> 34.07Improved
use decomposed grids instead of composed 5D audio-spatial grid (ablation)Ours (head model) · Self-driven (Obama)FPS: 35 -> 40Improved
use decomposed modulation instead of implicit MLP backbone (ablation)Ours (head model) · Self-driven (Obama)FPS: 28 -> 40Improved
use decomposed modulation instead of implicit MLP backbone (ablation)Ours (head model) · Self-driven (Obama)LPIPS: 0.0324 -> 0.0375Hurt
raise audio coordinate dimension from 2 to 3 (ablation)Ours · Self-driven (Obama)FPS / LMD: 40 -> 36 FPS; 2.696 -> 2.749 LMDHurt
raise audio coordinate dimension from 1 to 2 (ablation)Ours · Self-driven (Obama)LMD: 2.736 -> 2.696Improved
increase ray sampling count from 8 to 16 (ablation)Ours · Self-driven (Obama)FPS / LMD: 51 -> 40 FPS; 2.761 -> 2.696 LMDMixed
increase ray sampling count from 16 to 32 (ablation)Ours · Self-driven (Obama)LMD / FPS: 2.696 -> 2.661 LMD; 40 -> 35 FPSNo clear effect
use proposed method instead of AD-NeRFOurs · Self-driven (Obama)PSNR: 25.53 -> 34.00Improved
use proposed method instead of Wav2Lip for lip syncOurs · Cross-driven Testset ASync: 9.257 -> 6.218Hurt

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