Research map / 3D vision / Face and shape
Learning to Generate Customized Dynamic 3D Facial Expressions
Rolandos Alexandros Potamias, Jiali Zheng, Stylianos Ploumpis, Giorgos Bouritsas, Evangelos Ververas, Stefanos Zafeiriou
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 changed | Where | Measured | Effect |
|---|---|---|---|
| replace PCA blendshape decoder with spiral mesh-convolution decoder | LSTM expression generator · 4DFAB test, total per-vertex loss | per-vertex loss (mm): 0.44 -> 0.39 | Improved |
| replace PCA blendshape decoder with mesh decoder | LSTM expression generator · 4DFAB test, Happy | per-vertex loss (mm): 0.49 -> 0.37 | Improved |
| replace PCA blendshape decoder with mesh decoder | LSTM expression generator · 4DFAB test, Angry | per-vertex loss (mm): 0.37 -> 0.35 | No clear effect |
| replace PCA blendshape decoder with mesh decoder | LSTM expression generator · 4DFAB test, Fear | per-vertex loss (mm): 0.43 -> 0.42 | No clear effect |
| replace PCA blendshape decoder with mesh decoder | LSTM expression generator · 4DFAB expression classification, total | F1: 0.57 -> 0.68 | Improved |
| replace PCA blendshape decoder with mesh decoder | LSTM expression generator · 4DFAB expression classification, Surprise | F1: 0.65 -> 0.81 | Improved |
| replace PCA blendshape decoder with mesh decoder | LSTM expression generator · 4DFAB expression classification, Sad | F1: 0.49 -> 0.61 | Improved |
| use mesh decoder for apex phase of expression | LSTM expression generator · 4DFAB per-frame L1 error | L1 error (mm): stable loss around 0.45mm across the apex phase | Improved |
Closest papers by meaning
- AnimateMe: 4D Facial Expressions via Diffusion Models (ECCV 2024 · similarity 0.92)
- Sparse to Dense Dynamic 3D Facial Expression Generation (CVPR 2022 · similarity 0.92)
- Photo-Realistic Facial Details Synthesis From Single Image (ICCV 2019 · similarity 0.90)
- 3D Facial Expressions through Analysis-by-Neural-Synthesis (CVPR 2024 · similarity 0.89)
- DEEPTalk: Dynamic Emotion Embedding for Probabilistic Speech-Driven 3D Face Animation (AAAI 2025 · similarity 0.88)
- TEASER: Token Enhanced Spatial Modeling for Expressions Reconstruction (ICLR 2025 · similarity 0.88)
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.
