Research map / Video / Video generation
FixTalk: Taming Identity Leakage for High-Quality Talking Head Generation in Extreme Cases
Shuai Tan, Bill Gong, Bin Ji, 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 EMI and EDI to the EDTalk baseline | EDTalk · HDTF | CSIM: 0.594 -> 0.613 | Improved |
| add EMI and EDI to the EDTalk baseline | EDTalk · HDTF | NIQE: 42.41 -> 13.44 | Improved |
| add EMI and EDI to the EDTalk baseline | EDTalk · HDTF | CPBD: 0.221 -> 0.282 | Improved |
| add EMI and EDI to the EDTalk baseline | EDTalk · MEAD | PSNR: 21.628 -> 22.382 | Improved |
| add EMI and EDI to the EDTalk baseline | EDTalk · MEAD | Sync_conf: 8.115 -> 8.009 | Hurt |
| add EMI and EDI to the EDTalk baseline | EDTalk · MEAD | Acc_emo: 67.32 -> 68.25 | Improved |
| add EMI and EDI to the EDTalk baseline | EDTalk · User study identity | rating: 4.13 -> 4.47 | Improved |
| remove EMI only (keep EDI) (ablation) | FixTalk · qualitative | identity / artifacts | Mixed |
| remove EDI only (keep EMI) (ablation) | FixTalk · qualitative | identity / artifacts | Mixed |
| add EMI and EDI to plain baseline (ablation) | FixTalk · qualitative | identity / artifacts | Improved |
| plug FixTalk modules into LIA | LIA · qualitative | identity / quality | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- EDTalk — EDTalk: Efficient Disentanglement for Emotional Talking Head Synthesis (ECCV 2024) · better on 9 of 10 shared results
- Hallo3 — Hallo3: Highly Dynamic and Realistic Portrait Image Animation with Video Diffusion Transformer (CVPR 2025) · better on 1 of 1 shared result
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
- Synergizing Motion and Appearance: Multi-Scale Compensatory Codebooks for Talking Head Video Generation (CVPR 2025 · similarity 0.86)
- SyncTalk: The Devil is in the Synchronization for Talking Head Synthesis (CVPR 2024 · similarity 0.85)
- DAWN: Dynamic Frame Avatar with Non-autoregressive Diffusion Framework for Talking head Video Generation (ICLR 2025 · similarity 0.84)
- GoHD: Gaze-oriented and Highly Disentangled Portrait Animation with Rhythmic Poses and Realistic Expressions (AAAI 2025 · similarity 0.84)
- FaceChain-ImagineID: Freely Crafting High-Fidelity Diverse Talking Faces from Disentangled Audio (CVPR 2024 · similarity 0.84)
- AnyTalk: Multi-modal Driven Multi-domain Talking Head Generation (AAAI 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.
