pith. sign in

arxiv: 2606.22905 · v1 · pith:NSJK4GOYnew · submitted 2026-06-22 · 💻 cs.CV

InteractiveAvatar: Real-Time Streaming Video Generation for Consistent and Intent-Aware Avatars

Quanyue Song , Yishan He , Yanfei Zhang , Shihao Cheng , Zhixiang He , Zhizhi Guo , Chi Zhang , Xuelong Li
show 1 more author
Caigui Jiang
This is my paper

Pith reviewed 2026-06-26 09:07 UTC · model grok-4.3

classification 💻 cs.CV
keywords avatar generationvideo synthesisreal-time streamingvisual consistencyintent-aware interactiondiffusion modelsautoregressive distillation
0
0 comments X

The pith

InteractiveAvatar generates visually consistent avatar videos in real time over arbitrary lengths while aligning with user intent.

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

The paper presents a framework for real-time streaming video generation of human avatars that maintains visual consistency across long durations and responds to user intent in interactive settings. It relies on autoregressive distillation to support infinite streaming without quality drop. A Long-Short Visual Memory mechanism compresses past visual data into tokens to preserve both immediate and extended coherence. A Reasoning-Reaction Module incorporates state cycling and cache switching to match avatar speech and actions to detected user goals. Experiments across scenarios position the approach as superior to prior methods in consistency and real-time interaction capability.

Core claim

InteractiveAvatar is a real-time infinite-streaming video generation framework that supports visually consistent avatar video generation and intent-aware interactions. With autoregressive distillation, it achieves real-time streaming generation of human avatars over arbitrarily long durations. For visual consistency, the Long-Short Visual Memory mechanism flexibly compresses historical visual information into compact tokens, preserving both short-range coherence and long-term consistency. To generate avatars with speeches and actions aligned with user intent, the Reasoning-Reaction Module incorporates a State-Cycling strategy and a Cache-Switching mechanism.

What carries the argument

The Long-Short Visual Memory mechanism that compresses historical visual information into compact tokens to maintain coherence, together with the Reasoning-Reaction Module that uses State-Cycling and Cache-Switching to align outputs with user intent.

If this is right

  • Achieves state-of-the-art visual consistency in long-duration generation.
  • Enables complex user-avatar interaction in real time.
  • Supports arbitrarily long avatar video streams without interruption.
  • Produces speeches and actions aligned with user intent through explicit reasoning.

Where Pith is reading between the lines

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

  • The same memory compression approach could apply to other streaming video tasks like virtual meetings or game characters.
  • Intent alignment might extend to multi-user scenarios where the system tracks several participants simultaneously.
  • Integration with language models could strengthen the reasoning step for more nuanced intent detection.
  • Real-world deployment would require testing latency under varying network conditions to confirm the real-time claim.

Load-bearing premise

The Long-Short Visual Memory and Reasoning-Reaction Module mechanisms will deliver the claimed consistency and intent alignment when implemented.

What would settle it

A side-by-side comparison of generated videos that shows visible drift in avatar appearance or clothing after extended streaming, or actions and speech that do not match explicit user commands in complex multi-turn interactions.

Figures

Figures reproduced from arXiv: 2606.22905 by Caigui Jiang, Chi Zhang, Quanyue Song, Shihao Cheng, Xuelong Li, Yanfei Zhang, Yishan He, Zhixiang He, Zhizhi Guo.

Figure 1
Figure 1. Figure 1: We propose InteractiveAvatar, a real-time streaming audio-driven avatar gen￾eration framework that enables intent-aware interaction. InteractiveAvatar interprets user intent to generate contextually relevant actions throughout the dialogue while maintaining long-range visual consistency. The RRM enhances the realism of user-avatar interaction by leveraging a large language model for intent understanding an… view at source ↗
Figure 2
Figure 2. Figure 2: Overview of InteractiveAvatar, which consists of (a) The Reasoning-Reaction Module (RRM) performs intent-aware interaction with user; (b) Streaming Inference with Long-Short Visual Memory (LSVM) mechanism to enhance the visual consistency; and (c) DMD training for real-time streaming generation. cues, with synchronized but simple gestures. Recent works [8, 19] on interac￾tive avatars have explored audio-dr… view at source ↗
Figure 3
Figure 3. Figure 3: LSVM Mechanism.(a) During training, long-term memory frames are randomly sampled, while short-term memory retains all recent frames. (b) During inference, Dynamic Key-Frame Selection adaptively updates memory to retain critical visual information. generated frames to ensure local temporal coherence, while the long-term mem￾ory stores compact representations of globally salient visual states to stabilize ov… view at source ↗
Figure 4
Figure 4. Figure 4: Qualitative comparisons with state-of-the-art methods. Our method exhibits better visual consistency and following of action instructions. and aesthetic appeal (ASE). Distribution-level fidelity is measured by FID [11] for frame-wise realism and FVD [29] for overall spatio-temporal coherence. For video consistency, we measure audio-visual synchronization using SynC and SynD [5], capturing the correspondenc… view at source ↗
Figure 5
Figure 5. Figure 5: Qualitative ablation of InteractiveAvatar. Ablation studies show that our Full model maintains the best visual consistency and enables more realistic interactions. Selection with random sampling (w/o DKFS) causes slight distortions in the watch face, highlighting the advantage of informed memory updates. Remov￾ing the entire LSVM module (w/o LSVM) leads to a significant drop in OBJ, confirming its importan… view at source ↗
read the original abstract

Recent diffusion-based models have enabled realistic audio-driven avatar generation in real-time streaming. However, existing approaches struggle to maintain visual temporal consistency and fail to explicitly perceive user intent in complex interactive streaming scenarios. To address these challenges, we propose InteractiveAvatar, a real-time infinite-streaming video generation framework that supports visually consistent avatar video generation and intent-aware interactions. With autoregressive distillation, InteractiveAvatar achieves real-time str-eaming generation of human avatars over arbitrarily long durations. For visual consistency, we introduce a Long-Short Visual Memory (LSVM) mechanism that flexibly compresses historical visual information into compact tokens, preserving both short-range coherence and long-term consistency. To generate avatars with speeches and actions aligned with user intent, we propose a Reasoning-Reaction Module (RRM), which incorporates a State-Cycling strategy and a Cache-Switching mechanism. Extensive experimental results over diverse scenarios demonstrate that our method achieves state-of-the-art visual consistency in long-duration generation, while enabling complex user-avatar interaction in real time.

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

0 major / 3 minor

Summary. The paper claims to introduce InteractiveAvatar, a real-time infinite-streaming video generation framework for consistent and intent-aware avatars. It addresses limitations in visual temporal consistency and user intent perception in diffusion-based models using autoregressive distillation, a Long-Short Visual Memory (LSVM) mechanism for compressing historical visual information to preserve short-range and long-term consistency, and a Reasoning-Reaction Module (RRM) incorporating State-Cycling and Cache-Switching for intent-aligned speeches and actions. Extensive experiments demonstrate state-of-the-art performance in long-duration generation and real-time complex interactions.

Significance. If the central claims hold, this work offers a significant contribution to the field of real-time avatar video generation by enabling arbitrarily long consistent streaming and explicit intent-aware interactions, which prior methods struggle with. The LSVM token compression and RRM strategies, supported by autoregressive distillation, provide a practical and internally consistent solution to the identified challenges. The reported experimental results over diverse scenarios add credibility to the approach, potentially advancing applications in interactive virtual environments.

minor comments (3)
  1. [Abstract] Abstract: the phrase 'real-time str-eaming generation' contains an apparent typographical error and should read 'streaming'.
  2. [Method] The description of the LSVM token compression in the method section would benefit from explicit discussion of the compression ratios used and their effect on memory usage to support reproducibility.
  3. [Experiments] Figure captions in the experimental section are often brief; expanding them to note which specific consistency or interaction aspects are visualized would improve reader comprehension.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for their positive assessment of InteractiveAvatar and for recommending minor revision. The summary correctly identifies the core challenges addressed (visual temporal consistency and intent perception) as well as the proposed solutions via autoregressive distillation, LSVM, and RRM.

Circularity Check

0 steps flagged

No circularity; architectural proposals are self-contained descriptions without derivations or self-referential reductions.

full rationale

The manuscript introduces InteractiveAvatar as a framework using autoregressive distillation plus two new modules (LSVM for memory compression and RRM with State-Cycling/Cache-Switching). No equations, fitted parameters renamed as predictions, or load-bearing self-citations appear in the provided text. The consistency and intent-alignment claims rest on the explicit design of the modules rather than any reduction to prior fitted results or self-defined quantities. The argument is therefore internally consistent and non-circular by the stated criteria.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

No free parameters, axioms, or invented entities can be identified from the abstract alone.

pith-pipeline@v0.9.1-grok · 5734 in / 999 out tokens · 18970 ms · 2026-06-26T09:07:30.161473+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

48 extracted references · 12 linked inside Pith

  1. [1]

    arXiv preprint arXiv:2506.12479 (2025)

    An, H., Hu, W., Huang, S., Huang, S., Li, R., Liang, Y., Shao, J., Song, Y., Wang, Z., Yuan, C., et al.: Ai flow: Perspectives, scenarios, and approaches (2025). arXiv preprint arXiv:2506.12479 (2025)

    arXiv 2025

  2. [2]

    arXiv preprint arXiv:2505.20156 (2025)

    Chen, Y., Liang, S., Zhou, Z., Huang, Z., Ma, Y., Tang, J., Lin, Q., Zhou, Y., Lu, Q.: Hunyuanvideo-avatar: High-fidelity audio-driven human animation for multiple characters. arXiv preprint arXiv:2505.20156 (2025)

    arXiv 2025

  3. [3]

    In: Proceedings of the AAAI Conference on Artificial Intelligence

    Chen, Z., Cao, J., Chen, Z., Li, Y., Ma, C.: Echomimic: Lifelike audio-driven portrait animations through editable landmark conditions. In: Proceedings of the AAAI Conference on Artificial Intelligence. vol. 39, pp. 2403–2410 (2025)

    2025

  4. [4]

    In: INTERSPEECH (2018)

    Chung, J.S., Nagrani, A., Zisserman, A.: Voxceleb2: Deep speaker recognition. In: INTERSPEECH (2018)

    2018

  5. [5]

    In: Asian conference on computer vision

    Chung, J.S., Zisserman, A.: Out of time: automated lip sync in the wild. In: Asian conference on computer vision. pp. 251–263. Springer (2016)

    2016

  6. [6]

    In: Proceedings of the Computer Vision and Pattern Recognition Conference

    Cui, J., Li, H., Zhan, Y., Shang, H., Cheng, K., Ma, Y., Mu, S., Zhou, H., Wang, J., Zhu, S.: Hallo3: Highly dynamic and realistic portrait image animation with video diffusion transformer. In: Proceedings of the Computer Vision and Pattern Recognition Conference. pp. 21086–21095 (2025)

    2025

  7. [7]

    arXiv preprint arXiv:2510.02283 (2025)

    Cui, J., Wu, J., Li, M., Yang, T., Li, X., Wang, R., Bai, A., Ban, Y., Hsieh, C.J.: Self-forcing++: Towards minute-scale high-quality video generation. arXiv preprint arXiv:2510.02283 (2025)

    Pith/arXiv arXiv 2025

  8. [8]

    arXiv preprint arXiv:2505.10238 (2025)

    Ding, Y., Hu, X., Guo, Z., Zhang, C., Wang, Y.: Mtvcrafter: 4d motion tokenization for open-world human image animation. arXiv preprint arXiv:2505.10238 (2025)

    arXiv 2025

  9. [9]

    arXiv preprint arXiv:2506.18866 (2025)

    Gan, Q., Yang, R., Zhu, J., Xue, S., Hoi, S.: Omniavatar: Efficient audio- driven avatar video generation with adaptive body animation. arXiv preprint arXiv:2506.18866 (2025)

    arXiv 2025

  10. [10]

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

    Ginosar, S., Bar, A., Kohavi, G., Chan, C., Owens, A., Malik, J.: Learning individ- ual styles of conversational gesture. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 3497–3506 (2019)

    2019

  11. [11]

    Advances in neural information processing systems30(2017)

    Heusel,M.,Ramsauer,H.,Unterthiner,T.,Nessler,B.,Hochreiter,S.:Ganstrained by a two time-scale update rule converge to a local nash equilibrium. Advances in neural information processing systems30(2017)

    2017

  12. [12]

    arXiv preprint arXiv:2506.08009 (2025)

    Huang, X., Li, Z., He, G., Zhou, M., Shechtman, E.: Self forcing: Bridging the train-test gap in autoregressive video diffusion. arXiv preprint arXiv:2506.08009 (2025)

    Pith/arXiv arXiv 2025

  13. [13]

    arXiv preprint arXiv:2512.04677 (2025)

    Huang, Y., Guo, H., Wu, F., Zhang, S., Huang, S., Gan, Q., Liu, L., Zhao, S., Chen, E., Liu, J., et al.: Live avatar: Streaming real-time audio-driven avatar generation with infinite length. arXiv preprint arXiv:2512.04677 (2025)

    Pith/arXiv arXiv 2025

  14. [14]

    Vicinagearth1(1), 8 (2024)

    Jiang, W., Zhang, Y., Zheng, S., Liu, S., Yan, S.: Data augmentation in human- centric vision. Vicinagearth1(1), 8 (2024)

    2024

  15. [15]

    arXiv preprint arXiv:2505.22647 (2025)

    Kong, Z., Gao, F., Zhang, Y., Kang, Z., Wei, X., Cai, X., Chen, G., Luo, W.: Let them talk: Audio-driven multi-person conversational video generation. arXiv preprint arXiv:2505.22647 (2025)

    arXiv 2025

  16. [16]

    arXiv preprint arXiv:2412.00115 (2024)

    Li, H., Xu, M., Zhan, Y., Mu, S., Li, J., Cheng, K., Chen, Y., Chen, T., Ye, M., Wang, J., et al.: Openhumanvid: A large-scale high-quality dataset for enhancing human-centric video generation. arXiv preprint arXiv:2412.00115 (2024)

    arXiv 2024

  17. [17]

    Vicinagearth1(1), 9 (2024) InteractiveAvatar 17

    Li, X., Wang, S., Zeng, S., Wu, Y., Yang, Y.: A survey on llm-based multi-agent systems: workflow, infrastructure, and challenges. Vicinagearth1(1), 9 (2024) InteractiveAvatar 17

    2024

  18. [18]

    IEEE Transactions on Neural Networks and Learn- ing Systems35(6), 8708–8714 (2022)

    Li, X.: Positive-incentive noise. IEEE Transactions on Neural Networks and Learn- ing Systems35(6), 8708–8714 (2022)

    2022

  19. [19]

    arXiv preprint arXiv:2506.03099 (2025)

    Low, C., Wang, W.: Talkingmachines: Real-time audio-driven facetime-style video via autoregressive diffusion models. arXiv preprint arXiv:2506.03099 (2025)

    arXiv 2025

  20. [20]

    arXiv preprint arXiv:2507.03905 (2025)

    Meng, R., Wang, Y., Wu, W., Zheng, R., Li, Y., Ma, C.: Echomimicv3: 1.3 b pa- rameters are all you need for unified multi-modal and multi-task human animation. arXiv preprint arXiv:2507.03905 (2025)

    arXiv 2025

  21. [21]

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

    Ng, E., Joo, H., Hu, L., Li, H., Darrell, T., Kanazawa, A., Ginosar, S.: Learning to listen: Modeling non-deterministic dyadic facial motion. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 20395– 20405 (2022)

    2022

  22. [22]

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

    Prajwal, K., Mukhopadhyay, R., Namboodiri, V.P., Jawahar, C.: 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. pp. 484–492 (2020)

    2020

  23. [23]

    Journal of machine learning research21(140), 1–67 (2020)

    Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., Zhou, Y., Li, W., Liu, P.J.: Exploring the limits of transfer learning with a unified text-to-text transformer. Journal of machine learning research21(140), 1–67 (2020)

    2020

  24. [24]

    Siméoni, O., Vo, H.V., Seitzer, M., Baldassarre, F., Oquab, M., Jose, C., Khalidov, V., Szafraniec, M., Yi, S., Ramamonjisoa, M., Massa, F., Haziza, D., Wehrstedt, L., Wang, J., Darcet, T., Moutakanni, T., Sentana, L., Roberts, C., Vedaldi, A., Tolan, J., Brandt, J., Couprie, C., Mairal, J., Jégou, H., Labatut, P., Bojanowski, P.: DINOv3 (2025),https://ar...

    Pith/arXiv arXiv 2025

  25. [25]

    arXiv preprint arXiv:2512.22065 (2025)

    Sun, Z., Peng, Z., Ma, Y., Chen, Y., Zhou, Z., Zhou, Z., Zhang, G., Zhang, Y., Zhou, Y., Lu, Q., et al.: Streamavatar: Streaming diffusion models for real-time interactive human avatars. arXiv preprint arXiv:2512.22065 (2025)

    arXiv 2025

  26. [26]

    In: European Conference on Computer Vision

    Tian, L., Wang, Q., Zhang, B., Bo, L.: Emo: Emote portrait alive generating ex- pressive portrait videos with audio2video diffusion model under weak conditions. In: European Conference on Computer Vision. pp. 244–260. Springer (2024)

    2024

  27. [27]

    arXiv preprint arXiv:2502.14786 (2025)

    Tschannen, M., Gritsenko, A., Wang, X., Naeem, M.F., Alabdulmohsin, I., Parthasarathy, N., Evans, T., Beyer, L., Xia, Y., Mustafa, B., et al.: Siglip 2: Multilingual vision-language encoders with improved semantic understanding, lo- calization, and dense features. arXiv preprint arXiv:2502.14786 (2025)

    Pith/arXiv arXiv 2025

  28. [28]

    Tu, S., Pan, Y., Huang, Y., Han, X., Xing, Z., Dai, Q., Luo, C., Wu, Z., Jiang, Y.G.: Stableavatar: Infinite-length audio-driven avatar video generation (2025), https://arxiv.org/abs/2508.08248

    arXiv 2025

  29. [29]

    arXiv preprint arXiv:1812.01717 (2018)

    Unterthiner, T., Van Steenkiste, S., Kurach, K., Marinier, R., Michalski, M., Gelly, S.: Towards accurate generative models of video: A new metric & challenges. arXiv preprint arXiv:1812.01717 (2018)

    Pith/arXiv arXiv 2018

  30. [30]

    arXiv preprint arXiv:2503.20314 (2025)

    Wan, T., Wang, A., Ai, B., Wen, B., Mao, C., Xie, C.W., Chen, D., Yu, F., Zhao, H., Yang, J., et al.: Wan: Open and advanced large-scale video generative models. arXiv preprint arXiv:2503.20314 (2025)

    Pith/arXiv arXiv 2025

  31. [31]

    Wang, J., Wang, C., Huang, K., Huang, J., Jin, L.: Videoclip-xl: Advancing long description understanding for video clip models (2024),https://arxiv.org/abs/ 2410.00741

    arXiv 2024

  32. [32]

    arXiv preprint arXiv:2601.10103 (2026)

    Wang, L., Zhu, Y., Ge, Z., Zheng, Y., Zhang, L., Hu, T., Qin, S., Luo, M., Zhang, J., Chen, X., et al.: Flowact-r1: Towards interactive humanoid video generation. arXiv preprint arXiv:2601.10103 (2026)

    arXiv 2026

  33. [33]

    In: Proceedings of the 33rd ACM International Conference on Multimedia

    Wang, M., Wang, Q., Jiang, F., Fan, Y., Zhang, Y., Qi, Y., Zhao, K., Xu, M.: Fantasytalking: Realistic talking portrait generation via coherent motion synthesis. In: Proceedings of the 33rd ACM International Conference on Multimedia. pp. 9891–9900 (2025) 18 Q. Song et al

    2025

  34. [34]

    In: Proceedings of the Computer Vision and Pattern Recognition Conference

    Wang, Y., Fan, Y., Wang, X., Yu, G., Wang, F.: Diffusion-based realistic listening head generation via hybrid motion modeling. In: Proceedings of the Computer Vision and Pattern Recognition Conference. pp. 15885–15895 (2025)

    2025

  35. [35]

    arXiv preprint arXiv:2312.17090 (2023)

    Wu, H., Zhang, Z., Zhang, W., Chen, C., Liao, L., Li, C., Gao, Y., Wang, A., Zhang, E., Sun, W., et al.: Q-align: Teaching lmms for visual scoring via discrete text-defined levels. arXiv preprint arXiv:2312.17090 (2023)

    Pith/arXiv arXiv 2023

  36. [36]

    In: The IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) (2022)

    Xie, L., Wang, X., Zhang, H., Dong, C., Shan, Y.: Vfhq: A high-quality dataset and benchmark for video face super-resolution. In: The IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) (2022)

    2022

  37. [37]

    arXiv preprint arXiv:2509.21574 (2025)

    Xie, Y., Gu, T., Li, Z., Zhang, C., Song, G., Zhao, X., Liang, C., Jiang, J., Xu, H., Luo, L.: X-streamer: Unified human world modeling with audiovisual interaction. arXiv preprint arXiv:2509.21574 (2025)

    arXiv 2025

  38. [38]

    arXiv preprint arXiv:2509.17765 (2025)

    Xu, J., Guo, Z., Hu, H., Chu, Y., Wang, X., He, J., Wang, Y., Shi, X., He, T., Zhu, X., et al.: Qwen3-omni technical report. arXiv preprint arXiv:2509.17765 (2025)

    Pith/arXiv arXiv 2025

  39. [39]

    Advances in Neural Information Processing Systems37, 660–684 (2024)

    Xu, S., Chen, G., Guo, Y.X., Yang, J., Li, C., Zang, Z., Zhang, Y., Tong, X., Guo, B.: Vasa-1: Lifelike audio-driven talking faces generated in real time. Advances in Neural Information Processing Systems37, 660–684 (2024)

    2024

  40. [40]

    arXiv preprint arXiv:2509.22622 (2025)

    Yang, S., Huang, W., Chu, R., Xiao, Y., Zhao, Y., Wang, X., Li, M., Xie, E., Chen, Y., Lu, Y., et al.: Longlive: Real-time interactive long video generation. arXiv preprint arXiv:2509.22622 (2025)

    Pith/arXiv arXiv 2025

  41. [41]

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

    Yin, T., Zhang, Q., Zhang, R., Freeman, W.T., Durand, F., Shechtman, E., Huang, X.: From slow bidirectional to fast autoregressive video diffusion models. In: Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion. pp. 22963–22974 (2025)

    2025

  42. [42]

    In: CVPR (2023)

    Yu, J., Zhu, H., Jiang, L., Loy, C.C., Cai, W., Wu, W.: CelebV-Text: A large-scale facial text-video dataset. In: CVPR (2023)

    2023

  43. [43]

    IEEE Transactions on Pattern Analysis and Machine Intel- ligence (2025)

    Zhang, H., Huang, S., Guo, Y., Li, X.: Variational positive-incentive noise: How noise benefits models. IEEE Transactions on Pattern Analysis and Machine Intel- ligence (2025)

    2025

  44. [44]

    arXiv preprint arXiv:2512.23851 (2025)

    Zhang, L., Cai, S., Li, M., Zeng, C., Lu, B., Rao, A., Han, S., Wetzstein, G., Agrawala, M.: Pretraining frame preservation in autoregressive video memory com- pression. arXiv preprint arXiv:2512.23851 (2025)

    Pith/arXiv arXiv 2025

  45. [45]

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

    Zhang, W., Cun, X., Wang, X., Zhang, Y., Shen, X., Guo, Y., Shan, Y., Wang, F.: 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 pattern recognition. pp. 8652–8661 (2023)

    2023

  46. [46]

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

    Zhang, Z., Li, L., Ding, Y., Fan, C.: Flow-guided one-shot talking face generation with a high-resolution audio-visual dataset. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 3661–3670 (2021)

    2021

  47. [47]

    arXiv preprint arXiv:2304.11277 (2023)

    Zhao,Y.,Gu,A.,Varma,R.,Luo,L.,Huang,C.C.,Xu,M.,Wright,L.,Shojanazeri, H., Ott, M., Shleifer, S., et al.: Pytorch fsdp: experiences on scaling fully sharded data parallel. arXiv preprint arXiv:2304.11277 (2023)

    Pith/arXiv arXiv 2023

  48. [48]

    ACM Transactions On Graph- ics (TOG)39(6), 1–15 (2020)

    Zhou, Y., Han, X., Shechtman, E., Echevarria, J., Kalogerakis, E., Li, D.: Makelttalk: speaker-aware talking-head animation. ACM Transactions On Graph- ics (TOG)39(6), 1–15 (2020)

    2020