pith. sign in

arxiv: 2606.30347 · v1 · pith:ZW4L5ATXnew · submitted 2026-06-29 · 💻 cs.CV · cs.AI

FFAvatar: Feed-Forward 4D Head Avatar Reconstruction from Sparse Portrait Images

Jianjiang Yao , Ke Xian , Renxiang Dai , Robert Caiming Qiu This is my paper

Pith reviewed 2026-06-30 06:27 UTC · model grok-4.3

classification 💻 cs.CV cs.AI
keywords 4D head avatarsfeed-forward reconstruction3D Gaussian splattingalternating attentionincremental reconstructionanimatable avatarssparse portrait imagesmotion refinement
0
0 comments X

The pith

FFAvatar builds animatable 4D head avatars incrementally from one or more sparse portrait images by disentangling identity from expression and viewpoint.

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

The paper presents a feed-forward Transformer-based 3D Gaussian method that constructs high-quality, controllable 4D head avatars from variable numbers of reference photos. An alternating attention mechanism separates identity appearance from expression and viewpoint changes to form a pose-consistent canonical 3D representation. Coarse features are learned on sparse primitives tied to FLAME vertices and then densified in the UV domain for finer geometry and texture, while a plug-and-play module adds residual subject-specific motion. The approach supports progressive refinement as more images arrive and produces efficient, identity-consistent driving and rendering.

Core claim

FFAvatar is a Transformer-based 3D Gaussian framework that supports incremental reconstruction of animatable 4D head avatars from any number of reference portrait images, using an alternating attention mechanism to disentangle identity appearance from expression and viewpoint variations for a consistent canonical 3D appearance, a sparse-to-dense learning paradigm for balancing fidelity and efficiency, and a motion refinement module for subject-specific personalization.

What carries the argument

Alternating attention mechanism that disentangles identity appearance from expression and viewpoint variations to reconstruct a canonical 3D appearance consistent across poses and facial expressions.

If this is right

  • Reconstruction quality and detail improve progressively as additional reference images are provided.
  • The sparse-to-dense paradigm delivers both computational efficiency and fine-grained geometric and texture fidelity.
  • The plug-and-play motion refinement module allows subject-specific dynamic personalization beyond standard parametric deformation.
  • The resulting avatars support efficient driving and identity-consistent rendering from diverse expressions and viewpoints.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The disentanglement approach could be tested on full-body or non-human subjects to check generalization beyond heads.
  • Incremental updates might support live capture scenarios where new photos arrive over time.
  • The framework could be adapted to other parametric 3D models besides FLAME for different animation domains.

Load-bearing premise

The alternating attention mechanism successfully disentangles identity appearance from expression and viewpoint variations to produce a canonical 3D appearance consistent across poses and expressions.

What would settle it

Reconstruction of the same identity from multiple images under extreme expression and viewpoint changes where the output avatar shows visible identity drift or inconsistent texture when animated with unseen motions.

Figures

Figures reproduced from arXiv: 2606.30347 by Jianjiang Yao, Ke Xian, Renxiang Dai, Robert Caiming Qiu.

Figure 1
Figure 1. Figure 1: FFAvatar. We present FFAvatar, a feed-forward framework that reconstructs high-fidelity 4D portrait avatars from one or more reference portrait images. The ref￾erence images can be virtual characters synthesized by large-scale generative models from text prompts. Furthermore, by leveraging diffusion-based MMDM [56] to synthe￾size novel-view and novel-expression 2D images, our approach further enhances reco… view at source ↗
Figure 2
Figure 2. Figure 2: Overview of FFAvatar. Canonical field modeling. Visual features extracted by DINOv3 are concatenated with camera pose and expression encodings. An alternat￾ing attention mechanism performs intra- and inter-image matching to infer a consistent global appearance representation across all inputs. This representation is then aligned with the FLAME template through the proposed Sparse-to-Dense Cross-modal Align… view at source ↗
Figure 3
Figure 3. Figure 3: Novel expression synthesis. Qualitative comparison of novel expression synthesis on the VFHQ monocular test set. We compare our method with Real3D-Portrait [70], Portrait4D-v2 [13], GAGAvatar [8], LAM [23], GPAvatar [9], and FastAvatar [62] [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Novel View Synthesis. Qualitative comparison of novel view synthesis results on the NeRsemble multi-view subset. For the single-reference setting, we compare our method with Portrait4D-v2 [13], GAGAvatar [8], LAM [23], and FastAvatar [62]. For the multi-reference setting, we further compare against GaussianAvatars [52], FlashA￾vatar [63], GHA [66], and GPAvatar [9] [PITH_FULL_IMAGE:figures/full_fig_p011_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Qualitative comparisons with FastGHA and Avat3r. [PITH_FULL_IMAGE:figures/full_fig_p012_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Multi-view rendering results of our method under different input conditions: single-image input (left) and 16-image input (right). The rendered viewpoints corre￾spond to yaw angles of 30◦ , 10◦ , −10◦ , and −30◦ . From top to bottom, we show the RGB renderings, depth maps, and surface normal maps produced by 3D Gaussian splatting. 4.3 Ablation Studies Incremental Reconstruction. FFAvatar supports increment… view at source ↗
Figure 7
Figure 7. Figure 7: Ablation study of key design components. (a) Optimization using the original FLAME vertices as Gaussian primitives results in insufficient texture detail in several regions; (b) direct optimization of dense point clouds leads to suboptimal deformation, with some primitives not well optimized; (c) removing the motion refinement module causes inaccurate dynamic motion representation; (d) final results of the… view at source ↗
read the original abstract

We present FFAvatar, a Transformer-based 3D Gaussian framework for fast construction of high-quality and animatable 4D head avatars from one or more reference portrait images. Unlike existing feed-forward approaches that require a fixed number of input views, FFAvatar supports incremental reconstruction, progressively refining the avatar representation as additional reference images become available. At the core of our method is an alternating attention mechanism that disentangles identity appearance from expression and viewpoint variations, enabling the reconstruction of a canonical 3D appearance that remains consistent across poses and facial expressions. To balance visual fidelity and computational efficiency, we introduce a sparse-to-dense learning paradigm. Coarse appearance features are first learned using sparse primitives anchored to the FLAME vertex level and are subsequently densified in the UV domain to capture fine-grained geometric and texture details. We further propose a plug-and-play motion refinement module that enables subject-specific dynamic personalization by modeling residual motion beyond parametric deformation. Extensive experiments demonstrate that FFAvatar efficiently produces high-fidelity and controllable 4D head avatars, achieving superior flexibility, driving efficiency, and identity-consistent rendering across diverse expressions and viewpoints.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 0 minor

Summary. The manuscript presents FFAvatar, a Transformer-based 3D Gaussian framework for fast construction of high-quality and animatable 4D head avatars from one or more reference portrait images. The method supports incremental reconstruction, uses an alternating attention mechanism to disentangle identity appearance from expression and viewpoint variations for a canonical 3D appearance, introduces a sparse-to-dense learning paradigm with FLAME-anchored sparse primitives densified in the UV domain, and proposes a plug-and-play motion refinement module for subject-specific dynamic personalization. Extensive experiments are claimed to demonstrate superior flexibility, driving efficiency, and identity-consistent rendering.

Significance. If the claims are substantiated by rigorous experiments, this could be a significant contribution to 4D avatar reconstruction by providing a flexible feed-forward approach that handles variable numbers of input views and enables efficient personalization, potentially improving upon methods that require fixed view counts.

major comments (1)
  1. [Abstract] The abstract states the method and claims but supplies no quantitative results, ablation studies, or error analysis; without the full paper the support for the claims cannot be evaluated.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their review. We address the single major comment below.

read point-by-point responses

  1. Referee: [Abstract] The abstract states the method and claims but supplies no quantitative results, ablation studies, or error analysis; without the full paper the support for the claims cannot be evaluated.

    Authors: Abstracts in computer vision papers are intentionally concise to summarize the approach and contributions within length constraints and do not typically include specific numerical results or detailed ablations. The full manuscript substantiates all claims with quantitative comparisons (Tables 1–3), ablation studies on the alternating attention and sparse-to-dense components (Section 4.3), and error analyses (Section 4.4 and supplementary material). The referee summary already references these elements, indicating access to the complete paper where the supporting evidence appears. revision: no

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The abstract and available description outline an architectural pipeline (alternating attention for disentanglement, sparse-to-dense densification, plug-and-play motion refinement) without any equations, fitted parameters, or derivation steps that reduce to self-referential definitions or self-citations. No instances of self-definitional constructs, fitted inputs renamed as predictions, or load-bearing self-citation chains are present. The method's claims rest on described components and external experiments rather than internal circular reductions, making the derivation self-contained at the visible level.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

No free parameters, axioms, or invented entities are identifiable from the abstract alone.

pith-pipeline@v0.9.1-grok · 5735 in / 1047 out tokens · 29855 ms · 2026-06-30T06:27:46.938769+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

76 extracted references · 13 canonical work pages · 3 internal anchors

  1. [1]

    Gaussianspeech: Audio-driven gaussian avatars,

    Aneja, S., Sevastopolsky, A., Kirschstein, T., Thies, J., Dai, A., Nießner, M.: Gaus- sianspeech: Audio-driven gaussian avatars. arXiv preprint arXiv:2411.18675 (2024)

    work page arXiv 2024

  2. [2]

    In: European Conference on Computer Vision

    Athar, S., Saito, S., Yang, Z., Pidhorskyi, S., Cao, C.: Bridging the gap: Studio- like avatar creation from a monocular phone capture. In: European Conference on Computer Vision. pp. 72–88. Springer (2024)

    2024

  3. [3]

    In: Proceedings of the IEEE/CVF conference on Computer Vision and Pattern Recognition

    Athar,S.,Xu,Z.,Sunkavalli,K.,Shechtman,E.,Shu,Z.:Rignerf:Fullycontrollable neural 3d portraits. In: Proceedings of the IEEE/CVF conference on Computer Vision and Pattern Recognition. pp. 20364–20373 (2022)

    2022

  4. [4]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Bai, Z., Tan, F., Huang, Z., Sarkar, K., Tang, D., Qiu, D., Meka, A., Du, R., Dou, M., Orts-Escolano, S., et al.: Learning personalized high quality volumetric head avatars from monocular rgb videos. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 16890–16900 (2023)

    2023

  5. [5]

    In: ACM SIGGRAPH 2024 conference papers

    Chen, Y., Wang, L., Li, Q., Xiao, H., Zhang, S., Yao, H., Liu, Y.: Monogaussiana- vatar: Monocular gaussian point-based head avatar. In: ACM SIGGRAPH 2024 conference papers. pp. 1–9 (2024)

    2024

  6. [6]

    In: Proceedings of the 32nd ACM International Conference on Multimedia

    Cho, K., Lee, J., Yoon, H., Hong, Y., Ko, J., Ahn, S., Kim, S.: Gaussiantalker: Real-time talking head synthesis with 3d gaussian splatting. In: Proceedings of the 32nd ACM International Conference on Multimedia. pp. 10985–10994 (2024)

    2024

  7. [7]

    In: Proceedings of the SIGGRAPH Asia 2025 Conference Papers

    Chu, X., Goswami, N., Cui, Z., Wang, H., Harada, T.: Artalk: Speech-driven 3d head animation via autoregressive model. In: Proceedings of the SIGGRAPH Asia 2025 Conference Papers. pp. 1–9 (2025)

    2025

  8. [8]

    Advances in Neural Information Processing Systems37, 57642–57670 (2024)

    Chu, X., Harada, T.: Generalizable and animatable gaussian head avatar. Advances in Neural Information Processing Systems37, 57642–57670 (2024)

    2024

  9. [9]

    Gpavatar: Generaliz- able and precise head avatar from image (s).arXiv preprint arXiv:2401.10215, 2024

    Chu, X., Li, Y., Zeng, A., Yang, T., Lin, L., Liu, Y., Harada, T.: Gpavatar: Gener- alizable and precise head avatar from image (s). arXiv preprint arXiv:2401.10215 (2024)

    work page arXiv 2024

  10. [10]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Daněček, R., Black, M.J., Bolkart, T.: Emoca: Emotion driven monocular face capture and animation. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 20311–20322 (2022)

    2022

  11. [11]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Deng, J., Guo, J., Xue, N., Zafeiriou, S.: Arcface: Additive angular margin loss for deep face recognition. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 4690–4699 (2019)

    2019

  12. [12]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Deng, Y., Wang, D., Ren, X., Chen, X., Wang, B.: Portrait4d: Learning one-shot 4d head avatar synthesis using synthetic data. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 7119–7130 (2024)

    2024

  13. [13]

    In: European Conference on Computer Vision

    Deng, Y., Wang, D., Wang, B.: Portrait4d-v2: Pseudo multi-view data creates better 4d head synthesizer. In: European Conference on Computer Vision. pp. 316–333. Springer (2024)

    2024

  14. [14]

    In: Proceedings of the IEEE/CVF International conference on Computer Vision

    Doukas, M.C., Zafeiriou, S., Sharmanska, V.: Headgan: One-shot neural head syn- thesis and editing. In: Proceedings of the IEEE/CVF International conference on Computer Vision. pp. 14398–14407 (2021)

    2021

  15. [15]

    In: Proceed- ings of the 30th ACM International Conference on Multimedia

    Drobyshev, N., Chelishev, J., Khakhulin, T., Ivakhnenko, A., Lempitsky, V., Za- kharov, E.: Megaportraits: One-shot megapixel neural head avatars. In: Proceed- ings of the 30th ACM International Conference on Multimedia. pp. 2663–2671 (2022)

    2022

  16. [16]

    In: Proceedings of the Computer Vision and Pattern Recognition Conference

    Feng, W.Q., Han, D., Zhou, Z.K., Li, S., Liu, X., Wan, P., Zhang, D., Wang, M.: Gpavatar: High-fidelity head avatars by learning efficient gaussian projections. In: Proceedings of the Computer Vision and Pattern Recognition Conference. pp. 250–259 (2025) FFAvatar 17

    2025

  17. [17]

    ACM Transactions on Graphics (ToG)40(4), 1–13 (2021)

    Feng, Y., Feng, H., Black, M.J., Bolkart, T.: Learning an animatable detailed 3d face model from in-the-wild images. ACM Transactions on Graphics (ToG)40(4), 1–13 (2021)

    2021

  18. [18]

    Visual speech-aware perceptual 3d facial expression reconstruction from videos.arXiv preprint arXiv:2207.11094, 2022

    Filntisis, P.P., Retsinas, G., Paraperas-Papantoniou, F., Katsamanis, A., Roussos, A., Maragos, P.: Visual speech-aware perceptual 3d facial expression reconstruction from videos. arXiv preprint arXiv:2207.11094 (2022)

    work page arXiv 2022

  19. [19]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Gafni, G., Thies, J., Zollhofer, M., Nießner, M.: Dynamic neural radiance fields for monocular 4d facial avatar reconstruction. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 8649–8658 (2021)

    2021

  20. [20]

    In: Proceed- ings of the IEEE/CVF conference on computer vision and pattern recognition

    Giebenhain, S., Kirschstein, T., Georgopoulos, M., Rünz, M., Agapito, L., Nießner, M.: Mononphm: Dynamic head reconstruction from monocular videos. In: Proceed- ings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 10747–10758 (2024)

    2024

  21. [21]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Grassal, P.W., Prinzler, M., Leistner, T., Rother, C., Nießner, M., Thies, J.: Neu- ral head avatars from monocular rgb videos. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 18653–18664 (2022)

    2022

  22. [22]

    In: Proceedings of the IEEE/CVF international conference on computer vision

    Guo, Y., Chen, K., Liang, S., Liu, Y.J., Bao, H., Zhang, J.: Ad-nerf: Audio driven neural radiance fields for talking head synthesis. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 5784–5794 (2021)

    2021

  23. [23]

    In: Proceedings of the Special Interest Group on Computer Graphics and Interactive Techniques Conference Conference Papers

    He, Y., Gu, X., Ye, X., Xu, C., Zhao, Z., Dong, Y., Yuan, W., Dong, Z., Bo, L.: Lam: large avatar model for one-shot animatable gaussian head. In: Proceedings of the Special Interest Group on Computer Graphics and Interactive Techniques Conference Conference Papers. pp. 1–13 (2025)

    2025

  24. [24]

    In: Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition

    Hong, Y., Peng, B., Xiao, H., Liu, L., Zhang, J.: Headnerf: A real-time nerf-based parametric head model. In: Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition. pp. 20374–20384 (2022)

    2022

  25. [25]

    LRM: Large Reconstruction Model for Single Image to 3D

    Hong, Y., Zhang, K., Gu, J., Bi, S., Zhou, Y., Liu, D., Liu, F., Sunkavalli, K., Bui, T., Tan, H.: Lrm: Large reconstruction model for single image to 3d. arXiv preprint arXiv:2311.04400 (2023)

    work page internal anchor Pith review Pith/arXiv arXiv 2023

  26. [26]

    In: Proceedings of the 31st ACM International Conference on Multimedia

    Huang, Z., Shi, M., Liu, C., Xian, K., Cao, Z.: Simhmr: A simple query-based framework for parameterized human mesh reconstruction. In: Proceedings of the 31st ACM International Conference on Multimedia. pp. 6918–6927 (2023)

    2023

  27. [27]

    ACM Transactions on Graphics (ToG)34(4), 1–14 (2015)

    Ichim, A.E., Bouaziz, S., Pauly, M.: Dynamic 3d avatar creation from hand-held video input. ACM Transactions on Graphics (ToG)34(4), 1–14 (2015)

    2015

  28. [28]

    In: The Fourteenth International Conference on Learning Representations (2026)

    Ji, X., Weiss, S., Kansy, M., Naruniec, J., Cao, X., Solenthaler, B., Bradley, D.: Fastgha: Generalized few-shot 3d gaussian head avatars with real-time animation. In: The Fourteenth International Conference on Learning Representations (2026)

    2026

  29. [29]

    Knowledge-Based Systems320, 113470 (2025)

    Jiang, Y., Liao, Q., Li, X., Ma, L., Zhang, Q., Zhang, C., Lu, Z., Shan, Y.: Uv gaussians: Joint learning of mesh deformation and gaussian textures for human avatar modeling. Knowledge-Based Systems320, 113470 (2025)

    2025

  30. [30]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Kania, K., Yi, K.M., Kowalski, M., Trzciński, T., Tagliasacchi, A.: Conerf: Con- trollable neural radiance fields. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 18623–18632 (2022)

    2022

  31. [31]

    ACM Trans

    Kerbl, B., Kopanas, G., Leimkühler, T., Drettakis, G., et al.: 3d gaussian splatting for real-time radiance field rendering. ACM Trans. Graph.42(4), 139–1 (2023)

    2023

  32. [32]

    In: European Conference on Computer Vision

    Khakhulin, T., Sklyarova, V., Lempitsky, V., Zakharov, E.: Realistic one-shot mesh-based head avatars. In: European Conference on Computer Vision. pp. 345–

  33. [33]

    In: SIGGRAPH Asia 2024 Conference Papers

    Kirschstein, T., Giebenhain, S., Tang, J., Georgopoulos, M., Nießner, M.: Gghead: Fast and generalizable 3d gaussian heads. In: SIGGRAPH Asia 2024 Conference Papers. pp. 1–11 (2024) 18 J. Yao et al

    2024

  34. [34]

    ACM Transactions on Graphics (TOG)42(4), 1–14 (2023)

    Kirschstein, T., Qian, S., Giebenhain, S., Walter, T., Nießner, M.: Nersemble: Multi-view radiance field reconstruction of human heads. ACM Transactions on Graphics (TOG)42(4), 1–14 (2023)

    2023

  35. [35]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Kirschstein,T.,Romero,J.,Sevastopolsky,A., Nießner,M.,Saito,S.:Avat3r:Large animatable gaussian reconstruction model for high-fidelity 3d head avatars. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 12089–12100 (2025)

    2025

  36. [36]

    In: European Conference on Computer Vision

    Li, J., Zhang, J., Bai, X., Zheng, J., Ning, X., Zhou, J., Gu, L.: Talkinggaussian: Structure-persistent 3d talking head synthesis via gaussian splatting. In: European Conference on Computer Vision. pp. 127–145. Springer (2024)

    2024

  37. [37]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Li, J., Zhang, J., Bai, X., Zhou, J., Gu, L.: Efficient region-aware neural radiance fields for high-fidelity talking portrait synthesis. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 7568–7578 (2023)

    2023

  38. [38]

    ACM Trans

    Li, T., Bolkart, T., Black, M.J., Li, H., Romero, J.: Learning a model of facial shape and expression from 4d scans. ACM Trans. Graph.36(6), 194–1 (2017)

    2017

  39. [39]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Li, W., Zhang, L., Wang, D., Zhao, B., Wang, Z., Chen, M., Zhang, B., Wang, Z., Bo, L., Li, X.: One-shot high-fidelity talking-head synthesis with deformable neural radiance field. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 17969–17978 (2023)

    2023

  40. [40]

    Advances in Neural Information Processing Systems 36, 47239–47250 (2023)

    Li, X., De Mello, S., Liu, S., Nagano, K., Iqbal, U., Kautz, J.: Generalizable one- shot 3d neural head avatar. Advances in Neural Information Processing Systems 36, 47239–47250 (2023)

    2023

  41. [41]

    Feed- Forward Bullet-Time Reconstruction of Dynamic Scenes from Monocular Videos.arXiv preprint arXiv:2412.03526,

    Liang, H., Ren, J., Mirzaei, A., Torralba, A., Liu, Z., Gilitschenski, I., Fidler, S., Oztireli, C., Ling, H., Gojcic, Z., et al.: Feed-forward bullet-time reconstruction of dynamic scenes from monocular videos. arXiv preprint arXiv:2412.03526 (2024)

    work page arXiv 2024

  42. [42]

    arXiv preprint arXiv:2508.18389 (2025)

    Liang, H., Ge, Z., Majee, S., Tiwari, A., Godaliyadda, G., Veeraraghavan, A., Balakrishnan, G.: Fastavatar: Instant 3d gaussian splatting for faces from single unconstrained poses. arXiv preprint arXiv:2508.18389 (2025)

    work page arXiv 2025

  43. [43]

    In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision

    Ma, H., Zhang, T., Sun, S., Yan, X., Han, K., Xie, X.: Cvthead: One-shot con- trollable head avatar with vertex-feature transformer. In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. pp. 6131– 6141 (2024)

    2024

  44. [44]

    In: ACM SIGGRAPH 2024 Conference Papers

    Ma, S., Weng, Y., Shao, T., Zhou, K.: 3d gaussian blendshapes for head avatar animation. In: ACM SIGGRAPH 2024 Conference Papers. pp. 1–10 (2024)

    2024

  45. [45]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Ma, S., Simon, T., Saragih, J., Wang, D., Li, Y., De La Torre, F., Sheikh, Y.: Pixel codec avatars. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 64–73 (2021)

    2021

  46. [46]

    Advances in Neural Information Processing Systems37, 83008–83023 (2024)

    Martinez, J., Kim, E., Romero, J., Bagautdinov, T., Saito, S., Yu, S.I., Anderson, S., Zollhöfer, M., Wang, T.L., Bai, S., et al.: Codec avatar studio: Paired human captures for complete, driveable, and generalizable avatars. Advances in Neural Information Processing Systems37, 83008–83023 (2024)

    2024

  47. [47]

    Commu- nications of the ACM65(1), 99–106 (2021)

    Mildenhall, B., Srinivasan, P.P., Tancik, M., Barron, J.T., Ramamoorthi, R., Ng, R.: Nerf: Representing scenes as neural radiance fields for view synthesis. Commu- nications of the ACM65(1), 99–106 (2021)

    2021

  48. [48]

    ACM transactions on graphics (TOG)41(4), 1–15 (2022)

    Müller,T.,Evans,A.,Schied,C.,Keller,A.:Instantneuralgraphicsprimitiveswith a multiresolution hash encoding. ACM transactions on graphics (TOG)41(4), 1–15 (2022)

    2022

  49. [49]

    In: Proceedings of the IEEE/CVF international conference on computer vision

    Park, K., Sinha, U., Barron, J.T., Bouaziz, S., Goldman, D.B., Seitz, S.M., Martin- Brualla, R.: Nerfies: Deformable neural radiance fields. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 5865–5874 (2021) FFAvatar 19

    2021

  50. [50]

    In: Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion

    Peng, Z., Hu, W., Shi, Y., Zhu, X., Zhang, X., Zhao, H., He, J., Liu, H., Fan, Z.: Synctalk: The devil is in the synchronization for talking head synthesis. In: Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion. pp. 666–676 (2024)

    2024

  51. [51]

    arXiv preprint arXiv:2501.16617 (2025)

    Qi, D., Yang, T., Wang, B., Zhang, X., Zhang, W.: Predicting 3d representations for dynamic scenes. arXiv preprint arXiv:2501.16617 (2025)

    work page arXiv 2025

  52. [52]

    Qian, S., Kirschstein, T., Schoneveld, L., Davoli, D., Giebenhain, S., Nießner, M.: Gaussianavatars:Photorealisticheadavatarswithrigged3dgaussians.In:Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 20299–20309 (2024)

    2024

  53. [53]

    Advances in Neural Information Processing Systems37, 56828–56858 (2024)

    Ren, J., Xie, C., Mirzaei, A., Kreis, K., Liu, Z., Torralba, A., Fidler, S., Kim, S.W., Ling, H., et al.: L4gm: Large 4d gaussian reconstruction model. Advances in Neural Information Processing Systems37, 56828–56858 (2024)

    2024

  54. [54]

    FastVGGT: Training-Free Acceleration of Visual Geometry Transformer

    Shen, Y., Zhang, Z., Qu, Y., Zheng, X., Ji, J., Zhang, S., Cao, L.: Fastvggt: Training-free acceleration of visual geometry transformer. arXiv preprint arXiv:2509.02560 (2025)

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  55. [55]

    DINOv3

    Siméoni, O., Vo, H.V., Seitzer, M., Baldassarre, F., Oquab, M., Jose, C., Khali- dov, V., Szafraniec, M., Yi, S., Ramamonjisoa, M., et al.: Dinov3. arXiv preprint arXiv:2508.10104 (2025)

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  56. [56]

    In: 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

    Taubner, F., Zhang, R., Tuli, M., Lindell, D.B.: Cap4d: Creating animatable 4d portrait avatars with morphable multi-view diffusion models. In: 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). pp. 5318–5330. IEEE Computer Society (2025)

    2025

  57. [57]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recog- nition

    Tewari, A., Elgharib, M., Bharaj, G., Bernard, F., Seidel, H.P., Pérez, P., Zollhofer, M., Theobalt, C.: Stylerig: Rigging stylegan for 3d control over portrait images. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recog- nition. pp. 6142–6151 (2020)

    2020

  58. [58]

    In: Proceedings of the Computer Vision and Pattern Recognition Conference

    Wang, J., Chen, M., Karaev, N., Vedaldi, A., Rupprecht, C., Novotny, D.: Vggt: Visual geometry grounded transformer. In: Proceedings of the Computer Vision and Pattern Recognition Conference. pp. 5294–5306 (2025)

    2025

  59. [59]

    IEEE Transactions on Visualization and Computer Graphics (2025)

    Wang, J., Xie, J.C., Li, X., Xu, F., Pun, C.M., Gao, H.: Gaussianhead: High- fidelity head avatars with learnable gaussian derivation. IEEE Transactions on Visualization and Computer Graphics (2025)

    2025

  60. [60]

    IEEE transactions on image processing 13(4), 600–612 (2004)

    Wang, Z., Bovik, A.C., Sheikh, H.R., Simoncelli, E.P.: Image quality assessment: from error visibility to structural similarity. IEEE transactions on image processing 13(4), 600–612 (2004)

    2004

  61. [61]

    In: Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Wu, R., Gao, R., Poole, B., Trevithick, A., Zheng, C., Barron, J.T., Holynski, A.: Cat4d: Create anything in 4d with multi-view video diffusion models. In: Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 26057–26068 (2025)

    2025

  62. [62]

    arXiv preprint arXiv:2508.19754 (2025)

    Wu, Y., Chen, X., Wu, Y., Li, W., Lu, Y., Feng, K.: Fastavatar: Towards unified and fast 3d avatar reconstruction with large gaussian reconstruction transformers. arXiv preprint arXiv:2508.19754 (2025)

    work page arXiv 2025

  63. [63]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Xiang, J., Gao, X., Guo, Y., Zhang, J.: Flashavatar: High-fidelity head avatar with efficient gaussian embedding. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 1802–1812 (2024)

    2024

  64. [64]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Xie, L., Wang, X., Zhang, H., Dong, C., Shan, Y.: Vfhq: A high-quality dataset and benchmark for video face super-resolution. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 657–666 (2022) 20 J. Yao et al

    2022

  65. [65]

    arXiv preprint arXiv:2512.14677 (2025)

    Xu, S., Chen, G., Yang, J., Zhang, Y., Deng, Y., Lin, S., Guo, B.: Vasa-3d: Lifelike audio-driven gaussian head avatars from a single image. arXiv preprint arXiv:2512.14677 (2025)

    work page arXiv 2025

  66. [66]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Xu, Y., Chen, B., Li, Z., Zhang, H., Wang, L., Zheng, Z., Liu, Y.: Gaussian head avatar: Ultra high-fidelity head avatar via dynamic gaussians. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 1931– 1941 (2024)

    1931

  67. [67]

    STORM: Spatio-temporal reconstruction model for large-scale outdoor scenes.arXiv preprint arXiv:2501.00602,

    Yang, J., Huang, J., Chen, Y., Wang, Y., Li, B., You, Y., Sharma, A., Igl, M., Karkus, P., Xu, D., et al.: Storm: Spatio-temporal reconstruction model for large- scale outdoor scenes. arXiv preprint arXiv:2501.00602 (2024)

    work page arXiv 2024

  68. [68]

    In: European conference on computer vision

    Yang, K., Chen, K., Guo, D., Zhang, S.H., Guo, Y.C., Zhang, W.: Face2faceρ: Real-time high-resolution one-shot face reenactment. In: European conference on computer vision. pp. 55–71. Springer (2022)

    2022

  69. [69]

    In: Proceedings of the 28th ACM international conference on multimedia

    Yao, G., Yuan, Y., Shao, T., Zhou, K.: Mesh guided one-shot face reenactment using graph convolutional networks. In: Proceedings of the 28th ACM international conference on multimedia. pp. 1773–1781 (2020)

    2020

  70. [70]

    Real3d-portrait: One-shot realistic 3d talking portrait synthesis.arXiv preprint arXiv:2401.08503,

    Ye, Z., Zhong, T., Ren, Y., Yang, J., Li, W., Huang, J., Jiang, Z., He, J., Huang, R., Liu, J., et al.: Real3d-portrait: One-shot realistic 3d talking portrait synthesis. arXiv preprint arXiv:2401.08503 (2024)

    work page arXiv 2024

  71. [71]

    Advances in Neural Information Pro- cessing Systems35, 22451–22462 (2022)

    Zeng, B., Liu, B., Li, H., Liu, X., Liu, J., Chen, D., Peng, W., Zhang, B.: Fnevr: Neural volume rendering for face animation. Advances in Neural Information Pro- cessing Systems35, 22451–22462 (2022)

    2022

  72. [72]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Zhang, R., Isola, P., Efros, A.A., Shechtman, E., Wang, O.: The unreasonable effectiveness of deep features as a perceptual metric. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 586–595 (2018)

    2018

  73. [73]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Zheng, Y., Abrevaya, V.F., Bühler, M.C., Chen, X., Black, M.J., Hilliges, O.: Im avatar: Implicit morphable head avatars from videos. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 13545– 13555 (2022)

    2022

  74. [74]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Zheng, Y., Yifan, W., Wetzstein, G., Black, M.J., Hilliges, O.: Pointavatar: De- formable point-based head avatars from videos. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 21057–21067 (2023)

    2023

  75. [75]

    In: European conference on computer vision

    Zielonka, W., Bolkart, T., Thies, J.: Towards metrical reconstruction of human faces. In: European conference on computer vision. pp. 250–269. Springer (2022)

    2022

  76. [76]

    a portrait of Joe Biden

    Zielonka, W., Bolkart, T., Thies, J.: Instant volumetric head avatars. In: Proceed- ings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 4574–4584 (2023) FFAvatar 21 Supplementary Material This supplementary document provides additional implementation details, ex- perimental results, qualitative analyses, and discussions that co...

    2023