mlsanityData to 14 Oct 2026

FixTalk: Taming Identity Leakage for High-Quality Talking Head Generation in Extreme Cases

arXiv 2507.01390 ↗First posted 2 Jul 2025cs.CVcited by 14

Not on the map. Why?

Abstract (arXiv)

Talking head generation is gaining significant importance across various domains, with a growing demand for high-quality rendering. However, existing methods often suffer from identity leakage (IL) and rendering artifacts (RA), particularly in extreme cases. Through an in-depth analysis of previous approaches, we identify two key insights: (1) IL arises from identity information embedded within motion features, and (2) this identity information can be leveraged to address RA. Building on these findings, this paper introduces FixTalk, a novel framework designed to simultaneously resolve both issues for high-quality talking head generation. Firstly, we propose an Enhanced Motion Indicator (EMI) to effectively decouple identity information from motion features, mitigating the impact of IL on generated talking heads. To address RA, we introduce an Enhanced Detail Indicator (EDI), which utilizes the leaked identity information to supplement missing details, thus fixing the artifacts. Extensive experiments demonstrate that FixTalk effectively mitigates IL and RA, achieving superior performance compared to state-of-the-art methods.

Key facts. Cited by 14 papers in the map (data through 14 Oct 2026). Builds on 77 of its 103 references that are in the map. First posted on arXiv 2 Jul 2025.

Builds on

77 of 103 references.

References · 77

  1. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  3. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,177
  4. 2023BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Modelscited by 6,137
  5. 2022Generative Adversarial Networkscited by 5,627
  6. 2016Perceptual Losses for Real-Time Style Transfer and Super-Resolutioncited by 2,694
  7. 2023Animate Anyone: Consistent and Controllable Image-to-Video Synthesis for Character Animationcited by 688
  8. 2021SUPERB: Speech processing Universal PERformance Benchmarkcited by 666
  9. 2022SadTalker: Learning Realistic 3D Motion Coefficients for Stylized Audio-Driven Single Image Talking Face Animationcited by 382
  10. 2023MagicAnimate: Temporally Consistent Human Image Animation using Diffusion Modelcited by 330
  11. 2020MakeItTalk: Speaker-Aware Talking-Head Animationcited by 312
  12. 2019Few-Shot Adversarial Learning of Realistic Neural Talking Head Modelscited by 284
38 more
  1. 2024Champ: Controllable and Consistent Human Image Animation with 3D Parametric Guidancecited by 272
  2. 2019Hierarchical Cross-Modal Talking Face Generationwith Dynamic Pixel-Wise Losscited by 252
  3. 2019Neural Voice Puppetry: Audio-driven Facial Reenactmentcited by 250
  4. 2024EMO: Emote Portrait Alive -- Generating Expressive Portrait Videos with Audio2Video Diffusion Model under Weak Conditionscited by 235
  5. 2018Talking Face Generation by Adversarially Disentangled Audio-Visual Representationcited by 212
  6. 2022Thin-Plate Spline Motion Model for Image Animationcited by 210
  7. 2023InterGen: Diffusion-based Multi-human Motion Generation under Complex Interactionscited by 206
  8. 2024LivePortrait: Efficient Portrait Animation with Stitching and Retargeting Controlcited by 206
  9. 2024Hallo: Hierarchical Audio-Driven Visual Synthesis for Portrait Image Animationcited by 202
  10. 2024MimicMotion: High-Quality Human Motion Video Generation with Confidence-aware Pose Guidancecited by 198
  11. 2024EchoMimic: Lifelike Audio-Driven Portrait Animations through Editable Landmark Conditionscited by 196
  12. 2024AniPortrait: Audio-Driven Synthesis of Photorealistic Portrait Animationcited by 194
  13. 2024VASA-1: Lifelike Audio-Driven Talking Faces Generated in Real Timecited by 187
  14. 2022Depth-Aware Generative Adversarial Network for Talking Head Video Generationcited by 178
  15. 2022VFHQ: A High-Quality Dataset and Benchmark for Video Face Super-Resolutioncited by 176
  16. 2022Latent Image Animator: Learning to Animate Images via Latent Space Navigationcited by 175
  17. 2024Follow-Your-Emoji: Fine-Controllable and Expressive Freestyle Portrait Animationcited by 174
  18. 2024ControlNeXt: Powerful and Efficient Control for Image and Video Generationcited by 159
  19. 2018Nonlinear 3D Face Morphable Modelcited by 157
  20. 2018Lip Movements Generation at a Glancecited by 155
  21. 2022EAMM: One-Shot Emotional Talking Face via Audio-Based Emotion-Aware Motion Modelcited by 151
  22. 2021Audio2Head: Audio-driven One-shot Talking-head Generation with Natural Head Motioncited by 149
  23. 2018Face Alignment in Full Pose Range: A 3D Total Solutioncited by 148
  24. 2023DiffTalk: Crafting Diffusion Models for Generalized Audio-Driven Portraits Animationcited by 136
  25. 2022MegaPortraits: One-shot Megapixel Neural Head Avatarscited by 131
  26. 2024Hallo3: Highly Dynamic and Realistic Portrait Image Animation with Video Diffusion Transformercited by 122
  27. 2020Talking-head Generation with Rhythmic Head Motioncited by 112
  28. 2024UniAnimate: Taming Unified Video Diffusion Models for Consistent Human Image Animationcited by 102
  29. 2024Loopy: Taming Audio-Driven Portrait Avatar with Long-Term Motion Dependencycited by 102
  30. 2023Seeing What You Said: Talking Face Generation Guided by a Lip Reading Expertcited by 98
  31. 2023Identity-Preserving Talking Face Generation with Landmark and Appearance Priorscited by 98
  32. 2022Learning Dynamic Facial Radiance Fields for Few-Shot Talking Head Synthesiscited by 97
  33. 2024Hallo2: Long-Duration and High-Resolution Audio-Driven Portrait Image Animationcited by 97
  34. 2022Progressive Disentangled Representation Learning for Fine-Grained Controllable Talking Head Synthesiscited by 93
  35. 2024X-Portrait: Expressive Portrait Animation with Hierarchical Motion Attentioncited by 91
  36. 2023StyleSync: High-Fidelity Generalized and Personalized Lip Sync in Style-based Generatorcited by 89
  37. 2024Animate-X: Universal Character Image Animation with Enhanced Motion Representationcited by 88
  38. 2024V-Express: Conditional Dropout for Progressive Training of Portrait Video Generationcited by 86

50 most cited of 77 All 77 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 4; 2026: 10.

14 papers.

Sort:

By area:

  1. 2025Animate-X++: Universal Character Image Animation with Dynamic Backgroundscited by 10
  2. 2025MEDTalk: Multimodal Controlled 3D Facial Animation with Dynamic Emotions by Disentangled Embeddingcited by 6
  3. 2025EDTalk++: Full Disentanglement for Controllable Talking Head Synthesiscited by 5
  4. 2026CoDance: An Unbind-Rebind Paradigm for Robust Multi-Subject Animationcited by 3
  5. 2026EmoTaG: Emotion-Aware Talking Head Synthesis on Gaussian Splatting with Few-Shot Personalizationcited by 3
  6. 2026ESGaussianFace: Emotional and Stylized Audio-Driven Facial Animation via 3D Gaussian Splattingcited by 2
  7. 2025IMTalker: Efficient Audio-driven Talking Face Generation with Implicit Motion Transfercited by 1
  8. 2026Multi-Subspace Multi-Modal Modeling for Diffusion Models: Estimation, Convergence and Mixture of Expertscited by 1
  9. 2026PortraitDirector: A Hierarchical Disentanglement Framework for Controllable and Real-time Facial Reenactmentcited by 1
  10. 2026Bridging Brain and Semantics: A Hierarchical Framework for Semantically Enhanced fMRI-to-Video Reconstructioncited by 1
  11. 2026TransBiolab: A Real-World Multi-View Dataset of Cluttered Transparent Biomedical Objectscited by 1
  12. 2026PerformRecast: Expression and Head Pose Disentanglement for Portrait Video Editingcited by 0
2 more
  1. 2026Archon: A Unified Multimodal Model for Holistic Digital Human Generationcited by 0
  2. 2026SubtleTalk: Generating Controllable Weakly-correlated Facial Dynamics for 3D Talking Heads via Residual Flow Matchingcited by 0