AI Paper Atlas

Research map / 3D vision / Face and shape

TalkingGaussian: Structure-Persistent 3D Talking Head Synthesis via Gaussian Splatting

Jiahe Li, Jiawei Zhang, Xiao Bai, Jin Zheng, Xin Ning, Jun Zhou, Lin Gu

ECCV 2024 · 3D vision › Face and shape · approach: Gaussian and implicit head avatars · 61 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 deformation instead of appearance modification on the 3DGS backbone (ablation)TalkingGaussian · Self-reconstructionLPIPS: 0.0355 -> 0.0290Improved
use deformation instead of appearance modification on NeRF Tri-Hash backbone (ablation)ER-NeRF Tri-Hash · Self-reconstructionPSNR: 33.14 -> 31.50Hurt
add face-mouth decomposition to 3DGS deformation model (ablation)TalkingGaussian · Self-reconstructionSync-C: 5.676 -> 6.516Improved
add face-mouth decomposition to 3DGS appearance modification model (ablation)3DGS appearance · Self-reconstructionSync-C: 6.001 -> 6.332Improved
add incremental sampling to deformation + FMD (ablation)TalkingGaussian · Self-reconstructionSSIM: 0.906 -> 0.910Improved
additionally predict opacity adjustment alongside deformation (ablation)TalkingGaussian · Self-reconstructionSSIM: 0.910 -> 0.908No clear effect
additionally predict RGB color per primitive (ablation)TalkingGaussian · Self-reconstructionSSIM: 0.910 -> 0.907Hurt
use HuBERT instead of DeepSpeech audio features (ablation)TalkingGaussian · Self-reconstructionSync-C: 6.516 -> 6.667Improved
use TalkingGaussian instead of ER-NeRFtalking head synthesis · Self-reconstructionTraining time: 2.1h -> 0.5hImproved
use TalkingGaussian instead of ER-NeRF on cross-gender audiotalking head synthesis · Lip-sync, Test Audio A, MaySync-C: 3.639 -> 4.774Improved
use TalkingGaussian instead of IP-LAP on PSNRtalking head synthesis · Self-reconstructionPSNR: 35.34 -> 33.61Hurt

Later papers that compared with this method

6 later accepted papers report TalkingGaussian as a baseline in their results tables; 6 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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