Pith. sign in

REVIEW 3 major objections 4 minor 37 references

DEGSTalk: Decomposed Per-Embedding Gaussian Fields for Hair-Preserving Talking Face Synthesis

T0 review · 3 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read A 3D Gaussian talking-face model preserves long hair and beats four published baselines on all measured metrics.

desk verdict Promising incremental talking-face synthesis, but the defining hair-preserving fusion equation (7) is not a valid alpha composite as written and needs correction before the quantitative claims can be trusted. read the letter →

arxiv 2412.20148 v1 pith:BGRHSQOD submitted 2024-12-28 cs.CV cs.HC

Kaijun Deng , Dezhi Zheng , Jindong Xie , Jinbao Wang , Weicheng Xie , Linlin Shen , Siyang Song This is my paper · ORCID
classification cs.CVcs.HC
keywords 3DGaussianSplattingTalkingFaceSynthesisPre-EmbeddingHair-PreservingRenderingAudio-DrivenAnimationDeformableFieldsImplicit3DMMCoefficientsReal-Time
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

This paper tries to establish that the two persistent failure modes of talking-face synthesis--distorted dynamic regions and broken long-hair rendering--can both be fixed inside a 3D Gaussian Splatting pipeline without losing real-time speed. The proposed DEGSTalk gives every Gaussian primitive a learnable embedding and deforms those primitives with audio plus implicit 3DMM expression coefficients, so subtle eye and mouth motion is not smeared by smooth neural fields. It then renders hair as a separately preserved region and composites it back into the face with an opacity-weighted formula, which the authors argue removes the double-chin and boundary artifacts seen in earlier methods. If the paper is right, long-haired talking avatars can be trained in about half an hour and rendered at 114 FPS while scoring highest on PSNR, LPIPS, LMD, and SSIM against four published baselines. The paper itself acknowledges that some noisy primitives may still occur.

What carries the argument

The carrier is the Deformable Pre-Embedding Gaussian Field: each Gaussian primitive is assigned a learnable embedding $z \in \mathbb{R}^{32}$, and a shared MLP reads the encoded position-and-embedding combination $\mathcal{H}(\mu \oplus z)$, concatenates audio features $f_a$ and implicit 3DMM coefficients $f_e$, and predicts the per-primitive deformation $\{\Delta \mu, \Delta s, \Delta q\}$, which is added to the canonical position, scale, and rotation before splatting. The second carrier is Dynamic Hair-Preserving Portrait Rendering: the face and mouth regions are reconstructed separately with hair excluded and then fused with the preserved hair by $C = C_{hair} + C_{face}O_{face} + C_{mouth}(1 - O_{face})$. The ablation study identifies this hair-preserving compositing as the largest single contributor to the reported quality.

What would settle it

Re-run the four baselines and DEGSTalk on the same six clips using one shared preprocessing and metric implementation, and compare per-clip PSNR/LPIPS/LMD/SSIM; if the gap between DEGSTalk and TalkingGaussian shrinks to the noise level, the reported superiority is a protocol artifact rather than a property of the model. A complementary test is to train DEGSTalk on a held-out long-haired identity not among the six and inspect the hair/face boundary for ghosting or double edges, which would indicate the compositing does not generalize.

Watch

Extended reading notes

Core claim

The paper's central claim is that the distortions and hair artifacts that plague talking-face synthesis can be removed by giving each 3D Gaussian primitive a learnable embedding and rendering hair as a separately preserved region. In DEGSTalk, audio features $f_a$ and implicit 3DMM coefficients $f_e = \{\psi_{id}, \psi_{s}, \psi_{exp}, \psi_{eye}\}$ drive an MLP that predicts per-Gaussian deformations $\{\Delta \mu, \Delta s, \Delta q\}$ from the primitive's position and embedding, so dynamic regions such as eyes and mouth are represented by primitives that know their own identity. Hair is reconstructed with the face and mouth regions excluded, then composited by the opacity-weighted formula $C = C_{hair} + C_{face}O_{face} + C_{mouth}(1 - O_{face})$, which the authors argue prevents the double-chin and boundary noise seen in earlier methods. On six HD long-hair portraits, the paper reports the best scores on all four measured metrics (PSNR, LPIPS, LMD, SSIM), with about 0.5 hours of training and 114 FPS inference, and it closes by noting that some noisy primitives may still occur.

Load-bearing premise

The reported advantage over AD-NeRF, ER-NeRF, SyncTalk, and TalkingGaussian assumes all four baselines were measured on the same clips with the same crop, pose, and metric code; the paper does not say whether these baselines were re-run under that protocol.

Editorial extensions

If this is right

  • Long-hair talking avatars can be trained in about half an hour and rendered at 114 FPS, removing the main practical barrier to real-time 3DGS avatars for subjects with long hair.
  • The ablation attributes the largest quality drop (PSNR from 37.30 to 33.89) to removing the hair-preserving rendering, so the compositing stage is the key component to reproduce in follow-up work.
  • The best-reported numbers indicate that, on the tested six-clip dataset, per-Gaussian embeddings and implicit 3DMM coefficients together push PSNR from 32.39 (TalkingGaussian) to 37.30.
  • Since fine-tuning updates only color parameters, the deformation and embedding modules are kept from overfitting during the final stage, which the paper says is what makes the fused result realistic.

Reading between the lines

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

  • Our inference: the per-embedding deformation mechanism is not inherently face-specific, so the same recipe could be tried on other region-structured dynamic scenes such as hands or clothed bodies, but the paper provides no evidence for that transfer.
  • Our inference: because the reported metrics are per-frame, temporal flicker at the hair boundary is not measured; a short-term consistency test on long sequences would be the natural next evaluation.
  • Our inference: the SOTA claim is data-limited--six clips from one set of speakers--so the margin should be re-checked on unseen identities and hairstyles before being treated as a general property; the paper's own closing remark about noisy primitives points in the same direction.
  • Our inference: the opacity-weighted compositing suggests a matting interpretation; replacing the hand-chosen fusion order with learned alpha compositing could be a testable extension.
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

3 major / 4 minor

Summary. The paper proposes DEGSTalk, a 3D Gaussian Splatting (3DGS) based method for audio-driven talking face synthesis with emphasis on long-hair preservation. The method introduces per-Gaussian learnable embeddings z used by a deformation network, and a Dynamic Hair-Preserving Portrait Rendering stage that fuses separately rendered face, mouth, and hair regions. Experiments on six talking-head videos report consistent improvements over AD-NeRF, ER-NeRF, SyncTalk, and TalkingGaussian on PSNR, LPIPS, LMD, and SSIM, with inference at 114 FPS.

Significance. If the technical details are validated, the paper would make a useful contribution to 3DGS-based talking face synthesis, especially for the under-addressed case of long hair. The paper explicitly provides the loss formulations, training schedule, and an ablation study, and it promises public code. The main strengths are the novelty of per-Gaussian embeddings in this setting and the attempt to preserve hair through a dedicated rendering composition. However, the central rendering equation is not a coherent image-formation model as written, and the quantitative comparison lacks the protocol details needed to verify the claimed gains. The contribution is promising but the current evidence is insufficient for acceptance.

major comments (3)
  1. [II-C, Eq. (7)] The defining composition rule of the Dynamic Hair-Preserving Portrait Rendering is not a valid alpha-compositing formula as printed. In C = Chair + Cface × Oface + Cmouth × (1 − Oface), the hair color Chair is added without any opacity modulation or multiplication by (1 − Oface), so in pixels where the face is opaque the hair radiance is still added on top of the face color, which double-counts energy and can produce out-of-range RGB values. The mouth term is weighted by (1 − Oface), so it is strongest where the face is transparent, i.e., outside the face region, rather than in the mouth area. Moreover, Chair is never defined and no hair opacity term appears anywhere in the paper. Since the large H-versus-w/o-H gaps in Table II (PSNR 33.89 vs. 37.30, SSIM 0.9354 vs. 0.9664) and the qualitative hair-preservation claims are attributed to exactly this stage, Eq. (7) must be corrected and fully specified; otherwise the central mechanism is not assessable from the text.
  2. [III-B / Table I] The headline comparison in Table I is not verifiable as reported. The paper does not state whether AD-NeRF, ER-NeRF, SyncTalk, and TalkingGaussian were re-run by the authors under the same dataset, cropping, pose estimation, and metric protocol, or whether the numbers were taken from their original papers. Different test splits, preprocessing, or pose alignment can easily change the reported metric values by more than the margins shown here (for example, PSNR 33.73 vs. 35.66 and SSIM 0.9361 vs. 0.9678 for SyncTalk vs. the proposed method). The paper also reports no error bars or significance tests, so the consistent-looking gains could be within run-to-run variability. The authors should either state that all baselines were evaluated under identical conditions and provide the experimental setup, or re-run the baselines and report variances.
  3. [III-D / Table II] The ablation study contains internal inconsistencies that weaken its support for the component claims. The row 'w/o F' reports LPIPS of 0.150, which is an order of magnitude worse than the full model's 0.0140 and also much worse than every other ablation row; this looks like a typo, and without correction it undermines the ablation analysis. In addition, 'w/o E' achieves SSIM 0.9712 and 'w/o F' achieves SSIM 0.9692, both above the full model's 0.9664, yet the paper states 'Our method achieves strong performance across multiple evaluation metrics' without discussing these non-monotonic results. The paper should correct the table and either show that the full model is consistently best or temper the claim accordingly.
minor comments (4)
  1. [III-A] The phrase 'head reconstruction setting' appears in the metric section and in Table I, but the setting itself is never explicitly defined; please clarify what it means and whether the same setting applies to all baselines.
  2. [III-D / Table II] The last row of Table II is labeled 'w E, F, R, H', but the standard notation used elsewhere in the table is 'w/o'; this should be corrected to 'w/ E, F, R, H' or a similar unambiguous label.
  3. [II-C] The sentence 'Initially, the face and mouth regions undergo dilate operations during training to better capture their features, while also preserving the hair region Chair' is confusing: 'Chair' appears to be a term rather than a region name, and the connection between dilation and the preservation of Chair is not explained.
  4. [IV] The conclusion mentions 'Some noisy primitives may still occur' as future work, but this limitation is not introduced or discussed in the body; a short limitation paragraph would help contextualize the method's failure modes.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: DEGSTalk's reported gains are empirical measurements against external benchmarks, not quantities derived from its own fitted parameters.

full rationale

Walked the claimed derivation chain. The method is an empirical 3DGS pipeline: deformable per-embedding Gaussian fields are trained with image reconstruction losses (Eqs. 8-9) against ground-truth video frames, and the reported claims are evaluated with PSNR, LPIPS, LMD, SSIM on external benchmark videos. No predicted constant is recycled from a fitted value. The deformation model (Eqs. 5-6) maps audio features, 3DMM coefficients, and per-Gaussian embeddings to Gaussian parameter offsets; these quantities are all inputs, and the output renderings are compared pixel-wise to held-out ground truth, so the central realism claim is not circular by construction. The hair-preserving rendering equation (Eq. 7) is formally underspecified and arguably not a valid alpha-compositing rule, but that is a correctness/specification defect, not circularity: the equation does not define its output metric in terms of itself, nor does it fit a parameter that is later reported as a prediction. Baseline numbers in Table I may or may not share the same protocol, but a possible protocol mismatch concerns measurement validity, not logical circularity. No load-bearing self-citation was found: the only self-citations are unrelated prior affective-computing references [4,5], and the architecture builds on TalkingGaussian [25], which is an external, independently published baseline, not a result of the present authors. The paper contains no uniqueness theorem, no ansatz smuggled in via self-citation, and no renaming of a known result presented as unification. The honest finding is therefore no significant circularity with score 0.

Assumptions & free parameters 4 free parameters · 5 assumptions · 1 invented entities

The central claim is an empirical performance claim. The ledger shows the model relies on borrowed components (3DGS, 3DMM, hash encoding) plus an invented per-embedding latent. The equation-level novelty is small, and the evaluation rests on an unverified assumption that baseline numbers are comparable. The hand-set hyperparameters and the composition formula are load-bearing for the reported metrics.

free parameters (4)
  • embedding dimension d = 32
    Set manually in Implementation Details; no sensitivity study is reported.
  • loss weights lambda, beta, gamma = lambda=0.5, beta=0.001, gamma=0.2
    Chosen by hand in Eqs. 8 and 9; no ablation or tuning analysis is presented.
  • training iterations = 50,000 motion + 10,000 fine-tune
    Set in Implementation Details; effect on final quality is not studied.
  • hair and face dilate or fusion parameters = not specified
    Dynamic Hair-Preserving Portrait Rendering depends on dilation operations and region masks, but no values are given.
assumptions (5)
  • domain assumption 3D Gaussian Splatting is an adequate renderer for faces and hair
    Inherited from 3DGS via Section II-A; no independent justification is given for hair geometry.
  • domain assumption Predicted 3DMM coefficients from the cited reconstruction method are accurate enough to drive expression and jaw deformation
    Used as fe in Eqs. 3 and 5; noisy coefficients would break the deformation model.
  • domain assumption The tri-plane hash encoder plus MLP captures smooth facial motion
    Borrowed from TalkingGaussian without analysis in this paper.
  • ad hoc to paper The composition formula in Eq. 7 preserves hair while fusing face and mouth
    This is the proposed rendering rule; it is not derived and is validated only indirectly by final metrics.
  • domain assumption Baseline numbers in Table I are comparable to DEGSTalk under identical protocols
    The paper does not state whether the baselines were re-run; if they were not, the comparison is invalid.
invented entities (1)
  • per-Gaussian learnable embedding z in R^32
    purpose: Adds a latent code to each Gaussian primitive so the deformation MLP can specialize to facial regions
    Introduced in Section II-B; its only evidence is the internal ablation in Table II.

how reviews work

0 comments
Cite this review

Pith. "Pith review of DEGSTalk: Decomposed Per-Embedding Gaussian Fields for Hair-Preserving Talking Face Synthesis." pith.science (2026). https://pith.science/paper/BGRHSQOD

@misc{pith2026241220148,
  author       = {Pith},
  title        = {Pith review of: DEGSTalk: Decomposed Per-Embedding Gaussian Fields for Hair-Preserving Talking Face Synthesis},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BGRHSQOD}},
  note         = {Machine review of arXiv:2412.20148}
}
read the original abstract

Accurately synthesizing talking face videos and capturing fine facial features for individuals with long hair presents a significant challenge. To tackle these challenges in existing methods, we propose a decomposed per-embedding Gaussian fields (DEGSTalk), a 3D Gaussian Splatting (3DGS)-based talking face synthesis method for generating realistic talking faces with long hairs. Our DEGSTalk employs Deformable Pre-Embedding Gaussian Fields, which dynamically adjust pre-embedding Gaussian primitives using implicit expression coefficients. This enables precise capture of dynamic facial regions and subtle expressions. Additionally, we propose a Dynamic Hair-Preserving Portrait Rendering technique to enhance the realism of long hair motions in the synthesized videos. Results show that DEGSTalk achieves improved realism and synthesis quality compared to existing approaches, particularly in handling complex facial dynamics and hair preservation. Our code will be publicly available at https://github.com/CVI-SZU/DEGSTalk.

Figures

Figures reproduced from arXiv: 2412.20148 by the authors.

Figure 1
Figure 1. Overview of DEGSTalk. Given a cropped reference video of a talking face and its corresponding speech, our DEGSTalk first extracts the audio feature fa and performs 3D face reconstruction [6] to obtain 3DMM coefficients, including identity, shape, expression, eyelid movement, and jaw. Secondly, Gaussian primitives are pre-embedded to construct the deformable pre-emebdding Gaussian fields and then optimize the coarse … view at source ↗
Figure 2
Figure 2. Visual results of the comparative experiments. We show the generated [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Visual results of the ablation Study. Methods PSNR ↑ LPIPS ↓ LMD ↓ SSIM ↑ w/o F 36.79 0.150 2.630 0.9692 w/o R 35.66 0.0300 2.359 0.9678 w/o E 37.21 0.0141 2.357 0.9712 w/o H 33.89 0.0301 2.628 0.9354 w E, F, R, H 37.30 0.0140 2.349 0.9664 TABLE II THE QUANTITATIVE RESULTS OF ABLATION STUDY. IV. CONCLUSIONS In this paper, we propose a novel Deformable Pre￾Embedding Gaussian Fields (DEGSTalk) enhancing the real￾ism a… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

37 extracted references · 26 canonical work pages

  1. [1]

    Real-time talking head driven by voice and its application to communication and entertainment,

    Shigeo Morishima, “Real-time talking head driven by voice and its application to communication and entertainment,” in A VSP’98 Interna- tional Conference on Auditory-Visual Speech Processing , 1998

    work page 1998

  2. [2]

    The acceptability of virtual characters as social skills trainers: usability study,

    Hiroki Tanaka, Satoshi Nakamura, et al., “The acceptability of virtual characters as social skills trainers: usability study,” JMIR human factors, vol. 9, no. 1, pp. e35358, 2022

    work page 2022

  3. [3]

    Seeing is believing but is hearing? comparing audio and video communication for young children,

    Joanne Tarasuik, Roslyn Galligan, and Jordy Kaufman, “Seeing is believing but is hearing? comparing audio and video communication for young children,” Frontiers in psychology , vol. 4, pp. 64, 2013

    work page 2013

  4. [4]

    Learning person-specific cognition from facial reactions for automatic personality recognition,

    Siyang Song, Zilong Shao, Shashank Jaiswal, Linlin Shen, Michel Val- star, and Hatice Gunes, “Learning person-specific cognition from facial reactions for automatic personality recognition,” IEEE Transactions on Affective Computing , vol. 14, no. 4, pp. 3048–3065, 2022

    work page 2022

  5. [5]

    React2023: The first multiple appropriate facial reaction generation challenge,

    Siyang Song, Micol Spitale, Cheng Luo, Germ ´an Barquero, Cristina Palmero, Sergio Escalera, Michel Valstar, Tobias Baur, Fabien Ringeval, Elisabeth Andr ´e, et al., “React2023: The first multiple appropriate facial reaction generation challenge,” in Proceedings of the 31st ACM International Conference on Multimedia , 2023, pp. 9620–9624

    work page 2023

  6. [6]

    3d facial expressions through analysis-by-neural-synthesis,

    George Retsinas, Panagiotis P Filntisis, Radek Danecek, Victoria F Abrevaya, Anastasios Roussos, Timo Bolkart, and Petros Maragos, “3d facial expressions through analysis-by-neural-synthesis,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion, 2024, pp. 2490–2501

    work page 2024

  7. [7]

    Gen- erative adversarial nets,

    Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio, “Gen- erative adversarial nets,” Advances in neural information processing systems, vol. 27, 2014

    2014

  8. [8]

    Image- to-image translation with conditional adversarial networks,

    Phillip Isola, Jun-Yan Zhu, Tinghui Zhou, and Alexei A. Efros, “Image- to-image translation with conditional adversarial networks,” 2018

    work page 2018

Show all 37 references
  1. [9]

    High-resolution image synthesis with latent diffu- sion models,

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bj ¨orn Ommer, “High-resolution image synthesis with latent diffu- sion models,” 2021

    2021

  2. [10]

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

    K R Prajwal, Rudrabha Mukhopadhyay, Vinay P. Namboodiri, and C.V . Jawahar, “A lip sync expert is all you need for speech to lip generation in the wild,” in Proceedings of the 28th ACM International Conference on Multimedia , New York, NY , USA, 2020, MM ’20, p. 484–492, Assoc...

    2020

  3. [11]

    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,” in Proceedings of the IEEE/CVF Conference on Computer Vision and ...

    2023

  4. [12]

    Identity-preserving talking face generation with landmark and appearance priors,

    Weizhi Zhong, Chaowei Fang, Yinqi Cai, Pengxu Wei, Gangming Zhao, Liang Lin, and Guanbin Li, “Identity-preserving talking face generation with landmark and appearance priors,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , June 20...

    2023

  5. [13]

    Makelttalk: speaker-aware talking-head animation,

    Yang Zhou, Xintong Han, Eli Shechtman, Jose Echevarria, Evangelos Kalogerakis, and Dingzeyu Li, “Makelttalk: speaker-aware talking-head animation,” ACM Transactions on Graphics , vol. 39, no. 6, pp. 1–15, Nov. 2020

    2020

  6. [14]

    Difftalk: Crafting diffusion models for generalized audio-driven portraits animation,

    Shuai Shen, Wenliang Zhao, Zibin Meng, Wanhua Li, Zheng Zhu, Jie Zhou, and Jiwen Lu, “Difftalk: Crafting diffusion models for generalized audio-driven portraits animation,” in CVPR, 2023

    2023

  7. [15]

    Diffclip: Leveraging stable diffusion for language grounded 3d classi- fication,

    Sitian Shen, Zilin Zhu, Linqian Fan, Harry Zhang, and Xinxiao Wu, “Diffclip: Leveraging stable diffusion for language grounded 3d classi- fication,” 2024

    2024

  8. [16]

    Hallo: Hierarchical audio- driven visual synthesis for portrait image animation,

    Mingwang Xu, Hui Li, Qingkun Su, Hanlin Shang, Liwei Zhang, Ce Liu, Jingdong Wang, Yao Yao, and Siyu Zhu, “Hallo: Hierarchical audio- driven visual synthesis for portrait image animation,” 2024

    2024

  9. [17]

    Nerf: Representing scenes as neural radiance fields for view synthesis,

    Ben Mildenhall, Pratul P. Srinivasan, Matthew Tancik, Jonathan T. Barron, Ravi Ramamoorthi, and Ren Ng, “Nerf: Representing scenes as neural radiance fields for view synthesis,” 2020

    2020

  10. [18]

    Ad-nerf: Audio driven neural radiance fields for talking head synthesis,

    Yudong Guo, Keyu Chen, Sen Liang, Yong-Jin Liu, Hujun Bao, and Juyong Zhang, “Ad-nerf: Audio driven neural radiance fields for talking head synthesis,” in Proceedings of the IEEE/CVF International Conference on Computer Vision , 2021, pp. 5784–5794

    2021

  11. [19]

    Real- time neural radiance talking portrait synthesis via audio-spatial decom- position,

    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 decom- position,” arXiv preprint arXiv:2211.12368 , 2022

    2022 arXiv

  12. [20]

    Efficient region-aware neural radiance fields for high-fidelity talking portrait synthesis,

    Jiahe Li, Jiawei Zhang, Xiao Bai, Jun Zhou, and Lin Gu, “Efficient region-aware neural radiance fields for high-fidelity talking portrait synthesis,” in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) , October 2023, pp. 7568–7578

    2023

  13. [21]

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

    Ziqiao Peng, Wentao Hu, Yue Shi, Xiangyu Zhu, Xiaomei Zhang, Jun He, Hongyan Liu, and Zhaoxin Fan, “Synctalk: The devil is in the synchronization for talking head synthesis,” arXiv preprint arXiv:2311.17590, 2023

    2023 arXiv

  14. [22]

    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,” in The Eleventh International Conference on Learning Representations, 2022

    2022

  15. [23]

    Gene- face++: Generalized and stable real-time audio-driven 3d talking face generation,

    Zhenhui Ye, Jinzheng He, Ziyue Jiang, Rongjie Huang, Jiawei Huang, Jinglin Liu, Yi Ren, Xiang Yin, Zejun Ma, and Zhou Zhao, “Gene- face++: Generalized and stable real-time audio-driven 3d talking face generation,” arXiv preprint arXiv:2305.00787 , 2023

    2023 arXiv

  16. [24]

    3d gaussian splatting for real-time radiance field rendering,

    Bernhard Kerbl, Georgios Kopanas, Thomas Leimk ¨uhler, and George Drettakis, “3d gaussian splatting for real-time radiance field rendering,” ACM Transactions on Graphics , vol. 42, no. 4, July 2023

    2023

  17. [25]

    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,” arXiv preprint arXiv:2404.15264 , 2024

    2024 arXiv

  18. [26]

    Gaussiantalker: 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, “Gaussiantalker: Real-time high- fidelity talking head synthesis with audio-driven 3d gaussian splatting,” 2024

    2024

  19. [27]

    Gaussiantalker: Speaker-specific talking head synthesis via 3d gaussian splatting,

    Hongyun Yu, Zhan Qu, Qihang Yu, Jianchuan Chen, Zhonghua Jiang, Zhiwen Chen, Shengyu Zhang, Jimin Xu, Fei Wu, Chengfei Lv, and Gang Yu, “Gaussiantalker: Speaker-specific talking head synthesis via 3d gaussian splatting,” 2024

    2024

  20. [28]

    Gstalker: Real-time audio-driven talking face generation via deformable gaussian splatting,

    Bo Chen, Shoukang Hu, Qi Chen, Chenpeng Du, Ran Yi, Yanmin Qian, and Xie Chen, “Gstalker: Real-time audio-driven talking face generation via deformable gaussian splatting,” 2024

    2024

  21. [29]

    Semantic-aware implicit neural audio-driven video portrait generation,

    Xian Liu, Yinghao Xu, Qianyi Wu, Hang Zhou, Wayne Wu, and Bolei Zhou, “Semantic-aware implicit neural audio-driven video portrait generation,” in Computer Vision–ECCV 2022: 17th European Confer- ence, Tel Aviv, Israel, October 23–27, 2022, Proceedings, Part XXXVII . Springer, ...

    2022

  22. [30]

    Efficient region-aware neural radiance fields for high-fidelity talking portrait synthesis,

    Jiahe Li, Jiawei Zhang, Xiao Bai, Jun Zhou, and Lin Gu, “Efficient region-aware neural radiance fields for high-fidelity talking portrait synthesis,” in Proceedings of the IEEE/CVF International Conference on Computer Vision , 2023, pp. 7568–7578

    2023

  23. [31]

    Dssim: a structural similarity index for floating-point data,

    Allison H Baker, Alexander Pinard, and Dorit M Hammerling, “Dssim: a structural similarity index for floating-point data,” arXiv preprint arXiv:2202.02616, 2022

    2022 arXiv

  24. [32]

    Audio-driven talking head generation,

    Jingnan Gao, “Audio-driven talking head generation,” 2023

    2023

  25. [33]

    Adam: A method for stochastic optimization,

    Diederik P Kingma and Jimmy Ba, “Adam: A method for stochastic optimization,” arXiv preprint arXiv:1412.6980 , 2014

    2014 arXiv

  26. [34]

    Decoupled weight decay regulariza- tion,

    Ilya Loshchilov and Frank Hutter, “Decoupled weight decay regulariza- tion,” in International Conference on Learning Representations , 2018

    2018

  27. [35]

    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,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2018, pp. 586–595

    2018

  28. [36]

    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 Proceedings of the European conference on computer vision (ECCV) , 2018, pp. 520–535

    2018

  29. [37]

    Image quality assessment: from error visibility to structural similarity,

    Zhou Wang, Alan C Bovik, Hamid R Sheikh, and Eero P Simoncelli, “Image quality assessment: from error visibility to structural similarity,” IEEE transactions on image processing , vol. 13, no. 4, pp. 600–612, 2004

    2004

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.