Research map / Video / Video generation
AUHead: Realistic Emotional Talking Head Generation via Action Units Control
Jiayi Lyu, Leigang Qu, Wenjing Zhang, Hanyu Jiang, Kai Liu, Zhenglin Zhou, Xiaobo Xia, Jian Xue, Tat-Seng Chua
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 |
|---|---|---|---|
| generate emotion first then AU (chain of thought) instead of AU only (ablation) | Qwen-Audio-Chat · Stage 1 AU prediction | MAE: 0.2447 -> 0.2085 | Improved |
| generate emotion first then AU (chain of thought) instead of AU only (ablation) | Qwen-Audio-Chat · Stage 1 AU prediction | F1: 0.62 -> 0.69 | Improved |
| predict emotion before AU rather than after AU (ablation) | Qwen-Audio-Chat · Stage 1 emotion accuracy | ACC_emo (%): 51.76 -> 67.01 | Improved |
| use 2D landmark or mesh-rendering AU representation instead of 1D AU sequence (ablation) | MEMO · MEAD | FID: 11.1105 -> 10.8701 | Improved |
| use 2D landmark or mesh-rendering AU representation instead of 1D AU sequence (ablation) | MEMO · CREMA | FID: 8.4159 -> 8.3352 | Improved |
| use 2D landmark or mesh-rendering AU representation instead of 1D AU sequence (ablation) | MEMO · MEAD | Sync: 6.7445 -> 6.6095 | Hurt |
| add AU conditioning (AUHead) to MEMO | MEMO · MEAD | FID: 11.1237 -> 10.9671 | Improved |
| add AU conditioning (AUHead) to MEMO | MEMO · MEAD | Sync: 6.9885 -> 6.6311 | Hurt |
| add AU conditioning (AUHead) to HalloV1 | HalloV1 · MEAD | Sync: 4.9512 -> 6.0201 | Improved |
| add AU conditioning (AUHead) to HalloV1 | HalloV1 · CREMA | PSNR: 23.2809 -> 23.0818 | Hurt |
| add AU conditioning (AUHead) to HalloV1 | HalloV1 · MEAD | PSNR: 22.0258 -> 22.0132 | No clear effect |
| raise AU classifier-free guidance scale (ablation) | AUHead · MEAD | FID | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Sonic — Sonic: Shifting Focus to Global Audio Perception in Portrait Animation (CVPR 2025) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- EmotiveTalk: Expressive Talking Head Generation through Audio Information Decoupling and Emotional Video Diffusion (CVPR 2025 · similarity 0.91)
- Style2Talker: High-Resolution Talking Head Generation with Emotion Style and Art Style (AAAI 2024 · similarity 0.90)
- FG-EmoTalk: Talking Head Video Generation with Fine-Grained Controllable Facial Expressions (AAAI 2024 · similarity 0.89)
- Playmate: Flexible Control of Portrait Animation via 3D-Implicit Space Guided Diffusion (ICML 2025 · similarity 0.89)
- DAWN: Dynamic Frame Avatar with Non-autoregressive Diffusion Framework for Talking head Video Generation (ICLR 2025 · similarity 0.89)
- GoHD: Gaze-oriented and Highly Disentangled Portrait Animation with Rhythmic Poses and Realistic Expressions (AAAI 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.
