Pith. sign in

REVIEW 4 major objections 4 minor 33 references

HM-Talker: Hybrid Motion Modeling for High-Fidelity Talking Head Synthesis

T0 review · 4 major / 4 minor · reviewed 2026-08-05 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read HM-Talker claims to resolve the personalization-generalization trade-off in audio-driven talking heads by fusing explicit articulatory cues with implicit prosodic features through stochastic feature pairing.

desk verdict Plausible architecture, honest framing, but no artifacts and a circular lip-sync metric; worth referee time with demands for code and cleaner evaluation. read the letter →

arxiv 2508.10566 v3 pith:L62PEPYQ submitted 2025-08-14 cs.CV

Shiyu Liu , Kui Jiang , Junjun Jiang , Xianming Liu , Xiaocheng Feng , Hongxun Yao , Qi Tian This is my paper
classification cs.CV
keywords talkingheadsynthesisaudio-drivenanimationlipsynchronizationhybridmotionmodelingstochasticfeaturepairingfaciallandmarksactionunitspersonalizationandgeneralization
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

HM-Talker tries to settle a trade-off that runs through audio-driven talking-head synthesis: models that learn motion directly from speech (implicit models) generalize to new speakers but produce jittery or deformed faces, while models that use anatomical priors such as landmarks and action units (explicit models) stay structurally stable but look stiff and neutral. The paper's claim is that both can be had by extracting a rich vocabulary of motion cues from audio and video, then fusing the explicit and implicit streams with a stochastic pairing mechanism and alternating between speaker-specific and speaker-agnostic training objectives. If the claim holds, talking-head systems would no longer have to choose between lip-sync fidelity and natural expression, and few-shot personalized avatars could be driven by arbitrary speech. Reported experiments across several benchmarks indicate better visual realism (SSIM, LPIPS) and lip-sync accuracy (sync confidence) than current state-of-the-art methods.

What carries the argument

The load-bearing mechanism is the Stochastic Feature Pairing (SFP) strategy, which dynamically pairs explicit and implicit feature streams rather than concatenating them in a fixed way. This forces the motion decoder to learn a flexible mapping between anatomical articulatory cues (landmarks, action units) and prosodic, expressive audio features. The Cross-Modal Mapping Module (CMMM) supplies the cue vocabulary, and the alternating two-objective optimization ties the fused motion to both the specific identity and the driving audio.

What would settle it

A falsifying experiment: retrain or fine-tune HM-Talker with artificially perturbed landmark and action-unit sequences while keeping audio and appearance fixed. If visual realism and lip-sync scores do not drop materially, the explicit articulatory branch is not carrying the claimed weight. Similarly, replace SFP with deterministic fixed pairing; a large drop would confirm the stochastic mechanism, while no drop would falsify its necessity.

Watch

Extended reading notes

Core claim

The central discovery, as the authors present it, is that the apparent conflict between structural coherence and expressive generality is a modeling artifact, not a necessary trade-off. HM-Talker introduces a Cross-Modal Mapping Module (CMMM) that builds a vocabulary of motion cues from both audio and video, and a Hybrid Motion Modeling Module (HMMM) whose Stochastic Feature Pairing (SFP) strategy merges explicit articulatory features—landmarks and action units encoding what the mouth and face are doing—with implicit prosodic features encoding how the audio is delivered. The lower-face motion is then optimized iteratively, alternating between an identity-specific objective that preserves the

Load-bearing premise

The approach assumes that automatically estimated 2D facial landmarks and action units are accurate enough to serve as the explicit vocabulary for facial dynamics; if those estimates are noisy or miss important motions on in-the-wild faces, the hybrid fusion will inherit their errors.

Editorial extensions

If this is right

  • Talking-head systems can be built that stay structurally stable for new speakers without freezing into generic expressions.
  • Audio-driven avatars could preserve an individual's speaking style from a few seconds of reference video while remaining responsive to arbitrary input audio.
  • Lip-sync accuracy and visual realism can be improved simultaneously, rather than traded off.
  • The alternating identity-specific and identity-agnostic training schedule offers a template for other audio-to-motion tasks that need both personalization and generalization.
  • The reported gains on standard quality metrics suggest practical use in video dubbing and avatar animation.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the stochastic pairing is the active ingredient, an ablation that swaps it for deterministic concatenation and measures the drop in lip-sync confidence would isolate the mechanism; the paper does not report that isolation explicitly.
  • Because the explicit cues are 2D landmarks and action units, the method's ceiling is tied to the tracker's accuracy; a natural testable extension is to feed 3D landmarks or learned per-speaker landmarks and watch whether the gap to prior methods grows or shrinks.
  • The CMMM's cross-modal motion-cue vocabulary, if it captures articulatory state, could be reused as a disentangled motion representation in other rendering backbones, not just the one used here.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. HM-Talker proposes a hybrid motion modeling framework for audio-driven talking head synthesis. The method combines explicit articulatory cues (OpenFace landmarks and action units) with implicit prosodic features via a Cross-Modal Mapping Module (CMMM) and a Hybrid Motion Modeling Module (HMMM) that uses a Stochastic Feature Pairing (SFP) strategy. The authors state that this design resolves the trade-off between personalization and generalization and claim state-of-the-art visual realism and lip-sync accuracy across diverse settings. However, the supplied manuscript contains only the abstract, introduction, acknowledgement, and references; the method and experimental sections are absent. Consequently, the central claim is not verifiable from the submitted text.

Significance. The conceptual direction is appealing: feature-level fusion of explicit anatomical cues with implicit prosodic features could indeed address the personalization/generalization trade-off in talking-head synthesis. The introduction provides a clear taxonomy of implicit, explicit, and spatially compromised models, and the proposed framework is well motivated. However, the submission as provided contains no technical details, no training objectives, no experiments, no quantitative comparisons, and no artifacts such as code or videos. If the full paper were available and the potential circularity in lip-sync evaluation were addressed, this could be a useful contribution. As it stands, the significance cannot be assessed beyond the abstract-level promise.

major comments (4)
  1. [Overall (Section 1 to Section 5)] The manuscript text jumps from Section 1 (Introduction) directly to Section 5 (Acknowledgement). Sections 2–4, which would contain the proposed method (CMMM, HMMM, SFP), the training losses, the datasets, the comparisons, and the ablations, are missing. Without these sections, the central claim that HM-Talker outperforms state-of-the-art methods is unsupported. This is the primary blocker for any positive recommendation.
  2. [Abstract and references [7,8,23,26]] The lip-sync evaluation may be circular. The references indicate the use of SyncNet-style estimators both as a training objective (e.g., Wav2Lip [23], lip-reading-guided methods [26]) and as an evaluation metric (SyncNet confidence [8]). Because the training-loss and evaluation sections are omitted, it cannot be verified whether the same estimator is used on both sides. If it is, the reported lip-sync advantage may partly reflect overfitting to the exact statistics of the evaluation metric. Please state explicitly which losses are used in training and which metrics are reported, and include at least one lip-sync metric not used in training.
  3. [Method / explicit visual cues] The approach relies on OpenFace-estimated landmarks and action units as explicit articulatory cues. These estimators are known to be noisy on in-the-wild data, yet no ablation is provided to isolate the contribution of the explicit branch or to test sensitivity to landmark/AU errors. Without such an ablation, the source of the claimed gain—explicit cues, implicit prosody, or the stochastic pairing mechanism—remains ambiguous. A noise-injection study and an ablation that removes the explicit branch would be needed to support the hybrid-modeling claim.
  4. [Experiments (missing)] No experimental results, variance estimates, statistical significance tests, or qualitative comparisons are included. Even after the missing sections are restored, the evaluation should report multiple random seeds and confidence intervals for SSIM, LPIPS, and SyncNet scores, since differences among talking-head methods are typically small and dataset-dependent. Without these, the claim of 'outperforms state-of-the-art methods across diverse settings' cannot be critically assessed.
minor comments (4)
  1. [Figure 1] Figure 1 is described but not included in the supplied text, so the qualitative illustration of the 'dilemma' cannot be assessed.
  2. [References] The reference list contains formatting errors, e.g., 'V olker' (Blanz), 'Yao Chong Lim' (OpenFace author), and inconsistent spacing in 'etal.' and 'Conference on Computer Vision and Pattern Recognition' entries.
  3. [Affiliation] Affiliation 1 reads 'Harbin Institute of Technology University'; the standard name is 'Harbin Institute of Technology'.
  4. [Section numbering] The jump from Section 1 to Section 5 confirms an omission in the submitted version. Please ensure the arXiv listing includes the complete manuscript.

Circularity Check

0 steps flagged · score 0.0 of 10

No demonstrable circularity in the supplied manuscript text; training-loss/evaluation sections are not present, so the SyncNet train/eval overlap cannot be established as a circular step.

full rationale

The available text (abstract, introduction, acknowledgments, references) contains no derivation chain that reduces a predicted quantity to a fitted input or to a self-citation by construction. The claimed superiority is presented as an empirical result against external benchmarks (SSIM/LPIPS/SyncNet confidence, with references [7,8,27,29] to external toolkits), and the method's components (CMMM, HMMM, SFP) are described as architectural designs rather than as definitions of the evaluation metrics. No equation is included in the supplied portion, so the skeptic's concern that a SyncNet-style loss in Section 3.4 may coincide with the SyncNet evaluation metric cannot be verified; per the hard rules, unverifiable speculation about training/evaluation overlap is not circularity. There is also no load-bearing self-citation chain: the citation to InsTaG [19] is to prior work by other authors and is used as motivation, not as a uniqueness theorem or as justification that HM-Talker's output is correct. Therefore the honest finding is no significant circularity.

Assumptions & free parameters 2 free parameters · 3 assumptions · 1 invented entities

The central claim rests on unstated loss weights and network hyperparameters, on the reliability of OpenFace-based explicit cues, and on the trustworthiness of SyncNet as both a training signal and an evaluation metric. There are no open artifacts and no independent external benchmarks beyond the standard HDTF/VoxCeleb2 test sets, so the degree to which the paper itself contributes the claimed capabilities is not directly auditable.

free parameters (2)
  • Multi-scale temporal convolution architecture hyperparameters = Not specified in the manuscript
    The convolutional kernels and stride settings are not reported, so architectural degrees of freedom are unstated. These affect the model capacity and the reported results. This is a free parameter in the practical sense, not a fitted physical constant.
  • Loss weights for implicit/explicit objectives and SyncNet lip-sync loss = Not reported in manuscript
    The relative weighting of the identity-specific and audio-only objectives and the lip-sync loss must be tuned; the values are not given. These weights directly influence the trade-off claimed by the paper.
assumptions (3)
  • domain assumption OpenFace landmark and action unit extraction is sufficiently accurate for the talking-head task.
    The explicit motion-cue vocabulary is extracted with OpenFace; no validation of the accuracy of these cues on the target videos is given. In-the-wild video can degrade OpenFace accuracy.
  • domain assumption The paired implicit and explicit motion-cue vocabularies are complementary and jointly sufficient to represent lower-face talking motion.
    The entire Hybrid Motion Modeling Module assumes that no important motion information is lost between the two representations.
  • domain assumption SyncNet confidence is a reliable proxy for lip-sync accuracy.
    The paper uses SyncNet both as an evaluation metric and as a training loss; this assumes the metric is trustworthy and not gameable.
invented entities (1)
  • Stochastic Feature Pairing (SFP) strategy
    purpose: Dynamically merges paired implicit and explicit features during motion synthesis.
    This is an internal mechanism introduced for the proposed model; it is not a new physical entity and no falsifiable prediction outside the paper is provided.

how reviews work

0 comments
Cite this review

Pith. "Pith review of HM-Talker: Hybrid Motion Modeling for High-Fidelity Talking Head Synthesis." pith.science (2026). https://pith.science/paper/L62PEPYQ

@misc{pith2026250810566,
  author       = {Pith},
  title        = {Pith review of: HM-Talker: Hybrid Motion Modeling for High-Fidelity Talking Head Synthesis},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/L62PEPYQ}},
  note         = {Machine review of arXiv:2508.10566}
}
read the original abstract

Audio-driven talking head generation faces a fundamental trade-off between personalization and generalization, limiting its practical application. Implicit models often achieve generalization at the cost of structural incoherence, resulting in unstable head motion and inaccurate lip synchronization. While explicit methods incorporate geometric and anatomical priors such as 3D Morphable Models (3DMMs), which parameterize facial geometry, or Action Units (AUs), which code facial muscle movements--they tend to produce overly neutral expressions or suffer from limited generalization. To resolve this conflict, we present HM-Talker, an audio-driven talking head framework that synergistically integrates explicit articulatory cues with implicit prosodic features to characterize identity-specific dynamics while enabling audio-driven generalization. Its distinctive features can be summarized as: i) the Cross-Modal Mapping Module (CMMM) that extracts a comprehensive vocabulary of motion cues from audio and video, and ii) the Hybrid Motion Modeling Module (HMMM) that employs a Stochastic Feature Pairing (SFP) strategy to dynamically merge paired implicit and explicit features for motion synthesis. This design facilitates an iterative optimization of the lower face motion, alternating between identity-specific and identity-agnostic (audio-only) objectives. Extensive experiments demonstrate that HM-Talker outperforms state-of-the-art methods in both visual realism and lip-sync accuracy across diverse settings.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

33 extracted references · 28 canonical work pages

  1. [23]

    A lip sync expert is all you need for speech to lip generation in the wild

    KR Prajwal, Rudrabha Mukhopadhyay, Vinay P Nambood- iri, and CV Jawahar. A lip sync expert is all you need for speech to lip generation in the wild. InACM International Conference on Multimedia, pages 484–492, 2020. 1

    work page 2020

  2. [26]

    Seeing what you said: Talking face genera- tion guided by a lip reading expert

    Jiadong Wang, Xinyuan Qian, Malu Zhang, Robby T Tan, and Haizhou Li. Seeing what you said: Talking face genera- tion guided by a lip reading expert. InIEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 14653– 14662, 2023. 5, 6

    work page 2023

  3. [8]

    Out of time: auto- mated lip sync in the wild

    Joon Son Chung and Andrew Zisserman. Out of time: auto- mated lip sync in the wild. InAsian Conference on Computer Vision, pages 251–263. Springer, 2017. 5

    work page 2017

  4. [1]

    Openface 2.0: Facial behavior analysis toolkit

    Tadas Baltrusaitis, Amir Zadeh, Yao Chong Lim, and Louis- Philippe Morency. Openface 2.0: Facial behavior analysis toolkit. InInternational Conference on Automatic Face and Gesture Recognition, pages 59–66. IEEE, 2018. 5

    work page 2018

  5. [2]

    A morphable model for the synthesis of 3d faces

    V olker Blanz and Thomas Vetter. A morphable model for the synthesis of 3d faces. InProceedings of the 26th Annual Conference on Computer Graphics and Interactive Tech- niques, page 187–194, 1999. 2

    work page 1999

  6. [3]

    A morphable model for the synthesis of 3d faces

    V olker Blanz and Thomas Vetter. A morphable model for the synthesis of 3d faces. InSeminal Graphics Papers: Pushing the Boundaries, Volume 2, pages 157–164. Association for Computing Machinery, 2023. 2

    work page 2023

  7. [4]

    Lip movements generation at a glance

    Lele Chen, Zhiheng Li, Ross K Maddox, Zhiyao Duan, and Chenliang Xu. Lip movements generation at a glance. In European Conference on Computer Vision, pages 520–535,

    work page

  8. [5]

    Gaus- siantalker: Real-time high-fidelity talking head synthesis with audio-driven 3d gaussian splatting

    Kyusun Cho, Joungbin Lee, Heeji Yoon, Yeobin Hong, Jaehoon Ko, Sangjun Ahn, and Seungryong Kim. Gaus- siantalker: Real-time high-fidelity talking head synthesis with audio-driven 3d gaussian splatting. InACM Inter- national Conference on Multimedia, pages 10985–10994,

    work page

Show all 33 references
  1. [6]

    Generalizable and ani- matable gaussian head avatar, 2025

    Xuangeng Chu and Tatsuya Harada. Generalizable and ani- matable gaussian head avatar, 2025. 2

    2025

  2. [7]

    Lip reading in the wild

    Joon Son Chung and Andrew Zisserman. Lip reading in the wild. InAsian Conference on Computer Vision, pages 87–

  3. [9]

    Facial action coding system.Environmental Psychology & Nonverbal Behavior,

    Paul Ekman and Wallace V Friesen. Facial action coding system.Environmental Psychology & Nonverbal Behavior,

  4. [10]

    Monocular and generalizable gaussian talking head animation

    Shengjie Gong, Haojie Li, Jiapeng Tang, Dongming Hu, Shuangping Huang, Hao Chen, Tianshui Chen, and Zhuo- man Liu. Monocular and generalizable gaussian talking head animation. InIEEE/CVF Conference on Computer Vision and Pattern Recognition, 2025. 2

    2025

  5. [11]

    Beyond self-attention: External attention using two linear layers for visual tasks.IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(5):5436–5447, 2022

    Meng-Hao Guo, Zheng-Ning Liu, Tai-Jiang Mu, and Shi- Min Hu. Beyond self-attention: External attention using two linear layers for visual tasks.IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(5):5436–5447, 2022. 4

    2022

  6. [12]

    Ad-nerf: Audio driven neural ra- diance fields for talking head synthesis

    Yudong Guo, Keyu Chen, Sen Liang, Yong-Jin Liu, Hujun Bao, and Juyong Zhang. Ad-nerf: Audio driven neural ra- diance fields for talking head synthesis. InIEEE/CVF Inter- national Conference on Computer Vision, pages 5784–5794,

  7. [13]

    Deep speech: Scaling up end-to-end speech recognition.arXiv preprint arXiv:1412.5567, 2014

    Awni Hannun, Carl Case, Jared Casper, Bryan Catanzaro, Greg Diamos, Erich Elsen, Ryan Prenger, Sanjeev Satheesh, Shubho Sengupta, Adam Coates, et al. Deep speech: Scaling up end-to-end speech recognition.arXiv preprint arXiv:1412.5567, 2014. 7

    2014 arXiv

  8. [14]

    3d gaussian splatting for real-time radiance field rendering.ACM Transactions on Graphics, 42 (4), 2023

    Bernhard Kerbl, Georgios Kopanas, Thomas Leimk ¨uhler, and George Drettakis. 3d gaussian splatting for real-time radiance field rendering.ACM Transactions on Graphics, 42 (4), 2023. 1

    2023

  9. [15]

    Adam: A method for stochastic optimization.arXiv preprint arXiv:1412.6980,

    Diederik P Kingma and Jimmy Ba. Adam: A method for stochastic optimization.arXiv preprint arXiv:1412.6980,

    arXiv

  10. [16]

    J. P. Lewis, K. Anjyo, and T. Rhee. Practice and theory of blendshape facial models.The Eurographics Association,

  11. [17]

    Ef- ficient region-aware neural radiance fields for high-fidelity talking portrait synthesis

    Jiahe Li, Jiawei Zhang, Xiao Bai, Jun Zhou, and Lin Gu. Ef- ficient region-aware neural radiance fields for high-fidelity talking portrait synthesis. InIEEE/CVF International Con- ference on Computer Vision, pages 7568–7578, 2023. 1, 2, 5, 6

    2023

  12. [18]

    Talkinggaussian: Structure-persistent 3d talking head synthesis via gaussian splatting

    Jiahe Li, Jiawei Zhang, Xiao Bai, Jin Zheng, Xin Ning, Jun Zhou, and Lin Gu. Talkinggaussian: Structure-persistent 3d talking head synthesis via gaussian splatting. InEuropean Conference on Computer Vision, pages 127–145, 2025. 1, 2, 5, 6

    2025

  13. [19]

    Instag: Learning personalized 3d talking head from few-second video

    Jiahe Li, Jiawei Zhang, Xiao Bai, Jin Zheng, Jun Zhou, and Lin Gu. Instag: Learning personalized 3d talking head from few-second video. InIEEE/CVF Conference on Computer Vision and Pattern Recognition, 2025. 1

    2025

  14. [20]

    Decoupled weight de- cay regularization

    Ilya Loshchilov and Frank Hutter. Decoupled weight de- cay regularization. InInternational Conference on Learning Representations, 2018. 5

    2018

  15. [21]

    Nerf: Representing scenes as neural radiance fields for view syn- thesis.Communications of the ACM, 65(1):99–106, 2021

    Ben Mildenhall, Pratul P Srinivasan, Matthew Tancik, Jonathan T Barron, Ravi Ramamoorthi, and Ren Ng. Nerf: Representing scenes as neural radiance fields for view syn- thesis.Communications of the ACM, 65(1):99–106, 2021. 1

    2021

  16. [22]

    Synctalk: The devil is in the synchronization for talking head synthesis

    Ziqiao Peng, Wentao Hu, Yue Shi, Xiangyu Zhu, Xiaomei Zhang, Hao Zhao, Jun He, Hongyan Liu, and Zhaoxin Fan. Synctalk: The devil is in the synchronization for talking head synthesis. InIEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 666–676, 2024. 1, 4, 5, 6

    2024

  17. [24]

    Facial action coding system.Environmental Psychology & Nonverbal Behavior, 1, 2015

    Emily B Prince, Katherine B Martin, Daniel S Messinger, and M Allen. Facial action coding system.Environmental Psychology & Nonverbal Behavior, 1, 2015. 5

    2015

  18. [25]

    Real-time neural radiance talking portrait synthesis via audio-spatial decomposition.arXiv preprint arXiv:2211.12368, 2022

    Jiaxiang Tang, Kaisiyuan Wang, Hang Zhou, Xiaokang Chen, Dongliang He, Tianshu Hu, Jingtuo Liu, Gang Zeng, and Jingdong Wang. Real-time neural radiance talking portrait synthesis via audio-spatial decomposition.arXiv preprint arXiv:2211.12368, 2022. 1, 5, 6

    2022 arXiv

  19. [27]

    Image quality assessment: from error visibility to structural similarity.IEEE Transactions on Image Process- ing, 13(4):600–612, 2004

    Zhou Wang, Alan C Bovik, Hamid R Sheikh, and Eero P Si- moncelli. Image quality assessment: from error visibility to structural similarity.IEEE Transactions on Image Process- ing, 13(4):600–612, 2004. 5

    2004

  20. [28]

    Geneface: Generalized and high-fidelity audio-driven 3d talking face synthesis

    Zhenhui Ye, Ziyue Jiang, Yi Ren, Jinglin Liu, Jinzheng He, and Zhou Zhao. Geneface: Generalized and high-fidelity audio-driven 3d talking face synthesis. InThe Eleventh In- ternational Conference on Learning Representations, 2023. 5

    2023

  21. [29]

    The unreasonable effectiveness of deep features as a perceptual metric

    Richard Zhang, Phillip Isola, Alexei A Efros, Eli Shechtman, and Oliver Wang. The unreasonable effectiveness of deep features as a perceptual metric. InIEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 586–595,

  22. [30]

    Sadtalker: Learning realistic 3d motion coefficients for stylized audio- driven single image talking face animation

    Wenxuan Zhang, Xiaodong Cun, Xuan Wang, Yong Zhang, Xi Shen, Yu Guo, Ying Shan, and Fei Wang. Sadtalker: Learning realistic 3d motion coefficients for stylized audio- driven single image talking face animation. InIEEE/CVF Conference on Computer Vision and Pattern Recognition,

  23. [31]

    Dinet: Deformation inpainting network for realistic face visually dubbing on high resolution video

    Zhimeng Zhang, Zhipeng Hu, Wenjin Deng, Changjie Fan, Tangjie Lv, and Yu Ding. Dinet: Deformation inpainting network for realistic face visually dubbing on high resolution video. InAAAI Conference on Artificial Intelligence, pages 3543–3551, 2023. 1, 5, 6

    2023

  24. [32]

    Identity- preserving talking face generation with landmark and ap- pearance priors

    Weizhi Zhong, Chaowei Fang, Yinqi Cai, Pengxu Wei, Gangming Zhao, Liang Lin, and Guanbin Li. Identity- preserving talking face generation with landmark and ap- pearance priors. InIEEE/CVF Conference on Computer Vi- sion and Pattern Recognition, pages 9729–9738, 2023. 1, 5, 6

    2023

  25. [103]

    Springer, 2017. 5, 6

    2017

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.