Pith. sign in

REVIEW 4 major objections 5 minor 3 cited by

LIA-X: Interpretable Latent Portrait Animator

T0 review · 4 major / 5 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read LIA-X makes portrait animation controllable by learning a sparse motion dictionary whose vectors are human-interpretable, enabling an 'edit-warp-render' pipeline that improves reenactment quality.

desk verdict LIA-X is a simple, plausible sparsity tweak on LIA; the interpretability story is supported mostly by cherry-picked visuals and the SOTA comparisons are confounded by much larger training data. read the letter →

arxiv 2508.09959 v1 pith:3R3TDXQ2 submitted 2025-08-13 cs.CV

Yaohui Wang , Di Yang , Xinyuan Chen , Francois Bremond , Yu Qiao , Antitza Dantcheva This is my paper
classification cs.CV
keywords portraitanimationlatentimageanimatorsparsemotiondictionaryinterpretablerepresentationseditingvideo3D-awaremanipulationscalability
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

LIA-X is an autoencoder-based portrait animator that models facial motion as linear navigation through a learned sparse motion dictionary. The paper argues that adding an L1 sparsity penalty to the motion coefficients forces each reconstruction to use few motion vectors, and that the resulting vectors become interpretable, corresponding to semantic factors like head yaw, pitch, roll, mouth, eyes, and eyebrows. This interpretability supports a novel edit-warp-render strategy: before transferring motion from a driving video, the user can edit the source portrait using the learned vectors to align its pose and expression with the driving frame. The paper claims this reduces the large pose/expression gap that degrades prior warp-render methods, and that LIA-X outperforms state-of-the-art approaches on self-reenactment and cross-reenactment benchmarks. It also demonstrates that the architecture scales to about one billion parameters with improved performance.

What carries the argument

The Sparse Motion Dictionary $D_m = \{d_1, \dots, d_M\}$ is a set of orthogonal motion vectors in latent space, with an L1 sparsity penalty $S(A_{r\to d})$ on the coefficients used to reconstruct each image. The sparsity constraint is the central mechanism: it forces the network to use a minimal set of motion vectors for each reconstruction, which empirically disentangles facial dynamics into interpretable factors. The linear editing operation $z_{s\to E(s)} = z_{s\to s} + a_i d_i$ then leverages these vectors for direct, user-controllable manipulation of the source portrait before animation.

What would settle it

A quantitative editing fidelity test would settle the claim: apply Eq. (11) with a 'smile' vector to a large, diverse set of neutral-expression source portraits, then measure smile intensity with a pretrained facial attribute classifier. If the edit succeeds for only a small fraction of identities or produces inconsistent magnitudes, the linear editing assumption fails. Alternatively, compare animation quality (PSNR, LPIPS, identity similarity) with and without the pre-editing step across many source-driving pairs with large pose/expression differences; if the improvement is not consistent and

Watch

Extended reading notes

Core claim

LIA-X discovers that applying an L1 sparsity constraint to the motion coefficient vector $A_{r\to d}$ in a latent autoencoder produces a Sparse Motion Dictionary whose individual vectors are interpretable. During self-supervised training, this penalty encourages the network to reconstruct each frame by activating only a few motion vectors; at inference, these vectors can be manipulated directly. Editing the source portrait is formulated as $z_{s\to E(s)} = z_{s\to s} + a_i d_i$, adding a scaled learned motion vector to the source's own reconstruction code. This edit is then followed by the standard warp-render animation step, yielding the 'edit-warp-render' pipeline. The paper claims this al

Load-bearing premise

The edit-warp-render advantage rests on the assumption that simply adding a scaled learned motion vector to the source's own reconstruction code (Eq. 11) reliably edits the portrait's expression or pose; this linear editing property is never trained for or quantitatively evaluated, and the examples shown were selected post hoc from vectors that happened to work.

Editorial extensions

If this is right

  • Portrait animation becomes a two-stage process: first edit the source to match the driving frame's pose and expression, then warp-render; this narrows the initial gap that degrades prior methods.
  • Users can perform fine-grained image editing (e.g., open/close eyes, smile, frown) by adjusting individual motion vectors, and video editing by applying the same linear edit to a video's first frame and then animating.
  • The model exhibits 3D-aware manipulation of yaw, pitch, and roll without any explicit 3D representation, since these transforms emerge as interpretable motion vectors.
  • Scaling the autoencoder from 0.05B to 0.9B parameters monotonically improves reconstruction metrics on VoxCelebHQ and TalkingHead-1KH, though gains diminish at the largest size.
  • Autoencoder-based latent navigation provides faster inference than diffusion-based animators, making LIA-X a practical complement for controllable generation.

Reading between the lines

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

  • The linear editing property of Eq. (11) may generalize beyond the qualitative examples: if the motion dictionary is truly interpretable, then composing vectors (e.g., 'smile' + 'yaw') should produce combined edits, and editing should work consistently across arbitrary source identities. This is testable with attribute classifiers on a large set of portraits.
  • The sparsity-induced interpretability might extend to other video-to-video tasks such as full-body animation or speech-driven avatars, where a sparse motion dictionary could reveal interpretable control axes for body pose or lip sync.
  • The diminishing returns from 0.3B to 0.9B parameters suggest the model is dataset-limited rather than architecture-limited; training on larger, more diverse datasets may unlock further gains, an explicit hypothesis the paper leaves for future work.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The paper introduces LIA-X, a self-supervised autoencoder for portrait animation built on LIA's linear latent navigation. The main novelty is a 'Sparse Motion Dictionary': an L1 penalty on the motion coefficients in the training loss (Eq. 9) is claimed to produce semantically disentangled, interpretable motion vectors. At inference, users can 'edit' a source portrait by adding a scaled dictionary vector to its self-reconstruction code (Eq. 11), then animate with an 'edit-warp-render' formula (Eq. 10). The paper reports state-of-the-art quantitative results on self-reenactment (Table 1) and cross-reenactment (Table 2), a scalability study up to 0.9B parameters (Tables 3-4), and qualitative demonstrations of editing and 3D-aware manipulation.

Significance. If the interpretability and controllability claims hold, LIA-X would be a meaningful step toward fast, controllable portrait animation without explicit 3D representations. The sparse-dictionary idea is simple and potentially useful, and the large-scale training study is a concrete engineering contribution. However, the central evidence is observational: the editing operation in Eq. (11) is never trained for or quantitatively measured, no controlled ablations isolate the effect of the sparsity penalty, and the quantitative comparisons are confounded by substantially larger model and dataset scale. The significance is therefore currently conditional rather than demonstrated.

major comments (4)
  1. [Section 5.2, Eq. (11)] The central editing operation is neither trained for nor quantitatively validated. The training loss (Eq. 9) supervises reconstruction of x_{s->d} and adds L1 on coefficients; nothing enforces that z_{s->s}+a_i d_i changes only the named semantic attribute, nor that the edit transfers across identities. The paper states 'Surprisingly, we found...' and shows selected examples. This is post-hoc observational evidence. Please add quantitative tests: attribute classifiers (pose angles, mouth/eye openness) on edited outputs, monotonicity of attribute magnitude with a_i, cross-identity consistency, and negative controls (random directions, a dense dictionary, or random sparse directions). Without such tests, the 'interpretable Sparse Motion Dictionary' claim is unsupported.
  2. [Section 4.3, Eq. (10)] The edit-warp-render formula is under-specified. z_{s->E(s)} is defined via Eq. (11), but no edited image x_{E(s)} is ever rendered or used as an actual source. Eq. (10) composes the original source's self-reconstruction code with the original driving motion difference. Please specify: (i) how x_{E(s)} is obtained, (ii) how a_i is chosen automatically or by the user to match the driving frame, and (iii) why substituting z_{s->E(s)} for z_{s->s} is equivalent to animating a genuinely edited portrait. If this is an inference-time heuristic, that should be stated and validated separately from the reconstruction training.
  3. [Tables 1 and 2] The reported comparisons are confounded by training scale and data. LIA-X is trained on 0.5M sequences / 94M frames / 55k identities with up to 0.9B parameters, whereas the baselines use smaller models and datasets. The wins do not isolate the sparse dictionary or the edit-warp-render mechanism. Provide ablations with matched data and model size: LIA-X with and without the sparsity penalty, with a dense dictionary, and with editing disabled. Also report confidence intervals or repeated-run variability. Table 2's Identity Similarity and no-reference Image Quality metrics do not measure semantic correctness of motion transfer; include pose/expression alignment metrics or a user study to attribute the gain to editing.
  4. [Section 5.1, Fig. 3] Sparsity is demonstrated on only two videos and with no quantitative metric (e.g., fraction of active coefficients, entropy, dictionary utilization). More importantly, coefficient sparsity is neither necessary nor sufficient for semantic disentanglement; a random sparse dictionary can also activate few vectors. Please report dictionary statistics over the validation set and show that each vector corresponds to the same semantic attribute across many identities and source images, rather than only the curated examples in Figs. 4-6. The labels attached to motion vectors (yaw, pitch, smile, etc.) appear to be assigned post hoc to vectors selected for illustration.
minor comments (5)
  1. [Figure 2 caption] Typo: 'enrouraging' should be 'encouraging'.
  2. [Eq. (5)] The perceptual loss weight is written λ without a subscript, while Eq. (9) uses λ1 and λ2. Align the notation.
  3. [Section 5.2] The perturbation range a_i ∈ [-0.5, 0.5] is given without context: no typical coefficient magnitudes or normalization of dictionary vectors are reported, so 'small perturbation' is undefined.
  4. [Tables 3 and 4] The improvements from Middle (0.3B) to Large (0.9B) are within 0.01 LPIPS and 0.08 PSNR on VoxCelebHQ; the statement that these results 'clearly demonstrate' the effectiveness of scaling is overstated without error bars or statistical testing.
  5. [Section 5.3] The text says 'To ensure a fair comparison, we train LIA-X at two different resolutions,' but LIA-X is also trained on substantially more data and with much larger capacity than the baselines. A fair comparison is not established by matching resolution alone.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the sparse dictionary and edit-warp-render contributions are new and evaluated against external baselines; the post-hoc interpretability evidence is a validation gap, not a circular derivation.

full rationale

LIA-X builds on the authors' prior LIA work, but the core new claims—the sparse motion dictionary and the edit-warp-render strategy—are introduced and evaluated in this paper rather than imported by citation. The sparsity penalty S(A_{r->d}) = L1 is a new regularizer added to the reconstruction objective (Eq. 9); the claim that it yields interpretable vectors is supported by activation visualizations (Fig. 3) and by qualitative manipulation (Figs. 4–6). The editing equation (11), z_{s->E(s)} = z_{s->s} + a_i d_i, defines an inference-time operation on the learned dictionary; it is an extrapolation, not a fitted parameter renamed as a prediction. No equation reduces the predicted output to the training target by construction: self-reenactment is standard reconstruction on held-out frames, and cross-reenactment has no ground truth and is evaluated via identity similarity and image quality. The self-citations to LIA provide background architecture and a baseline, and LIA itself is externally published; the load-bearing evidence for LIA-X's improvement is the quantitative comparison in Tables 1–2, not the self-citation. The main weakness—that semantic labels for motion vectors are assigned post hoc from the same examples used to illustrate them—is an empirical validation concern, not a circular derivation. Therefore no circular step is present.

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

The central method inherits the linear navigation assumption from LIA and adds a sparsity regularization whose interpretability benefit is taken as a hypothesis rather than derived. The user-facing semantic labels are invented after training and lack independent validation. The free parameters, especially lambda_2 and dictionary size, are not reported, which hampers replication.

free parameters (5)
  • lambda_2 (sparsity penalty weight) = not reported
    Controls the strength of the L1 penalty on motion coefficients (Eq. 9); central to the claimed sparsity and interpretability, but its value is not given.
  • lambda_1 (perceptual loss weight) = not reported
    Weight for the VGG perceptual loss in Eq. 9; not specified in the text.
  • Motion dictionary size M = larger than LIA, exact value not reported
    The number of motion vectors is a design choice that directly affects how fine-grained the semantics can be, but the exact M is not stated.
  • Perturbation range a_i for editing = [-0.5, 0.5] with step 0.1
    User-chosen range for semantic editing in Eq. 11; the paper does not show sensitivity to this choice.
  • Architecture widths/depths for Base, Middle, Large = 0.05B, 0.3B, 0.9B parameters
    The scaling study varies residual block counts and channel numbers, but exact architectural configurations are not specified.
assumptions (4)
  • domain assumption Any image transformation can be represented as a linear navigation in latent space via an implicit reference image (Eqs. 1-3).
    Inherited from LIA without proof in this paper; the whole edit-warp-render approach depends on this linearity.
  • ad hoc to paper An L1 sparsity penalty on motion coefficients produces semantically disentangled, human-interpretable vectors.
    This is the paper's central hypothesis, asserted in Section 4.2 and evidenced only by qualitative examples.
  • domain assumption Self-supervised reconstruction with L1, VGG, and adversarial losses is sufficient to learn transferable and editable motion codes.
    Standard for this line of work, but no theoretical or quantitative justification is given for editability.
  • domain assumption The mixed public plus internal training set is diverse and large enough to support generalization and fair comparison.
    The internal dataset is not described, and baselines were likely trained with different data, which affects the validity of comparisons.
invented entities (1)
  • Semantic labels attached to motion vectors (e.g., yaw, pitch, roll, smile, open/close mouth)
    purpose: Enable user-guided editing by manipulating specific learned vectors via Eq. 11.
    Labels are assigned post hoc by visual inspection of selected examples; no external benchmark or user study verifies that each vector corresponds stably to the same semantic across identities.

how reviews work

0 comments
Cite this review

Pith. "Pith review of LIA-X: Interpretable Latent Portrait Animator." pith.science (2026). https://pith.science/paper/3R3TDXQ2

@misc{pith2026250809959,
  author       = {Pith},
  title        = {Pith review of: LIA-X: Interpretable Latent Portrait Animator},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3R3TDXQ2}},
  note         = {Machine review of arXiv:2508.09959}
}
read the original abstract

We introduce LIA-X, a novel interpretable portrait animator designed to transfer facial dynamics from a driving video to a source portrait with fine-grained control. LIA-X is an autoencoder that models motion transfer as a linear navigation of motion codes in latent space. Crucially, it incorporates a novel Sparse Motion Dictionary that enables the model to disentangle facial dynamics into interpretable factors. Deviating from previous 'warp-render' approaches, the interpretability of the Sparse Motion Dictionary allows LIA-X to support a highly controllable 'edit-warp-render' strategy, enabling precise manipulation of fine-grained facial semantics in the source portrait. This helps to narrow initial differences with the driving video in terms of pose and expression. Moreover, we demonstrate the scalability of LIA-X by successfully training a large-scale model with approximately 1 billion parameters on extensive datasets. Experimental results show that our proposed method outperforms previous approaches in both self-reenactment and cross-reenactment tasks across several benchmarks. Additionally, the interpretable and controllable nature of LIA-X supports practical applications such as fine-grained, user-guided image and video editing, as well as 3D-aware portrait video manipulation.

Discussion (0). Sign in to comment.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Talker-T2AV: Joint Talking Audio-Video Generation with Autoregressive Diffusion Modeling

    cs.CV 2026-04 unverdicted novelty 7.0 of 10

    Talker-T2AV achieves better lip-sync accuracy, video quality, and audio quality than dual-branch baselines by separating high-level shared autoregressive modeling from modality-specific low-level diffusion refinement ...

  2. Instant Expressive Gaussian Head Avatars at Over 100 FPS

    cs.CV 2025-12 conditional novelty 7.0 of 10

    A single-photo avatar encoder with per-Gaussian feature-space deformation animates faces at 107 FPS with expression quality competitive with diffusion models.

  3. THEval. Evaluation Framework for Talking Head Video Generation

    cs.CV 2025-11 conditional novelty 6.0 of 10

    THEval proposes eight metrics for evaluating talking head videos on quality, naturalness, and synchronization, tested on 85,000 videos from 17 models with a new curated dataset.

Reference graph

Works this paper leans on

58 extracted references · 46 canonical work pages · cited by 3 Pith papers

  1. [1]

    Align your latents: High-resolution video synthesis with la- tent diffusion models

    Andreas Blattmann, Robin Rombach, Huan Ling, Tim Dock- horn, Seung Wook Kim, Sanja Fidler, and Karsten Kreis. Align your latents: High-resolution video synthesis with la- tent diffusion models. In CVPR, 2023. 2

    work page 2023

  2. [2]

    Generating long videos of dynamic scenes

    Tim Brooks, Janne Hellsten, Miika Aittala, Ting-Chun Wang, Timo Aila, Jaakko Lehtinen, Ming-Yu Liu, Alexei A Efros, and Tero Karras. Generating long videos of dynamic scenes. 2022

    work page 2022

  3. [3]

    Video generation models as world simulators

    Tim Brooks, Bill Peebles, Connor Holmes, Will DePue, Yufei Guo, Li Jing, David Schnurr, Joe Taylor, Troy Luh- man, Eric Luhman, Clarence Ng, Ricky Wang, and Aditya Ramesh. Video generation models as world simulators

  4. [4]

    Everybody dance now

    Caroline Chan, Shiry Ginosar, Tinghui Zhou, and Alexei A Efros. Everybody dance now. In ICCV, 2019. 2

    work page 2019

  5. [5]

    Magicdance: Realistic human dance video gen- eration with motions & facial expressions transfer

    Di Chang, Yichun Shi, Quankai Gao, Jessica Fu, Hongyi Xu, Guoxian Song, Qing Yan, Xiao Yang, and Mohammad So- leymani. Magicdance: Realistic human dance video gen- eration with motions & facial expressions transfer. arXiv preprint arXiv:2311.12052, 2023. 2

    arXiv 2023

  6. [6]

    Joint generative and con- trastive learning for unsupervised person re-identification

    Hao Chen, Yaohui Wang, Benoit Lagadec, Antitza Dantcheva, and Francois Bremond. Joint generative and con- trastive learning for unsupervised person re-identification. In CVPR, 2021. 2

    work page 2021

  7. [7]

    Videocrafter2: Overcoming data limitations for high-quality video diffusion models

    Haoxin Chen, Yong Zhang, Xiaodong Cun, Menghan Xia, Xintao Wang, Chao Weng, and Ying Shan. Videocrafter2: Overcoming data limitations for high-quality video diffusion models. In CVPR, 2024. 2

    work page 2024

  8. [8]

    Seine: Short-to-long video diffusion model for generative transition and prediction

    Xinyuan Chen, Yaohui Wang, Lingjun Zhang, Shaobin Zhuang, Xin Ma, Jiashuo Yu, Yali Wang, Dahua Lin, Yu Qiao, and Ziwei Liu. Seine: Short-to-long video diffusion model for generative transition and prediction. In ICLR, 2023

    work page 2023

Show all 58 references
  1. [9]

    Adver- sarial video generation on complex datasets

    Aidan Clark, Jeff Donahue, and Karen Simonyan. Adver- sarial video generation on complex datasets. arXiv preprint arXiv:1907.06571, 2019. 2

    1907 arXiv

  2. [10]

    Generative adversarial nets

    Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. Generative adversarial nets. In NIPS, 2014. 2

    2014

  3. [11]

    Livepor- trait: Efficient portrait animation with stitching and retarget- ing control

    Jianzhu Guo, Dingyun Zhang, Xiaoqiang Liu, Zhizhou Zhong, Yuan Zhang, Pengfei Wan, and Di Zhang. Livepor- trait: Efficient portrait animation with stitching and retarget- ing control. arXiv preprint arXiv:2407.03168, 2024. 2, 8

    2024 arXiv

  4. [12]

    Animatediff: Animate your personalized text-to- image diffusion models without specific tuning

    Yuwei Guo, Ceyuan Yang, Anyi Rao, Zhengyang Liang, Yaohui Wang, Yu Qiao, Maneesh Agrawala, Dahua Lin, and Bo Dai. Animatediff: Animate your personalized text-to- image diffusion models without specific tuning. Interna- tional Conference on Learning Representations, 2024. 2 9

    2024

  5. [13]

    Denoising diffu- sion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffu- sion probabilistic models. Advances in Neural Information Processing Systems, 33:6840–6851, 2020. 2

    2020

  6. [14]

    Imagen video: High definition video generation with diffusion mod- els

    Jonathan Ho, William Chan, Chitwan Saharia, Jay Whang, Ruiqi Gao, Alexey Gritsenko, Diederik P Kingma, Ben Poole, Mohammad Norouzi, David J Fleet, et al. Imagen video: High definition video generation with diffusion mod- els. arXiv preprint arXiv:2210.02303, 2022. 2

    2022 arXiv

  7. [15]

    Implicit identity representation conditioned memory compensation network for talking head video generation

    Fa-Ting Hong and Dan Xu. Implicit identity representation conditioned memory compensation network for talking head video generation. In ICCV, 2023. 8

    2023

  8. [16]

    Depth-aware generative adversarial network for talking head video generation

    Fa-Ting Hong, Longhao Zhang, Li Shen, and Dan Xu. Depth-aware generative adversarial network for talking head video generation. 2022. 2, 8

    2022

  9. [17]

    Flow-grounded spatial-temporal video prediction from still images

    Yijun Li, Chen Fang, Jimei Yang, Zhaowen Wang, Xin Lu, and Ming-Hsuan Yang. Flow-grounded spatial-temporal video prediction from still images. In ECCV, 2018. 2

    2018

  10. [18]

    Blendgan: Implicitly gan blending for arbitrary stylized face generation

    Mingcong Liu, Qiang Li, Zekui Qin, Guoxin Zhang, Pengfei Wan, and Wen Zheng. Blendgan: Implicitly gan blending for arbitrary stylized face generation. In Advances in Neural Information Processing Systems, 2021. 9

    2021

  11. [19]

    Liquid warping gan: A unified framework for human motion imitation, appearance transfer and novel view synthesis

    Wen Liu, Zhixin Piao, Jie Min, Wenhan Luo, Lin Ma, and Shenghua Gao. Liquid warping gan: A unified framework for human motion imitation, appearance transfer and novel view synthesis. In CVPR, 2019. 2

    2019

  12. [20]

    Latte: La- tent diffusion transformer for video generation.Transactions on Machine Learning Research, 2025

    Xin Ma, Yaohui Wang, Xinyuan Chen, Gengyun Jia, Ziwei Liu, Yuan-Fang Li, Cunjian Chen, and Yu Qiao. Latte: La- tent diffusion transformer for video generation.Transactions on Machine Learning Research, 2025. 2

    2025

  13. [21]

    Follow-your-emoji: Fine-controllable and expressive freestyle portrait animation

    Yue Ma, Hongyu Liu, Hongfa Wang, Heng Pan, Yingqing He, Junkun Yuan, Ailing Zeng, Chengfei Cai, Heung-Yeung Shum, Wei Liu, et al. Follow-your-emoji: Fine-controllable and expressive freestyle portrait animation. arXiv preprint arXiv:2406.01900, 2024. 2

    2024 arXiv

  14. [22]

    Snap video: Scaled spatiotemporal transformers for text-to-video synthesis

    Willi Menapace, Aliaksandr Siarohin, Ivan Skorokhodov, Ekaterina Deyneka, Tsai-Shien Chen, Anil Kag, Yuwei Fang, Aleksei Stoliar, Elisa Ricci, Jian Ren, et al. Snap video: Scaled spatiotemporal transformers for text-to-video synthesis. In CVPR, 2024. 2

    2024

  15. [23]

    Hierarchical video generation from or- thogonal information: Optical flow and texture

    Katsunori Ohnishi, Shohei Yamamoto, Yoshitaka Ushiku, and Tatsuya Harada. Hierarchical video generation from or- thogonal information: Optical flow and texture. In AAAI,

  16. [24]

    Emergence of simple- cell receptive field properties by learning a sparse code for natural images

    Bruno A Olshausen and David J Field. Emergence of simple- cell receptive field properties by learning a sparse code for natural images. Nature, 381(6583):607–609, 1996. 2

    1996

  17. [25]

    Scalable diffusion models with transformers

    William Peebles and Saining Xie. Scalable diffusion models with transformers. arXiv preprint arXiv:2212.09748, 2022. 9

    2022 arXiv

  18. [26]

    Tempo- ral generative adversarial nets with singular value clipping

    Masaki Saito, Eiichi Matsumoto, and Shunta Saito. Tempo- ral generative adversarial nets with singular value clipping. In ICCV, 2017. 2

    2017

  19. [27]

    StyleGAN-T: Unlocking the power of GANs for fast large-scale text-to-image synthesis

    Axel Sauer, Tero Karras, Samuli Laine, Andreas Geiger, and Timo Aila. StyleGAN-T: Unlocking the power of GANs for fast large-scale text-to-image synthesis. 2023. 4

    2023

  20. [28]

    First order motion model for image animation

    Aliaksandr Siarohin, St ´ephane Lathuili`ere, Sergey Tulyakov, Elisa Ricci, and Nicu Sebe. First order motion model for image animation. In NeurIPS, 2019. 2, 4, 8

    2019

  21. [29]

    Motion representations for ar- ticulated animation

    Aliaksandr Siarohin, Oliver Woodford, Jian Ren, Menglei Chai, and Sergey Tulyakov. Motion representations for ar- ticulated animation. In CVPR, 2021. 2, 4

    2021

  22. [30]

    Make-a-video: Text-to-video generation without text-video data

    Uriel Singer, Adam Polyak, Thomas Hayes, Xi Yin, Jie An, Songyang Zhang, Qiyuan Hu, Harry Yang, Oron Ashual, Oran Gafni, Devi Parikh, Sonal Gupta, and Yaniv Taigman. Make-a-video: Text-to-video generation without text-video data. In ICLR, 2023. 2

    2023

  23. [31]

    Stylegan-v: A continuous video generator with the price, image quality and perks of stylegan2

    Ivan Skorokhodov, Sergey Tulyakov, and Mohamed Elho- seiny. Stylegan-v: A continuous video generator with the price, image quality and perks of stylegan2. In CVPR, 2022. 2

    2022

  24. [32]

    Denois- ing diffusion implicit models

    Jiaming Song, Chenlin Meng, and Stefano Ermon. Denois- ing diffusion implicit models. In International Conference on Learning Representations, 2021. 2

    2021

  25. [33]

    Blindly assess image qual- ity in the wild guided by a self-adaptive hyper network

    Shaolin Su, Qingsen Yan, Yu Zhu, Cheng Zhang, Xin Ge, Jinqiu Sun, and Yanning Zhang. Blindly assess image qual- ity in the wild guided by a self-adaptive hyper network. In CVPR, 2020. 9

    2020

  26. [34]

    Metaxas, and Sergey Tulyakov

    Yu Tian, Jian Ren, Menglei Chai, Kyle Olszewski, Xi Peng, Dimitris N. Metaxas, and Sergey Tulyakov. A good image generator is what you need for high-resolution video synthe- sis. In ICLR, 2021. 2

    2021

  27. [35]

    MoCoGAN: Decomposing motion and content for video generation

    Sergey Tulyakov, Ming-Yu Liu, Xiaodong Yang, and Jan Kautz. MoCoGAN: Decomposing motion and content for video generation. In CVPR, 2018

    2018

  28. [36]

    Generating videos with scene dynamics

    Carl V ondrick, Hamed Pirsiavash, and Antonio Torralba. Generating videos with scene dynamics. In NIPS, 2016. 2

    2016

  29. [37]

    Mead: A large-scale audio-visual dataset for emotional talking-face generation

    Kaisiyuan Wang, Qianyi Wu, Linsen Song, Zhuoqian Yang, Wayne Wu, Chen Qian, Ran He, Yu Qiao, and Chen Change Loy. Mead: A large-scale audio-visual dataset for emotional talking-face generation. In ECCV, 2020. 6

    2020

  30. [38]

    Video-to- video synthesis

    Ting-Chun Wang, Ming-Yu Liu, Jun-Yan Zhu, Guilin Liu, Andrew Tao, Jan Kautz, and Bryan Catanzaro. Video-to- video synthesis. In NeurIPS, 2018. 2

    2018

  31. [39]

    Few-shot video-to-video synthesis

    Ting-Chun Wang, Ming-Yu Liu, Andrew Tao, Guilin Liu, Jan Kautz, and Bryan Catanzaro. Few-shot video-to-video synthesis. In NeurIPS, 2019. 2

    2019

  32. [40]

    One-shot free-view neural talking-head synthesis for video conferenc- ing

    Ting-Chun Wang, Arun Mallya, and Ming-Yu Liu. One-shot free-view neural talking-head synthesis for video conferenc- ing. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2021. 2, 6, 8

    2021

  33. [41]

    Learning to Generate Human Videos

    Yaohui Wang. Learning to Generate Human Videos. Theses, Inria - Sophia Antipolis ; Universit´e Cote d’Azur, 2021. 2

    2021

  34. [42]

    G3AN: Disentangling appearance and motion for video generation

    Yaohui Wang, Piotr Bilinski, Francois Bremond, and Antitza Dantcheva. G3AN: Disentangling appearance and motion for video generation. In CVPR, 2020

    2020

  35. [43]

    ImaGINator: Conditional Spatio-Temporal GAN for Video Generation

    Yaohui Wang, Piotr Bilinski, Francois F Bremond, and An- titza Dantcheva. ImaGINator: Conditional Spatio-Temporal GAN for Video Generation. In WACV, 2020

    2020

  36. [44]

    Inmodegan: Interpretable motion decomposition generative adversarial network for video generation

    Yaohui Wang, Francois Bremond, and Antitza Dantcheva. Inmodegan: Interpretable motion decomposition generative adversarial network for video generation. arXiv preprint arXiv:2101.03049, 2021. 2

    2021 arXiv

  37. [45]

    Latent image animator: Learning to animate im- ages via latent space navigation

    Yaohui Wang, Di Yang, Francois Bremond, and Antitza Dantcheva. Latent image animator: Learning to animate im- ages via latent space navigation. In ICLR, 2022. 2, 3, 5 10

    2022

  38. [46]

    Lavie: High-quality video gener- ation with cascaded latent diffusion models

    Yaohui Wang, Xinyuan Chen, Xin Ma, Shangchen Zhou, Ziqi Huang, Yi Wang, Ceyuan Yang, Yinan He, Jiashuo Yu, Peiqing Yang, et al. Lavie: High-quality video gener- ation with cascaded latent diffusion models. arXiv preprint arXiv:2309.15103, 2023. 2

    2023 arXiv

  39. [47]

    Leo: Generative latent im- age animator for human video synthesis

    Yaohui Wang, Xin Ma, Xinyuan Chen, Antitza Dantcheva, Bo Dai, and Yu Qiao. Leo: Generative latent im- age animator for human video synthesis. arXiv preprint arXiv:2305.03989, 2023. 2

    2023 arXiv

  40. [48]

    Lia: Latent image animator

    Yaohui Wang, Di Yang, Francois Bremond, and Antitza Dantcheva. Lia: Latent image animator. IEEE Transactions on Pattern Analysis and Machine Intelligence , pages 1–16,

  41. [49]

    X-portrait: Expressive portrait anima- tion with hierarchical motion attention

    You Xie, Hongyi Xu, Guoxian Song, Chao Wang, Yichun Shi, and Linjie Luo. X-portrait: Expressive portrait anima- tion with hierarchical motion attention. 2024. 3, 8

    2024

  42. [50]

    Omniavatar: Geometry-guided controllable 3d head synthesis

    Hongyi Xu, Guoxian Song, Zihang Jiang, Jianfeng Zhang, Yichun Shi, Jing Liu, Wanchun Ma, Jiashi Feng, and Lin- jie Luo. Omniavatar: Geometry-guided controllable 3d head synthesis. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 1281...

    2023

  43. [51]

    Videogpt: Video generation using vq-vae and trans- formers

    Wilson Yan, Yunzhi Zhang, Pieter Abbeel, and Aravind Srinivas. Videogpt: Video generation using vq-vae and trans- formers. arXiv preprint arXiv:2104.10157, 2021. 2

    2021 arXiv

  44. [52]

    Transmomo: Invariance-driven unsupervised video motion retargeting

    Zhuoqian Yang, Wentao Zhu, Wayne Wu, Chen Qian, Qiang Zhou, Bolei Zhou, and Chen Change Loy. Transmomo: Invariance-driven unsupervised video motion retargeting. In CVPR, 2020. 2

    2020

  45. [53]

    Generating videos with dynamics-aware implicit generative adversarial net- works

    Sihyun Yu, Jihoon Tack, Sangwoo Mo, Hyunsu Kim, Junho Kim, Jung-Woo Ha, and Jinwoo Shin. Generating videos with dynamics-aware implicit generative adversarial net- works. In ICLR, 2022. 2

    2022

  46. [54]

    Toontalker: Cross-domain face reenactment, 2023

    Gong Yuan, Zhang Yong, Cun Xiaodong, Yin Fei, Fan Yanbo, Wang Xuan, Wu Baoyuan, and Yang Yujiu. Toontalker: Cross-domain face reenactment, 2023. 2

    2023

  47. [55]

    Few-shot adversarial learning of realis- tic neural talking head models

    Egor Zakharov, Aliaksandra Shysheya, Egor Burkov, and Victor Lempitsky. Few-shot adversarial learning of realis- tic neural talking head models. In ICCV, 2019. 2

    2019

  48. [56]

    Show-1: Marrying pixel and latent diffusion models for text-to-video generation

    David Junhao Zhang, Jay Zhangjie Wu, Jia-Wei Liu, Rui Zhao, Lingmin Ran, Yuchao Gu, Difei Gao, and Mike Zheng Shou. Show-1: Marrying pixel and latent diffusion models for text-to-video generation. International Journal of Com- puter Vision, pages 1–15, 2024. 2

    2024

  49. [57]

    Flow-guided one-shot talking face generation with a high- resolution audio-visual dataset

    Zhimeng Zhang, Lincheng Li, Yu Ding, and Changjie Fan. Flow-guided one-shot talking face generation with a high- resolution audio-visual dataset. In CVPR, 2021. 6, 9

    2021

  50. [58]

    Thin-plate spline motion model for image animation

    Jian Zhao and Hui Zhang. Thin-plate spline motion model for image animation. In CVPR, 2022. 2, 4, 8 11

    2022

Pith tools

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