Research map / 3D vision / Face and shape
EMOCA: Emotion Driven Monocular Face Capture and Animation
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Add emotion consistency loss during expression training (ablation) | EMOCA · AffectNet test | V-CCC: 0.66 -> 0.77 | Helped |
| Add emotion consistency loss during training on DECA data (ablation) | EMOCA · AffectNet test | V-CCC: 0.69 -> 0.76 | Helped |
| Train with AffectNet rather than DECA training data (ablation) | EMOCA · AffectNet test | V-CCC: 0.76 -> 0.77 | Helped |
| Add detailed displacement meshes (ablation) | EMOCA · AMT perceptual study | classification consistency: 0.68 -> 0.65 | Hurt |
| Add detailed displacement meshes (ablation) | EMOCA · AffectNet test | E-ACC: 0.68 -> 0.69 | Helped |
| Use ResNet-50 emotion features rather than EmoNet features for supervision (ablation) | EMOCA · 3D expression reconstruction | visual quality | Helped |
| Change emotion-feature distance from L2 to L1 or cosine similarity (ablation) | EMOCA · 3D expression reconstruction | visual quality | No clear effect |
| Remove relative landmark losses for eye closure, mouth closure, and lip width (ablation) | EMOCA · 3D expression reconstruction | visual quality | Hurt |
| Increase emotion consistency loss weight from too small to too high (ablation) | EMOCA · 3D expression reconstruction | expression naturalness | Mixed |
Closest papers by meaning
- DEEPTalk: Dynamic Emotion Embedding for Probabilistic Speech-Driven 3D Face Animation (AAAI 2025 · similarity 0.88)
- EmoNet-Face: An Expert-Annotated Benchmark for Synthetic Emotion Recognition (NeurIPS 2025 · similarity 0.88)
- 3D Facial Expressions through Analysis-by-Neural-Synthesis (CVPR 2024 · similarity 0.87)
- TEASER: Token Enhanced Spatial Modeling for Expressions Reconstruction (ICLR 2025 · similarity 0.87)
- EMOPortraits: Emotion-enhanced Multimodal One-shot Head Avatars (CVPR 2024 · similarity 0.87)
- Expressive Gaussian Human Avatars from Monocular RGB Video (NeurIPS 2024 · similarity 0.86)
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.