Research map / Vision-language and multimodal / Multimodal LLMs
Who is a Better Talker: Subjective and Objective Quality Assessment for AI-Generated Talking Heads
Yingjie Zhou, Jiezhang Cao, Zicheng Zhang, Farong Wen, Yanwei Jiang, Jun Jia, Xiaohong Liu, Xiongkuo Min, Guangtao Zhai
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 first-frame spatial features (ablation) | FSCD · THQA-10K | SRCC: 0.6510 -> 0.8066 | Improved |
| add first-frame spatial features (ablation) | FSCD · THQA | SRCC: 0.6830 -> 0.7812 | Improved |
| add first-frame spatial features (ablation) | FSCD · THQA-3D | SRCC: 0.7368 -> 0.8235 | Improved |
| add Y-T slice temporal features (ablation) | FSCD · THQA-10K | SRCC: 0.7330 -> 0.8066 | Improved |
| add Y-T slice temporal features (ablation) | FSCD · THQA | SRCC: 0.7205 -> 0.7812 | Improved |
| add Y-T slice temporal features (ablation) | FSCD · THQA-3D | SRCC: 0.7506 -> 0.8235 | Improved |
| add LSE-D lip-sync distance map (ablation) | FSCD · THQA-10K | SRCC: 0.7610 -> 0.8066 | Improved |
| add LSE-C lip-sync confidence map (ablation) | FSCD · THQA-10K | SRCC: 0.7927 -> 0.8066 | Improved |
| add LSE-C lip-sync confidence map (ablation) | FSCD · THQA | SRCC: 0.7660 -> 0.7812 | Improved |
| add LSE-C lip-sync confidence map (ablation) | FSCD · THQA-3D | SRCC: 0.8019 -> 0.8235 | Improved |
| use FSCD instead of best prior VQA method | FSCD · THQA-10K | SRCC: 0.7775 -> 0.8066 | Improved |
Closest papers by meaning
- AGAV-Rater: Adapting Large Multimodal Model for AI-Generated Audio-Visual Quality Assessment (ICML 2025 · similarity 0.85)
- AE-NeRF: Audio Enhanced Neural Radiance Field for Few Shot Talking Head Synthesis (AAAI 2024 · similarity 0.85)
- Style2Talker: High-Resolution Talking Head Generation with Emotion Style and Art Style (AAAI 2024 · similarity 0.85)
- FaceSpeak: Expressive and High-Quality Speech Synthesis from Human Portraits of Different Styles (AAAI 2025 · similarity 0.85)
- GAIA: Zero-shot Talking Avatar Generation (ICLR 2024 · similarity 0.85)
- TalkCuts: A Large-Scale Dataset for Multi-Shot Human Speech Video Generation (NeurIPS 2025 · similarity 0.84)
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.
