Research map / Video / Video generation
FACIAL: Synthesizing Dynamic Talking Face With Implicit Attribute Learning
Chenxu Zhang, Yifan Zhao, Yifei Huang, Ming Zeng, Saifeng Ni, Madhukar Budagavi, Xiaohu Guo
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 |
|---|---|---|---|
| add local phonetic network G^loc (ablation) | FACIAL · Ablation (AV offset/confidence) | AV confidence: 4.309 -> 5.127 | Improved |
| add temporal network G^tem (ablation) | FACIAL · Ablation (AV offset/confidence) | AV confidence: 4.051 -> 5.127 | Improved |
| add temporal network G^tem (ablation) | FACIAL · Ablation (head motion) | head motion | Improved |
| add eye attention maps (part-sensitive encoding) (ablation) | FACIAL · 1,569 test clips | blink frequency | Improved |
| implicit head pose and blink learning vs prior 3D method | Yi · Talking-face dataset test split | LMD: 3.82 -> 3.57 | Improved |
| FACIAL vs Zhou on personalization of head pose | Zhou · Talking-face dataset test split | personalization (head pose): 0.52 -> 0.85 | Improved |
| FACIAL vs Zhou on personalization of blinks | Zhou · Talking-face dataset test split | personalization (blinks): 0.40 -> 0.73 | Improved |
| FACIAL vs Yi on blink user rating | Yi · User study | blink score: 0.2 -> 1.4 | Improved |
| FACIAL vs Yi on image quality user rating | Yi · User study | image quality score: 1.6 vs 1.6 | No clear effect |
| FACIAL vs Yi on AV offset | Yi · Talking-face dataset test split | AV offset: -2 vs -2 | No clear effect |
Closest papers by meaning
- Realistic Speech-Driven Facial Animation with GANs (IJCV 2019 · similarity 0.93)
- Speech-driven Facial Animation using Cascaded GANs for Learning of Motion and Texture (ECCV 2020 · similarity 0.92)
- Hierarchical Cross-Modal Talking Face Generation With Dynamic Pixel-Wise Loss (CVPR 2019 · similarity 0.92)
- Talking Face Generation by Adversarially Disentangled Audio-Visual Representation (AAAI 2019 · similarity 0.91)
- GoHD: Gaze-oriented and Highly Disentangled Portrait Animation with Rhythmic Poses and Realistic Expressions (AAAI 2025 · similarity 0.89)
- SyncTalk: The Devil is in the Synchronization for Talking Head Synthesis (CVPR 2024 · 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.
