DEGSTalk combines per-embedding Gaussian deformations with a hair-preserving fusion step to synthesize talking faces that beat prior baselines on PSNR, LPIPS, LMD, and SSIM.
Real-time talking head driven by voice and its application to communication and entertainment,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
DEGSTalk: Decomposed Per-Embedding Gaussian Fields for Hair-Preserving Talking Face Synthesis
DEGSTalk combines per-embedding Gaussian deformations with a hair-preserving fusion step to synthesize talking faces that beat prior baselines on PSNR, LPIPS, LMD, and SSIM.