Research map / Video / Video generation
EDTalk: Efficient Disentanglement for Emotional Talking Head Synthesis
Shuai Tan, Bin Ji, Mengxiao Bi, ye pan
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 orthogonality constraint on bases (ablation) | EDTalk · MEAD | Acc_emo: 38.71 -> 67.32 | Improved |
| add base banks storing visual priors (ablation) | EDTalk · MEAD | Acc_emo: 9.677 -> 67.32 | Improved |
| add Emotion Enhancement Module (ablation) | EDTalk · MEAD | Acc_emo: 49.367 -> 67.32 | Improved |
| add feature-level loss L_fea (ablation) | EDTalk · MEAD | Acc_emo: 54.34 -> 67.32 | Improved |
| add self-reconstruction loss L_self (ablation) | EDTalk · MEAD | Acc_emo: 44.23 -> 67.32 | Improved |
| add sync loss L^m_sync (ablation) | EDTalk audio-to-lip · MEAD | Sync_conf: 4.007 -> 8.115 | Improved |
| add expression disentanglement to lip and pose spaces (ablation) | EDTalk · MEAD | Sync_conf: 8.474 -> 8.115 Sync_conf; 16.75 -> 67.32 Acc_emo | Mixed |
| add reconstruction loss L^m_rec (ablation) | EDTalk audio-to-lip · MEAD | PSNR: 21.955 -> 21.628 PSNR; 5.447 -> 8.115 Sync_conf | Mixed |
| increase mouth bank size from 5 to 20 bases (ablation) | EDTalk · MEAD | Acc_emo: 63.53 -> 67.32 (40 bases: 63.12) | Improved |
| increase expression bank size beyond 10 bases (ablation) | EDTalk · MEAD | Acc_emo: 67.32 -> 61.34 (20), 59.65 (40) | Hurt |
| train disentanglement only lightweight modules on small data | EDTalk · training cost | training time: about one hour on 2 GPUs for mouth-pose vs over 2 days for DPE | Improved |
Later papers that compared with this method
11 later accepted papers report EDTalk as a baseline in their results tables; 10 of them report a better number on most of the results they share with it.
- PortraitDirector: A Hierarchical Disentanglement Framework for Controllable and Real-time Facial Reenactment (CVPR 2026 · better on 9 of 9 shared results)
- Cross-Modal Emotion Transfer for Emotion Editing in Talking Face Video (CVPR 2026 · better on 3 of 6 shared results)
- EDTalk++: Full Disentanglement for Controllable Talking Head Synthesis (TPAMI 2026 · better on 5 of 5 shared results)
- PC-Talk: Precise Facial Animation Control for Audio-Driven Talking Face Generation (CVPR 2026 · better on 4 of 4 shared results)
- Emotion-Conditioned Motion Sub-spaces with Flow Matching for Real-Time Audio-Driven Talking Heads (AAAI 2026 · better on 3 of 3 shared results)
- Conversational Human Audio-visual Talking Dialogue Generation (ECCV 2026 · better on 2 of 2 shared results)
- FixTalk: Taming Identity Leakage for High-Quality Talking Head Generation in Extreme Cases (ICCV 2025 · better on 9 of 10 shared results)
- KeyFace: Expressive Audio-Driven Facial Animation for Long Sequences via KeyFrame Interpolation (CVPR 2025 · better on 2 of 3 shared results)
3 more not listed.
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.
- StyleTalk — StyleTalk: One-Shot Talking Head Generation with Controllable Speaking Styles (AAAI 2023) · better on 2 of 2 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 0.92)
- FaceTalk: Audio-Driven Motion Diffusion for Neural Parametric Head Models (CVPR 2024 · similarity 0.91)
- FaceChain-ImagineID: Freely Crafting High-Fidelity Diverse Talking Faces from Disentangled Audio (CVPR 2024 · similarity 0.91)
- GoHD: Gaze-oriented and Highly Disentangled Portrait Animation with Rhythmic Poses and Realistic Expressions (AAAI 2025 · similarity 0.90)
- Style2Talker: High-Resolution Talking Head Generation with Emotion Style and Art Style (AAAI 2024 · similarity 0.90)
- DEEPTalk: Dynamic Emotion Embedding for Probabilistic Speech-Driven 3D Face Animation (AAAI 2025 · similarity 0.90)
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.
