Pith. sign in

REVIEW 4 major objections 5 minor 36 references

Controllable Single-shot Animation Blending with Temporal Conditioning

T0 review · 4 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read The paper claims that a time-varying skeleton identity map lets a single-shot generative model blend two or more input animations smoothly in one forward pass, with direct control over which motion appears at each time segment.

desk verdict A plausible single-shot motion blending idea with sparse evidence; the central conditioning extrapolation is untested but the work is worth a serious look. read the letter →

arxiv 2508.18525 v1 pith:QW6HCBLN submitted 2025-08-25 cs.GR cs.CV

Eleni Tselepi , Spyridon Thermos , Gerasimos Potamianos This is my paper · ORCID
classification cs.GRcs.CV
keywords single-shotmotiongenerationblendingtemporalconditioningskeletonidentitymapSPADEskeleton-awarenormalizationGANimatorskeletalanimation
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

The paper claims to be the first single-shot motion blending framework: a model trained on two or more skeletal motion sequences can, at inference, generate a smooth transition between them in a single forward pass, driven only by a temporal label map that says which motion should be active at each frame. If correct, this gives animators a lightweight, interpretable control signal for blending motions without retraining, large motion datasets, or post-processing. The key move is to condition the generator with a skeleton identity map through a SPADE-like, skeleton-aware normalization layer, and to train the model with each input motion in a separate batch. The authors demonstrate the approach on Mixamo and 100STYLE motions, including three-way blends, and show that the resulting transitions are kinematically smooth.

What carries the argument

The central mechanism is the temporally-conditioned skeleton identity map feeding a SPADE-like block: a map with one channel per input motion, indicating which motion should be active at each frame, is passed through two skeleton-aware convolutional layers to produce per-location scale (gamma) and shift (beta) tensors that modulate the generator's motion features. Because the conditioning is applied at the first, coarsest level of the temporal hierarchy, the model learns the global structure of each motion before refining, which the authors argue is why the blend is coherent. The batched training scheme, where each input motion occupies a separate batch, lets the generator assign distinct semantic labels to different motions, something the original single-shot GANimator could not do.

What would settle it

Train the model on two motions with very different dynamics, such as breakdance freezes and salsa dancing, and run inference with a skeleton identity map that switches abruptly at the midpoint; if the generated transition shows a visible discontinuity in joint velocity or acceleration, or if the output simply switches instantly between the two motion styles instead of blending, the central claim fails. A second check: invert the identity map (assign ones to the first motion and zeros to the second) and compare the blended output; if the result is not temporally symmetric in a way consistent with the swap, the label map is encoding something other than the intended motion identity.

Watch

Extended reading notes

Core claim

The paper claims that a single-shot GAN, built on a batched version of GANimator and extended with a SPADE-inspired skeleton-aware conditioning mechanism, can learn to blend two or more input motions so that the user can choose, at inference, which motion drives each time segment. During training each motion gets a constant skeleton identity channel (zeros for one motion, ones for another); at inference a combined map assigns the first half of frames to one identity and the second half to the other, and the generator produces a smooth transition. The conditioning operates through modulation tensors gamma and beta computed by skeleton-aware convolutions, which scale and shift the motion features without applying normalization, preserving the structure needed for smooth transitions. The authors show quantitative gains over an unconditioned GANimator baseline and a FiLM alternative, and qualitative results for two- and three-motion blends.

Load-bearing premise

The main load-bearing premise is that a simple time-varying skeleton identity map, zeros for one motion and ones for the other, carries enough information for the network to produce a smooth, plausible transition between motions without transition-specific training data or post-processing.

Editorial extensions

If this is right

  • An animator can select which input motion drives each time segment at inference by editing the skeleton identity map, with no additional training.
  • The method extends beyond two inputs: a three-way blend is demonstrated, so the control signal scales to more motions in a single forward pass.
  • Because blending happens in one generative pass, the approach can run in a few seconds on modest hardware, making it suitable for interactive or resource-constrained animation workflows.
  • The framework inherits single-shot data efficiency: it learns from the input motion sequences alone, without large motion capture datasets.
  • Blending quality varies with motion similarity, as shown by FID scores, meaning animators should expect better results when the source motions share dynamics.

Reading between the lines

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

  • The skeleton identity map could be generalized to continuous values or soft scheduling curves, letting an animator control not only when but how gradually the blend occurs, which the paper does not explore.
  • A natural testable extension is to condition the same architecture with textual or semantic prompts instead of a binary map, a direction the paper explicitly leaves to future work.
  • If the identity map is sufficient, the approach could serve as a data augmentation or motion-retargeting tool by blending a source motion into a target skeleton's kinematic style, though the paper does not investigate this.
  • The similarity dependence in the FID results suggests a potential practical boundary: for very dissimilar motions, an expert animator may still need to design the transition schedule manually.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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 manuscript proposes a single-shot generative framework for blending two or more skeletal motion sequences. It extends a batched version of GANimator with SPADE-like skeleton-aware modulation layers, where a temporal skeleton identity map is used as conditioning. At training time each input motion is assigned a constant identity channel; at inference the map is modified (e.g., first half zeros, second half ones) to drive a transition. The authors report ablations on Mixamo and 100STYLE, with FID, coverage, diversity, and joint velocity/acceleration metrics, and claim the first controllable single-shot motion blending method with smooth, plausible, and controllable transitions.

Significance. The idea is timely and practically relevant: animators could blend motions in a single generative pass without retraining or large datasets. The skeleton-aware SPADE design and the honest ablation of SPADE placement and FiLM are useful. The paper also explicitly acknowledges that FID is similarity-biased and that expert evaluation is necessary. However, the central claim rests on an untested extrapolation from constant training labels to time-varying inference labels, and the quantitative evidence is partly unfavorable (Table 2) or lacks baselines (Fig. 4). With additional validation the approach could be a meaningful contribution.

major comments (4)
  1. [Sec. 3.2.2 and Fig. 3] The conditioning signal used at inference is a step or piecewise-constant skeleton identity map, whereas during training every batch is conditioned on a constant map (all zeros or all ones). The losses in Eqs. (1)-(4) therefore provide no explicit supervision for the transition region, and the SPADE layers compute gamma and beta from inputs they have never observed near the boundary. The paper would need to demonstrate that the network actually generalizes across this distribution shift; the current evidence is a few qualitative examples. I ask for experiments that vary the transition time and transition width, together with a comparison against a simple crossfade or motion-in-betweening baseline on the same data.
  2. [Table 2] On 100STYLE, the proposed method has FID 0.13, which is worse than the GANimator baseline's 0.09, and the coverage jumps to 1.00 while GDiv, LDiv, and Inter-Div collapse to roughly 0.51, 0.46, and 0.35. This pattern suggests a loss of diversity rather than a successful blend. The same caveat applies to the FiLM row. The authors should report per-motion evaluation, additional samples, and explain why coverage becomes 1.00 while diversity drops; as presented, the quantitative results do not support the claim of high-fidelity blending.
  3. [Sec. 4.1 and Fig. 4] The L2 velocity and acceleration plots are the only direct evidence for 'smooth' transitions, but they are shown for a single example with no baseline (e.g., crossfade, GANimator, or ground-truth concatenation), no error bars, and no statistical aggregation over multiple seeds or motion pairs. The curves cannot support the smoothness claim without such comparisons.
  4. [Sec. 4.2.2 and Fig. 6] The caption states that 'the qualitative evaluation of generative animation models from experts ... is a necessity,' but no expert or user evaluation is included. Since the paper's central perceptual claims are about plausibility and smoothness, the absence of any human evaluation is a significant gap. In addition, the controllability claim is never tested: all reported blends use fixed 50/50 or equal-third splits, with no variation of transition location or transition duration, so 'when and how motions blend' is not actually demonstrated.
minor comments (5)
  1. [Sec. 3.2 and Table 1] The phrase 'first level (i.e., the ones of the first two G,D pairs)' conflicts with the ablation rows labeled '2nd Level', '3rd Level', '4th Level' in Table 1 and with the text in Sec. 4.2.1 that mentions placing the SPADE layer at the first, second, third, and fourth levels; please clarify the level indexing.
  2. [Eq. (5)] The notation Delta v_{t,j} is defined as |v_{t,j} - v_{t-1,j}|, but the symbol v_{t,j} is used both for the velocity vector and its L2 norm; please use distinct notation for these two quantities.
  3. [References] Reference [8] lists a placeholder arXiv identifier (2301.12345) and should be updated with the correct venue and identifier; reference [13] also appears to have an incorrect title for the cited Isola et al. paper.
  4. [Sec. 1] The text says the project page 'can be found here' but no URL is included; please add the link or remove the sentence.
  5. [Fig. 6] Figure 6 would benefit from error bars and a description of how the similarity-based FID comparison is computed; the single bars and unlabeled axes are hard to interpret.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity; the only self-citation ([30]) is an implementation dependency, and the paper itself concedes its FID evidence is similarity-biased.

full rationale

The paper's central claim is that a temporally-varying skeleton identity map, processed by SPADE-like skeleton-aware convolutions inside a batched GANimator, produces smooth and controllable blends in one generative pass. Tracing the derivation: the generator is trained with constant per-motion labels (zeros and ones) under the adversarial, reconstruction, and foot-contact losses (Eqs. 1-4); at inference, only the user-supplied label map changes. No equation in Sec. 3 reduces a reported result to an input by construction, and the output is not a fitted quantity: the transition itself has no ground-truth supervision, so the claimed smoothness is an empirical extrapolation rather than a definitional equivalence. The reported FID, coverage, and diversity numbers are comparisons against GANimator and FiLM baselines, not predictions forced by the conditioning signal. The only self-citation is [30], the batched GANimator variant co-authored by one of the present authors; it is used as an implementation dependency for multi-batch training, not as a theorem, uniqueness result, or proof of the blending claim, so it is not load-bearing in a circularity sense. The paper explicitly flags an evaluation limitation in the Fig. 6 caption, conceding that FID is biased toward similarity and that expert evaluation is a necessity; that is a validity concern, not a circularity concern. No step equates an output with an input by definition, and no ansatz or uniqueness is smuggled in via self-citation. The mild score of 2 reflects only the presence of the minor, non-load-bearing self-citation [30].

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The work is an empirical architecture paper, not a derivation, so its 'axioms' are design assumptions about what a conditioning label map can do and about the stability of the underlying batched GANimator. The two most consequential assumptions are that per-motion labels learned in separate batches separate the internal distributions, and that a time-varying label map at inference produces a smooth transition. The loss weights are the main hand-chosen parameters; their values are not reported.

free parameters (3)
  • lambda_adv = not reported
    Adversarial loss weight in Eq. 4; chosen by hand, no sensitivity analysis.
  • lambda_rec = not reported
    Reconstruction loss weight in Eq. 4; no sensitivity analysis.
  • lambda_con = not reported
    Foot contact consistency loss weight in Eq. 4; no sensitivity analysis.
assumptions (5)
  • domain assumption The batched GANimator variant of Roditakis et al. [30] provides a stable generative backbone that supports simultaneous training on multiple motions.
    The paper does not detail this variant; it is cited as the foundation for multi-batch training (Sec. 3.1).
  • domain assumption Assigning a unique skeleton identity label to each motion during training lets the network learn separable internal distributions.
    Used in Sec. 3.2.2; if false, the conditioning signal cannot separate motions.
  • ad hoc to paper A time-varying skeleton identity map at inference produces a smooth, plausible transition between motions without additional transition data.
    This is the core control mechanism (Sec. 3.2.2, Fig. 3); the paper does not prove it, only demonstrates it on examples.
  • domain assumption Omitting normalization in the SPADE block preserves motion structure needed for blending.
    The paper argues that removing normalization avoids texture-like high-frequency signals; no independent evidence is provided.
  • domain assumption FID, coverage, and diversity metrics computed on the training motions are meaningful for single-shot blending quality.
    These metrics are inherited from prior single-shot literature [19, 29, 31] and applied without adaptation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Controllable Single-shot Animation Blending with Temporal Conditioning." pith.science (2026). https://pith.science/paper/QW6HCBLN

@misc{pith2026250818525,
  author       = {Pith},
  title        = {Pith review of: Controllable Single-shot Animation Blending with Temporal Conditioning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QW6HCBLN}},
  note         = {Machine review of arXiv:2508.18525}
}
read the original abstract

Training a generative model on a single human skeletal motion sequence without being bound to a specific kinematic tree has drawn significant attention from the animation community. Unlike text-to-motion generation, single-shot models allow animators to controllably generate variations of existing motion patterns without requiring additional data or extensive retraining. However, existing single-shot methods do not explicitly offer a controllable framework for blending two or more motions within a single generative pass. In this paper, we present the first single-shot motion blending framework that enables seamless blending by temporally conditioning the generation process. Our method introduces a skeleton-aware normalization mechanism to guide the transition between motions, allowing smooth, data-driven control over when and how motions blend. We perform extensive quantitative and qualitative evaluations across various animation styles and different kinematic skeletons, demonstrating that our approach produces plausible, smooth, and controllable motion blends in a unified and efficient manner.

Figures

Figures reproduced from arXiv: 2508.18525 by the authors.

Figure 1
Figure 1. We introduce a single-shot motion blending approach that uses a batched version of the GANimator [ [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. A SPADE-like block that receives the temporal condi [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. A schematic representation of the training (top) and inference (bottom) processes. During training each semantic label is [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: L2 velocity (top) and acceleration (bottom) per joint for [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: The result of blending two (top) and three (bottom) animations from the Mixamo set [ [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Blending quality as a function of motion similarity. We [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

36 extracted references · 35 canonical work pages

  1. [1]

    Skeleton- aware networks for deep motion retargeting

    Kfir Aberman, Peizhuo Li, Dani Lischinski, Olga Sorkine- Hornung, Daniel Cohen-Or, and Baoquan Chen. Skeleton- aware networks for deep motion retargeting. ACM Transac- tions on Graphics (TOG), 39(4), 2020. 1, 2, 4

    work page 2020

  2. [2]

    Adobe Systems Inc. 2021. Mixamo. https://www. mixamo.com/. 5, 6, 7, 8

    work page 2021

  3. [3]

    Okan Arikan and David A. Forsyth. Interactive motion gen- eration from examples. ACM Transactions on Graphics (TOG), 21(3):483–490, 2002. 1, 2

    work page 2002

  4. [4]

    Forsyth, and James F

    Okan Arikan, David A. Forsyth, and James F. O’Brien. Mo- tion synthesis from annotations. In Proc. ACM SIGGRAPH, pages 402–408, 2003. 1

    work page 2003

  5. [5]

    Black, and G¨ul Varol

    Nikos Athanasiou, Mathis Petrovich, Michael J. Black, and G¨ul Varol. TEACH: Temporal action composition for 3D humans. In Proc. International Conference on 3D Vision (3DV), pages 414–423, 2022. 2

    work page 2022

  6. [6]

    Badler, Cary B

    Norman I. Badler, Cary B. Phillips, and Bonnie Lynn Web- ber. Simulating Humans: Computer Graphics, Animation, and Control. Oxford University Press, Inc., 1993. 2

    work page 1993

  7. [7]

    Seamless human motion composition with blended posi- tional encodings

    German Barquero, Sergio Escalera, and Cristina Palmero. Seamless human motion composition with blended posi- tional encodings. In Proc. IEEE/CVF Conference on Com- puter Vision and Pattern Recognition (CVPR) , pages 457– 469, 2024. 2

    work page 2024

  8. [8]

    Single- shot motion completion with transformer

    Yinglin Duan, Tianyang Shi, Zhengxia Zou, Yenan Lin, Zhehui Qian, Bohan Zhang, and Yi Yuan. Single- shot motion completion with transformer. arXiv preprint arXiv:2301.12345, 2021. 3

    arXiv 2021

Show all 36 references
  1. [9]

    Synthesis of compositional animations from textual descriptions

    Anindita Ghosh, Noshaba Cheema, Cennet Oguz, Christian Theobalt, and Philipp Slusallek. Synthesis of compositional animations from textual descriptions. In Proc. IEEE/CVF International Conference on Computer Vision (ICCV), pages 1396–1406, 2021. 2

    2021

  2. [10]

    Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio

    Ian J. Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. Generative adversarial networks. In Ad- vances in Neural Information Processing Systems, 2014. 2

    2014

  3. [11]

    Courville

    Ishaan Gulrajani, Faruk Ahmed, Martin Arjovsky, Vincent Dumoulin, and Aaron C. Courville. Improved training of Wasserstein GANs. In Advances in Neural Information Pro- cessing Systems, pages 5769–5779, 2017. 4

    2017

  4. [12]

    Harvey, Mike Yurick, Derek Nowrouzezahrai, and Christopher Pal

    F ´elix G. Harvey, Mike Yurick, Derek Nowrouzezahrai, and Christopher Pal. Robust motion in-betweening. ACM Trans- actions on Graphics (TOG), 39(4), 2020. 2

    2020

  5. [13]

    Phillip Isola, JunYan Zhu, Tinghui Zhou, and Alexei A. Efros. Learned queries for efficient local attention. In Proc. IEEE/CVF Conference on Computer Vision and Pat- tern Recognition (CVPR), pages 1125–1134, 2017. 3

    2017

  6. [14]

    Guided motion diffusion for controllable human motion synthesis

    Korrawe Karunratanakul, Konpat Preechakul, Supasorn Suwajanakorn, and Siyu Tang. Guided motion diffusion for controllable human motion synthesis. In Proc. IEEE/CVF International Conference on Computer Vision (ICCV), pages 2151–2162, 2023. 2

    2023

  7. [15]

    BlendGAN: Learning and blending the in- ternal distributions of single images by spatial image-identity conditioning

    Idan Kligvasser, Tamar Rott Shaham, Noa Alkobi, and Tomer Michaeli. BlendGAN: Learning and blending the in- ternal distributions of single images by spatial image-identity conditioning. arXiv preprint arXiv:2212.01589, 2022. 3

    2022 arXiv

  8. [16]

    Automated extraction and parameterization of motions in large data sets

    Lucas Kovar and Michael Gleicher. Automated extraction and parameterization of motions in large data sets. ACM Transactions on Graphics (TOG), 23(3):559–568, 2004. 2

    2004

  9. [17]

    Mo- tion graphs

    Lucas Kovar, Michael Gleicher, and Fr ´ed´eric Pighin. Mo- tion graphs. ACM Transactions on Graphics (TOG), 21(3): 473–482, 2002. 1, 2

    2002

  10. [18]

    Precomputed real-time texture synthesis with Markovian generative adversarial networks

    Chuan Li and Michael Wand. Precomputed real-time texture synthesis with Markovian generative adversarial networks. In Proc. European Conference on Computer Vision (ECCV), pages 702–716, 2016. 3

    2016

  11. [19]

    GANimator: Neural motion synthe- sis from a single sequence

    Peizhuo Li, Kfir Aberman, Zihan Zhang, Rana Hanocka, and Olga Sorkine-Hornung. GANimator: Neural motion synthe- sis from a single sequence. ACM Transactions on Graphics (TOG), 41(4), 2022. 1, 3, 5, 6, 8

    2022

  12. [20]

    Character controllers using motion V AEs

    Hung Yu Ling, Fabio Zinno, George Cheng, and Michiel van de Panne. Character controllers using motion V AEs. ACM Transactions on Graphics (TOG), 39(4), 2020. 2

    2020

  13. [21]

    Ian Mason, Sebastian Starke, and Taku Komura. Real-time style modelling of human locomotion via feature-wise trans- formations and local motion phases.Proceedings of the ACM on Computer Graphics and Interactive Techniques (CGIT), 5 (1), 2022. 5, 6

    2022

  14. [22]

    Geostatistical mo- tion interpolation

    Tomohiko Mukai and Shigeru Kuriyama. Geostatistical mo- tion interpolation. ACM Transactions on Graphics (TOG) , 24(3):1062–1070, 2005. 2

    2005

  15. [23]

    Oreshkin, Antonios Valkanas, Felix G

    Boris N. Oreshkin, Antonios Valkanas, Felix G. Har- vey, Louis-Simon Menard, Florent Bocquelet, and Mark J. Coates. Motion in-betweening via deep ∆-interpolator. IEEE Transactions on Visualization and Computer Graphics (TVCG), 30(8):5693–5704, 2024. 2

    2024

  16. [24]

    Semantic image synthesis with spatially-adaptive nor- malization

    Taesung Park, Ming-Yu Liu, Ting-Chun Wang, and Jun-Yan Zhu. Semantic image synthesis with spatially-adaptive nor- malization. In Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pages 2332–2341,

  17. [25]

    PyTorch: An imper- ative style, high-performance deep learning library

    Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, Alban Desmaison, Andreas K ¨opf, Edward Yang, Zach DeVito, Martin Raison, Alykhan Tejani, Sasank Chilamkurthy, Benoit Steiner, Lu F...

    2019

  18. [26]

    FiLM: Visual reasoning with a general conditioning layer

    Ethan Perez, Florian Strub, Harm de Vries, Vincent Du- moulin, and Aaron Courville. FiLM: Visual reasoning with a general conditioning layer. In Proc. AAAI Conference on Artificial Intelligence, pages 3942–3951, 2018. 6

    2018

  19. [27]

    Black, and G ¨ul Varol

    Mathis Petrovich, Michael J. Black, and G ¨ul Varol. TEMOS: Generating diverse human motions from textual descriptions. In Proc. European Conference on Computer Vision (ECCV), pages 480–497, 2022. 2

    2022

  20. [28]

    Motion in-betweening via two-stage transformers

    Jia Qin, Youyi Zheng, and Kun Zhou. Motion in-betweening via two-stage transformers. ACM Transactions on Graphics (TOG), 41(6), 2022. 2

    2022

  21. [29]

    Bermano, and Daniel Cohen-Or

    Sigal Raab, Inbal Leibovitch, Guy Tevet, Moab Arar, Amit H. Bermano, and Daniel Cohen-Or. Single motion dif- fusion. In Proc. International Conference on Learning Rep- resentations (ICLR), 2024. 3, 6

    2024

  22. [30]

    Towards practical single-shot motion synthesis

    Konstantinos Roditakis, Spyridon Thermos, and Nikolaos Zioulis. Towards practical single-shot motion synthesis. In Proc. Workshop on AI for 3D Generation (AI3DG @ CVPR),

  23. [31]

    Sin- GAN: Learning a generative model from a single natural im- age

    Tamar Rott Shaham, Tali Dekel, and Tomer Michaeli. Sin- GAN: Learning a generative model from a single natural im- age. In Proc. IEEE/CVF International Conference on Com- puter Vision (ICCV), 2019. 3, 6

    2019

  24. [32]

    Human motion diffusion as a generative prior

    Yoni Shafir, Guy Tevet, Roy Kapon, and Amit Haim Bermano. Human motion diffusion as a generative prior. In Proc. International Conference on Learning Representations (ICLR), 2024. 2

    2024

  25. [33]

    InGAN: Capturing and retargeting the “DNA” of a natu- ral image

    Assaf Shocher, Shai Bagon, Phillip Isola, and Michal Irani. InGAN: Capturing and retargeting the “DNA” of a natu- ral image. In Proc. IEEE/CVF International Conference on Computer Vision (ICCV), pages 4491–4500, 2019. 3

    2019

  26. [34]

    SinGRAF: Learning a 3D generative radi- ance field for a single scene

    Minjung Son, Jeong Joon Park, Leonidas Guibas, and Gor- don Wetzstein. SinGRAF: Learning a 3D generative radi- ance field for a single scene. In Proc. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 8507–8517, 2023. 3

    2023

  27. [35]

    Neural state machine for character-scene interactions

    Sebastian Starke, He Zhang, Taku Komura, and Jun Saito. Neural state machine for character-scene interactions. ACM Transactions on Graphics (TOG), 38(6), 2019. 2

    2019

  28. [36]

    Sin3DM: Learning a diffusion model from a single 3D tex- tured shape

    Rundi Wu, Ruoshi Liu, Carl V ondrick, and Changxi Zheng. Sin3DM: Learning a diffusion model from a single 3D tex- tured shape. In Proc. International Conference on Learning Representations (ICLR), 2023. 3

    2023

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.