Pith. sign in

REVIEW 5 major objections 6 minor 53 references

A Plug-and-Play Multi-Criteria Guidance for Diverse In-Betweening Human Motion Generation

T0 review · 5 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read Treating a frozen motion generator's sampling as a two-criteria Pareto search can raise the diversity of the motions it produces, with no retraining or added parameters, the paper argues.

desk verdict A genuinely new sampling-time evolutionary recipe for in-betweening motion, but the key conditioning step is unspecified and the evaluation is partly circular, so the claims outrun the evidence. read the letter →

arxiv 2508.01590 v1 pith:GXWDRJF2 submitted 2025-08-03 cs.GR cs.CV

Hua Yu , Jiao Liu , Xu Gui , Melvin Wong , Yaqing Hou , Yew-Soon Ong This is my paper · ORCID
classification cs.GRcs.CV
keywords humanmotionin-betweeningdiversitymulti-criteriaoptimizationParetooptimalityplug-and-playguidancediffusionmodelsevolutionarysamplinggeneration
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper tries to establish that the diversity of in-betweening human motions produced by pretrained generative models can be improved at sampling time, without retraining or adding parameters. The authors reformulate the sampling process as a multi-criteria optimization problem with two objectives—one that encourages variety across and within action categories and one that enforces smooth joins to the given keyframes—and then search for Pareto-optimal motion sequences using an evolutionary loop. If this works, it matters because intra-batch diversity is a known weakness of conditional motion generators, and the guidance is designed to be model-agnostic across DDPM, VAE, and GAN backbones. The paper reports consistent improvements in diversity and accuracy on four human motion datasets.

What carries the argument

The central object is the multi-criteria optimization formulation with objectives $F_1(Y)=\alpha_1(Y)+\beta(Y)$ and $F_2(Y)=\alpha_2(Y)+\beta(Y)$, where $\alpha_1(Y)=\frac{1}{D}(C(Y)+P_c(Y))$, $\alpha_2(Y)=1-\alpha_1(Y)$, and $\beta(Y)=\|X_1[-1]-Y[0]\|+\|Y[-1]-X_2[0]\|$ measures endpoint smoothness. The diversity component $\alpha_1$ couples a classifier's category label $C(Y)$ with its class probabilities $P_c(Y)$, so that both inter-class and intra-class differences are rewarded. The theorems guarantee that minimizers of $\beta$ are Pareto-optimal and that separating solutions along the Pareto front in objective space separates them by action category, which is what turns an optimization loop into a diversity engine.

What would settle it

Apply MCG-IMM to a pretrained VAE or GAN backbone whose forward pass accepts only keyframe poses and has no conditioning pathway for a full previously generated sequence; if that backbone cannot produce the offspring described in the generation equation, the method's model-agnostic plug-and-play claim fails in practice.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central claim is that any pretrained in-betweening motion generator can be steered toward more diverse and still smooth outputs by turning its sampling into a two-criteria search. The first criterion uses a frozen action classifier's labels and class probabilities to separate both different action categories and subtle variations within a category; the second measures the discontinuity at the two endpoints where the generated transition meets the user-provided sequences. Solving this two-objective problem with an evolutionary process—generating offspring motions from elite sequences via the frozen generator, then selecting by nondominated sorting and crowding distance—yields a batch of transitions that are mutually diverse and temporally coherent, with no additional training parameters. Two supporting theorems state that any motion minimizing the smoothness term is Pareto-optimal, and that two Pareto-optimal motions whose objective-space distance exceeds $4/D$ must belong to different action classes. The empirical claim is that this guidance consistently improves both diversity and accuracy over prior methods on BABEL, HumanAct12, NTU RGB-D, and GRAB, using DDPM, VAE, or GAN as the frozen backbone.

Load-bearing premise

The load-bearing premise is that the pretrained generative model can be conditioned on a full previously generated motion sequence in addition to the two keyframe sequences, and that the frozen classifier's category probabilities provide a meaningful signal for within-category diversity; the paper does not spell out how VAE or GAN backbones would accept that sequence conditioning.

Editorial extensions

If this is right

  • Wrapping an existing pretrained in-betweening model with MCG-IMM should improve intra-batch diversity immediately, without the cost of fine-tuning or architectural changes.
  • Diversity and smoothness become inference-time controls: adjusting the criteria or the number of optimization iterations tunes the output batch without touching the trained weights.
  • The same Pareto-search wrapper could be dropped onto any conditional generative sequence model that accepts an additional conditioning sequence alongside the fixed keyframes, making the approach a general sampling-time plugin.
  • The diversity ceiling is set by the classifier's granularity: a more fine-grained action classifier would let the optimizer separate motions that share a coarse action label.
  • Variable-length transitions are handled by predicting the transition length from the cosine similarity of the endpoint poses, avoiding fixed-length assumptions in the backbone.

Reading between the lines

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

  • The plug-and-play claim quietly depends on the backbone accepting a full previously generated sequence as a conditioning input; generative models that only condition on keyframes or latent codes would need a wrapper to satisfy the offspring-generation step, which the paper does not specify.
  • Because the diversity term is driven by a classifier's category probabilities, the method's diversity ceiling is set by the classifier's ability to perceive differences; on noisy datasets such as NTU RGB-D, the guidance may struggle, which matches the weaker reported gains there.
  • The evolutionary loop is essentially using the frozen generator as a mutation operator, an idea that could transfer to other autoregressive or diffusion-based sequence tasks where a pretrained conditional model plus a classifier are available.
  • The two theorems rely on $\alpha_1+\alpha_2=1$ exactly, so any normalization change in the classifier probabilities would break the Pareto-optimality proof and the construction would need reworking.
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

5 major / 6 minor

Summary. The paper proposes MCG-IMM, a plug-and-play multi-criteria guidance method for diverse in-betweening human motion generation. It reformulates the sampling process of pretrained generative models (DDPM, VAE, GAN) as a multi-criteria optimization problem with objectives for diversity and smoothness, and solves it using an evolutionary algorithm with nondominated sorting and crowding distance. Experiments on BABEL, HumanAct12, NTU RGB-D, and GRAB are reported, and the paper claims consistent state-of-the-art performance without retraining or additional parameters.

Significance. If the central claim were established, the method would be a valuable lightweight add-on to existing generative models for motion in-betweening, addressing a real need for intra-batch diversity. The paper's ambition to be model-agnostic is attractive, and the inclusion of ablations and the attempt at theoretical support are commendable. However, the manuscript as written does not provide the necessary specification of the conditioning mechanism, and the evaluation is confounded by using the same classifier in both the objective and the metrics.

major comments (5)
  1. [III.B, Eq. (8), Algorithm 1 line 4] The offspring generation step conditions the generative model on a full previously generated motion sequence Y_{i-1}^{[l']}. For DDPM this could be realized by inpainting or classifier guidance, but no such mechanism is specified; for VAE and GAN, a standard pretrained generator does not accept a full motion sequence as a conditioning input. The paper refers to "the Appendix" for details, but no appendix is included in v1. If the backbones are not conditioned on Y_{i-1}, the offspring distribution is independent of the elite parent and the evolutionary selection degenerates to repeated i.i.d. sampling; if they are, the backbone must have been trained or modified to accept this input, contradicting the "no retraining / no additional parameters" claim. Either way, the reported APD improvements are not attributable to the proposed multi-criteria guidance.
  2. [IV.C and Eq. (1)] The evaluation metrics ACC and FID are computed using the same pretrained action classifier that is used in the diversity objective via C(Y) and P_c(Y). This is circular: the optimization directly moves generated motions toward the classifier's preferred categories and feature distribution, so the reported ACC and FID gains are at least partly an artifact of optimizing the evaluation signal. An independent evaluation classifier or metrics not derived from the same model are needed.
  3. [III.A, Eq. (1)] The diversity component is ill-typed: C(Y) is an integer in {0,...,D-1}, while P_c(Y) is described as "the probability that a given motion sequence belongs to each categorized motion type," which would be a D-dimensional vector; adding them is undefined. If P_c(Y) is instead the scalar probability of the predicted class, the notation should say so explicitly. Moreover, both α1 and α2 are per-sample functions; minimizing F1 and F2 does not directly measure or encourage pairwise diversity within the batch, so the connection between the multi-criteria problem and the APD metric is not established.
  4. [III.A, Theorems 1 and 2] These results are near-tautological consequences of the definitions (α1+α2=1 and the form of β) and do not provide independent support for the diversity claim. Theorem 1 only asserts that minimizers of β are Pareto optimal, and Theorem 2 only states that two solutions in B with sufficiently distant objective vectors have different class labels. Neither theorem guarantees that the evolutionary algorithm will find such solutions or that the resulting batch will be diverse in the sense measured by APD.
  5. [Table I, Section V.A] The empirical evidence is incomplete and overstates the case for consistency. No error bars or multiple seeds are reported. On NTU RGB-D, MCG-IMM (VAE) has FIDtr 81.64 versus 75.46 for MoFusion; on GRAB, MCG-IMM (VAE) has ACC 78.5 versus 79.21 for MoFusion. Also, the underlying pretrained backbones for the three MCG-IMM variants are not identified, so the comparison across methods is not controlled; differences in base models may account for the reported improvements.
minor comments (6)
  1. [Section IV.B vs. Eq. (2)] The symbol β is used both for the diffusion noise schedule in Section IV.B and for the smoothness function in Eq. (2), which is confusing and should be disambiguated.
  2. [Section V.A] The phrase "within a bath process" appears to be a typo for "within a batch process."
  3. [Abstract] The final sentence of the abstract is grammatically incomplete; it should read something like "consistently outperforms state-of-the-art methods in the in-betweening motion generation task."
  4. [Eq. (11)] Equation (11) uses Y for the ground truth and Ŷ for the prediction, while elsewhere Y denotes generated sequences; the notation should be unified.
  5. [Remark 1, Section III.B] The text repeatedly refers to an appendix for details, but the v1 manuscript contains no appendix; this should be fixed before resubmission.
  6. [Abstract and Section I] The claim "without introducing additional parameters" should be qualified to "no additional trainable parameters," since the classifier C(Y) has its own parameters, even if they are frozen.

Circularity Check

1 steps flagged · score 6.0 of 10

Same pretrained classifier defines the diversity objective and the ACC/FID evaluation metrics; accuracy gains are partly by construction, while APD diversity is independently measured.

  1. fitted input called prediction [Section III.A (Eq. 1), Algorithm 1, Section IV.C (FID/ACC)]
    "We assume the availability of a classifier C(Y) that can categorize the motion type represented by a generated motion sequence Y. ... To evaluate motion realism, we report the action recognition accuracy of the generated motions using the same pretrained action recognition model."

    The diversity objective (Eq. 1) is built entirely from the outputs of classifier C(Y): α1(Y)=(C(Y)+P_c(Y))/D and α2(Y)=1−α1(Y). Algorithm 1 selects offspring using nondominated sorting on F1,F2, i.e., it explicitly spreads the classifier's predicted labels and probabilities across the batch. The evaluation then measures ACC with "the same pretrained action recognition model" and FID with features from "a pretrained motion classifier model." Since the same fitted classifier defines the search objective and the reported metrics, the ACC/FID improvements over baselines are partly by construction; these metrics are not independent tests of generated motion quality. The diversity metric APD (Eq.

full rationale

The most defensible circular step is the shared classifier. Section III.A defines the diversity component through a pretrained action classifier C(Y) (Eq. 1), and Algorithm 1's selection criterion (F1,F2) is a function of C(Y) and P_c(Y). Section IV.C then evaluates the same method with ACC computed by "the same pretrained action recognition model" and FID from features of "a pretrained motion classifier model." Because the optimizer is explicitly spreading classifier outputs, the reported ACC (and, to a lesser extent, FID) is not an independent measure of quality; this is a partial, evaluation-level circularity. It does not erase the main contribution: APD (Eq. 12) is a label-free pairwise L2 diversity metric, and the plug-and-play/no-retraining property is structurally realized by the sampling-loop design. The two theorems are deductive consequences of the definitions (α1+α2=1) and therefore are not circular, but they add little independent empirical support. A separate concern is the omitted appendix: Eq. (8) conditions offspring on Y_{i-1}, and Section III.B defers the per-backbone construction to "the Appendix"; no appendix is present in v1, so the plug-and-play claim for VAE/GAN rests on an unstated conditioning mechanism. This is missing support, not circularity. Author self-citations ([33], [38]-[41]) appear as related-work and differentiation only, and are not load-bearing for the central claim. Overall score 6: one evaluation metric reduces by construction, while the core diversity claim has independent content.

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

The method relies on a trained classifier and hand-set parameters rather than a first-principles derivation. The classifier is a significant additional component that conflicts with the 'no additional parameters' claim.

free parameters (5)
  • Y_min = 5
    Hand-set minimum transition length in Eq. (9).
  • Y_max = 15
    Hand-set maximum transition length in Eq. (9).
  • I_max = 20
    Number of evolution iterations, set in Section IV.B.
  • population size N = 20
    Population size for NSGA-II, set in Section IV.B.
  • classifier parameters = unknown
    The GRU-MLP classifier in Fig. 3 is trained on the datasets; its parameters are not enumerated and are essential to the diversity criterion.
assumptions (3)
  • domain assumption Availability of a pretrained classifier C(Y) that can classify motion types with probabilities.
    Assumed in Section III.A; no training details or accuracy reported.
  • domain assumption The pretrained generative models can be conditioned on a full motion sequence Y_{i-1} plus the endpoints.
    Required by Eq. (8); no implementation details for VAE/GAN/DDPM are given in the main text.
  • domain assumption The distribution of generated motions is smooth enough that evolutionary selection on the two criteria yields realistic transitions.
    Implicit in the design; no formal guarantee beyond the trivial Theorems 1-2.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A Plug-and-Play Multi-Criteria Guidance for Diverse In-Betweening Human Motion Generation." pith.science (2026). https://pith.science/paper/GXWDRJF2

@misc{pith2026250801590,
  author       = {Pith},
  title        = {Pith review of: A Plug-and-Play Multi-Criteria Guidance for Diverse In-Betweening Human Motion Generation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GXWDRJF2}},
  note         = {Machine review of arXiv:2508.01590}
}
read the original abstract

In-betweening human motion generation aims to synthesize intermediate motions that transition between user-specified keyframes. In addition to maintaining smooth transitions, a crucial requirement of this task is to generate diverse motion sequences. It is still challenging to maintain diversity, particularly when it is necessary for the motions within a generated batch sampling to differ meaningfully from one another due to complex motion dynamics. In this paper, we propose a novel method, termed the Multi-Criteria Guidance with In-Betweening Motion Model (MCG-IMM), for in-betweening human motion generation. A key strength of MCG-IMM lies in its plug-and-play nature: it enhances the diversity of motions generated by pretrained models without introducing additional parameters This is achieved by providing a sampling process of pretrained generative models with multi-criteria guidance. Specifically, MCG-IMM reformulates the sampling process of pretrained generative model as a multi-criteria optimization problem, and introduces an optimization process to explore motion sequences that satisfy multiple criteria, e.g., diversity and smoothness. Moreover, our proposed plug-and-play multi-criteria guidance is compatible with different families of generative models, including denoised diffusion probabilistic models, variational autoencoders, and generative adversarial networks. Experiments on four popular human motion datasets demonstrate that MCG-IMM consistently state-of-the-art methods in in-betweening motion generation task.

Figures

Figures reproduced from arXiv: 2508.01590 by the authors.

Figure 1
Figure 1. The example of our method MCG-IMM. Given two human motion [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. The illustration of multi-criteria guidance for diverse in-betweening human motion sequences. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. The illustration of a multi-class classifier model for the Diversity [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: The framework of MCG-IMM. We first randomly sample from the Gaussian distribution, and use the pretrained generator to generate the initial [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: Pareto front of the optimization process under the “Sit” motion [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: The generated diverse in-betweening human motion sequences given [PITH_FULL_IMAGE:figures/full_fig_p009_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

53 extracted references · 44 canonical work pages

  1. [1]

    Denoising diffusion prob- abilistic models.Advances in neural information processing systems, 33:6840–6851, 2020

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffusion prob- abilistic models.Advances in neural information processing systems, 33:6840–6851, 2020

    2020

  2. [2]

    Human motion diffusion as a generative prior

    Yoni Shafir, Guy Tevet, Roy Kapon, and Amit Haim Bermano. Human motion diffusion as a generative prior. InThe Twelfth International Conference on Learning Representations, 2024

    work page 2024

  3. [3]

    An introduction to varia- tional autoencoders.Foundations and Trends® in Machine Learning, 12(4):307–392, 2019

    Diederik P Kingma, Max Welling, et al. An introduction to varia- tional autoencoders.Foundations and Trends® in Machine Learning, 12(4):307–392, 2019

    work page 2019

  4. [4]

    Gen- erative adversarial networks.Communications of the ACM, 63(11):139– 144, 2020

    Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. Gen- erative adversarial networks.Communications of the ACM, 63(11):139– 144, 2020

    work page 2020

  5. [5]

    Score-based generative modeling through stochastic differential equations.arXiv preprint arXiv:2011.13456, 2020

    Yang Song, Jascha Sohl-Dickstein, Diederik P Kingma, Abhishek Kumar, Stefano Ermon, and Ben Poole. Score-based generative modeling through stochastic differential equations.arXiv preprint arXiv:2011.13456, 2020

    arXiv 2011

  6. [6]

    Ta2v: Text-audio guided video generation.IEEE Transactions on Multimedia, 26:7250–7264, 2024

    Minglu Zhao, Wenmin Wang, Tongbao Chen, Rui Zhang, and Ruochen Li. Ta2v: Text-audio guided video generation.IEEE Transactions on Multimedia, 26:7250–7264, 2024

    2024

  7. [7]

    Interaction transformer for human reaction generation.IEEE Transactions on Multimedia, 25:8842–8854, 2023

    Baptiste Chopin, Hao Tang, Naima Otberdout, Mohamed Daoudi, and Nicu Sebe. Interaction transformer for human reaction generation.IEEE Transactions on Multimedia, 25:8842–8854, 2023

    work page 2023

  8. [8]

    Multi-semantics aggrega- tion network based on the dynamic-attention mechanism for 3d human motion prediction.IEEE Transactions on Multimedia, 26:5194–5206, 2024

    Junyu Shi, Jianqi Zhong, and Wenming Cao. Multi-semantics aggrega- tion network based on the dynamic-attention mechanism for 3d human motion prediction.IEEE Transactions on Multimedia, 26:5194–5206, 2024

    work page 2024

Show all 53 references
  1. [9]

    Fusepose: Imu-vision sensor fusion in kinematic space for parametric human pose estimation.IEEE Transactions on Multimedia, 25:7736–7746, 2023

    Yiming Bao, Xu Zhao, and Dahong Qian. Fusepose: Imu-vision sensor fusion in kinematic space for parametric human pose estimation.IEEE Transactions on Multimedia, 25:7736–7746, 2023

    2023

  2. [10]

    Cross- modal quantization for co-speech gesture generation.IEEE Transactions on Multimedia, 26:10251–10263, 2024

    Zheng Wang, Wei Zhang, Long Ye, Dan Zeng, and Tao Mei. Cross- modal quantization for co-speech gesture generation.IEEE Transactions on Multimedia, 26:10251–10263, 2024

    2024

  3. [11]

    I know how you move: Explicit motion estimation for human action recognition.IEEE Transactions on Multimedia, 27:1665–1676, 2025

    Zhongwei Shen, Xiao-Jun Wu, Hui Li, Tianyang Xu, and Cong Wu. I know how you move: Explicit motion estimation for human action recognition.IEEE Transactions on Multimedia, 27:1665–1676, 2025

    2025

  4. [12]

    Human motion diffusion model

    Guy Tevet, Sigal Raab, Brian Gordon, Yoni Shafir, Daniel Cohen-or, and Amit Haim Bermano. Human motion diffusion model. InThe Eleventh International Conference on Learning Representations, 2023

    2023

  5. [13]

    Xiaokai Yi, Hanli Wang, Sam Kwong, and C.-C. Jay Kuo. Task-driven video compression for humans and machines: Framework design and optimization.IEEE Transactions on Multimedia, 25:8091–8102, 2023

    2023

  6. [14]

    Implicit compositional generative network for length-variable co-speech gesture synthesis.IEEE Transactions on Multimedia, 26:6325–6335, 2024

    Chenghao Xu, Jiexi Yan, Yanhua Yang, and Cheng Deng. Implicit compositional generative network for length-variable co-speech gesture synthesis.IEEE Transactions on Multimedia, 26:6325–6335, 2024

    2024

  7. [15]

    Co- herent image animation using spatial-temporal correspondence.IEEE Transactions on Multimedia, 25:3397–3408, 2023

    Weilun Wang, Wengang Zhou, Jianmin Bao, and Houqiang Li. Co- herent image animation using spatial-temporal correspondence.IEEE Transactions on Multimedia, 25:3397–3408, 2023

    2023

  8. [16]

    Attention is all you need.Advances in neural information processing systems, 30, 2017

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. Attention is all you need.Advances in neural information processing systems, 30, 2017

    2017

  9. [17]

    Robust motion in-betweening.ACM Transactions on Graphics (TOG), 39(4):60–1, 2020

    F ´elix G Harvey, Mike Yurick, Derek Nowrouzezahrai, and Christopher Pal. Robust motion in-betweening.ACM Transactions on Graphics (TOG), 39(4):60–1, 2020. 10

    2020

  10. [18]

    Weakly-supervised action transition learning for stochastic human motion prediction

    Wei Mao, Miaomiao Liu, and Mathieu Salzmann. Weakly-supervised action transition learning for stochastic human motion prediction. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 8151–8160, 2022

    2022

  11. [19]

    SCORE-BASED GEN- ERATIVE MODELING THROUGH STOCHASTIC DIFFERENTIAL EQUATIONS

    Yang Song, Jascha Sohl-Dickstein, Diederik P Kingma, Abhishek Kumar, Stefano Ermon, and Ben Poole. SCORE-BASED GEN- ERATIVE MODELING THROUGH STOCHASTIC DIFFERENTIAL EQUATIONS. 2021

    2021

  12. [20]

    Improved techniques for training score- based generative models.Advances in neural information processing systems, 33:12438–12448, 2020

    Yang Song and Stefano Ermon. Improved techniques for training score- based generative models.Advances in neural information processing systems, 33:12438–12448, 2020

    2020

  13. [21]

    Denoising diffusion implicit models

    Jiaming Song, Chenlin Meng, and Stefano Ermon. Denoising diffusion implicit models. InInternational Conference on Learning Representa- tions, 2021

    2021

  14. [22]

    Diffusion models already have a semantic latent space.arXiv preprint arXiv:2210.10960, 2022

    Mingi Kwon, Jaeseok Jeong, and Youngjung Uh. Diffusion models already have a semantic latent space.arXiv preprint arXiv:2210.10960, 2022

    2022 arXiv

  15. [23]

    Motiondiffuser: Controllable multi-agent motion prediction using diffusion

    Chiyu Jiang, Andre Cornman, Cheolho Park, Benjamin Sapp, Yin Zhou, Dragomir Anguelov, et al. Motiondiffuser: Controllable multi-agent motion prediction using diffusion. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 9644– 9653, 2023

    2023

  16. [24]

    Action2motion: Conditioned generation of 3d human motions

    Chuan Guo, Xinxin Zuo, Sen Wang, Shihao Zou, Qingyao Sun, Annan Deng, Minglun Gong, and Li Cheng. Action2motion: Conditioned generation of 3d human motions. InProceedings of the 28th ACM International Conference on Multimedia, pages 2021–2029, 2020

    2021

  17. [25]

    Dlow: Diversifying latent flows for diverse human motion prediction

    Ye Yuan and Kris Kitani. Dlow: Diversifying latent flows for diverse human motion prediction. InComputer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part IX 16, pages 346–364. Springer, 2020

    2020

  18. [26]

    Convolutional autoencoders for human motion infilling

    Manuel Kaufmann, Emre Aksan, Jie Song, Fabrizio Pece, Remo Ziegler, and Otmar Hilliges. Convolutional autoencoders for human motion infilling. In2020 International Conference on 3D Vision (3DV), pages 918–927, 2020

    2020

  19. [27]

    Sd-nerf: Towards lifelike talking head animation via spatially- adaptive dual-driven nerfs.IEEE Transactions on Multimedia, 26:3221– 3234, 2024

    Shuai Shen, Wanhua Li, Xiaoke Huang, Zheng Zhu, Jie Zhou, and Jiwen Lu. Sd-nerf: Towards lifelike talking head animation via spatially- adaptive dual-driven nerfs.IEEE Transactions on Multimedia, 26:3221– 3234, 2024

    2024

  20. [28]

    Recurrent network models for human dynamics

    Katerina Fragkiadaki, Sergey Levine, Panna Felsen, and Jitendra Malik. Recurrent network models for human dynamics. InProceedings of the IEEE international conference on computer vision, pages 4346–4354, 2015

    2015

  21. [29]

    Sequential texts driven cohesive motions synthesis with natural transitions

    Shuai Li, Sisi Zhuang, Wenfeng Song, Xinyu Zhang, Hejia Chen, and Aimin Hao. Sequential texts driven cohesive motions synthesis with natural transitions. InProceedings of the IEEE/CVF International Conference on Computer Vision, pages 9498–9508, 2023

    2023

  22. [30]

    Diverse motion variations for physics-based character animation

    Shailen Agrawal, Shuo Shen, and Michiel Van de Panne. Diverse motion variations for physics-based character animation. InProceedings of the 12th ACM SIGGRAPH/Eurographics Symposium on Computer Animation, pages 37–44, 2013

    2013

  23. [31]

    Real-time controllable motion transition for characters

    Xiangjun Tang, He Wang, Bo Hu, Xu Gong, Ruifan Yi, Qilong Kou, and Xiaogang Jin. Real-time controllable motion transition for characters. ACM Transactions on Graphics (TOG), 41(4):1–10, 2022

    2022

  24. [32]

    Generative tweening: Long-term inbetweening of 3d human motions

    Yi Zhou, Jingwan Lu, Connelly Barnes, Jimei Yang, Sitao Xiang, et al. Generative tweening: Long-term inbetweening of 3d human motions. arXiv preprint arXiv:2005.08891, 2020

    2005 arXiv

  25. [33]

    Towards efficient and diverse generative model for unconditional human motion synthesis

    Hua Yu, Weiming Liu, Jiapeng Bai, Xu Gui, Yaqing Hou, YewSoon Ong, and Qiang Zhang. Towards efficient and diverse generative model for unconditional human motion synthesis. InProceedings of the 32nd ACM International Conference on Multimedia, MM ’24, page 2535–2544, New York, ...

    2024

  26. [34]

    Multiact: Long- term 3d human motion generation from multiple action labels

    Taeryung Lee, Gyeongsik Moon, and Kyoung Mu Lee. Multiact: Long- term 3d human motion generation from multiple action labels. InPro- ceedings of the AAAI Conference on Artificial Intelligence, volume 37, pages 1231–1239, 2023

    2023

  27. [35]

    Omnicontrol: Control any joint at any time for human motion generation, 2024

    Yiming Xie, Varun Jampani, Lei Zhong, Deqing Sun, and Huaizu Jiang. Omnicontrol: Control any joint at any time for human motion generation, 2024

    2024

  28. [36]

    Mofusion: A framework for denoising-diffusion- based motion synthesis

    Rishabh Dabral, Muhammad Hamza Mughal, Vladislav Golyanik, and Christian Theobalt. Mofusion: A framework for denoising-diffusion- based motion synthesis. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 9760–9770, 2023

    2023

  29. [37]

    Flexible motion in-betweening with diffusion models

    Setareh Cohan, Guy Tevet, Daniele Reda, Xue Bin Peng, and Michiel van de Panne. Flexible motion in-betweening with diffusion models. InACM SIGGRAPH 2024 Conference Papers, SIGGRAPH ’24, New York, NY , USA, 2024. Association for Computing Machinery

    2024

  30. [38]

    Divdiff: A conditional diffusion model for diverse human motion pre- diction.IEEE Transactions on Multimedia, 27:1848–1859, 2025

    Hua Yu, Yaqing Hou, Wenbin Pei, Yew-Soon Ong, and Qiang Zhang. Divdiff: A conditional diffusion model for diverse human motion pre- diction.IEEE Transactions on Multimedia, 27:1848–1859, 2025

    2025

  31. [39]

    Deterministic-to-stochastic diverse latent feature mapping for human motion synthesis

    Yu Hua, Weiming Liu, Gui Xu, Yaqing Hou, Yew-Soon Ong, and Qiang Zhang. Deterministic-to-stochastic diverse latent feature mapping for human motion synthesis. InProceedings of the Computer Vision and Pattern Recognition Conference (CVPR), pages 22724–22734, June 2025

    2025

  32. [40]

    Fast direct: Query-efficient online black-box guidance for diffusion-model target generation

    Kim Yong Tan, Yueming Lyu, Ivor Tsang, and Yew-Soon Ong. Fast direct: Query-efficient online black-box guidance for diffusion-model target generation. InThe Thirteenth International Conference on Learning Representations, 2025

    2025

  33. [41]

    Evolvable conditional diffusion.arXiv preprint arXiv:2506.13834, 2025

    Zhao Wei, Chin Chun Ooi, Abhishek Gupta, Jian Cheng Wong, Pao- Hsiung Chiu, Sheares Xue Wen Toh, and Yew-Soon Ong. Evolvable conditional diffusion.arXiv preprint arXiv:2506.13834, 2025

    2025 arXiv

  34. [42]

    Empirical evaluation of gated recurrent neural networks on sequence modeling

    Junyoung Chung, Caglar Gulcehre, Kyunghyun Cho, and Yoshua Ben- gio. Empirical evaluation of gated recurrent neural networks on sequence modeling. InNIPS 2014 Workshop on Deep Learning, December 2014, 2014

    2014

  35. [43]

    A generator for multiobjective test problems with difficult-to-approximate pareto front boundaries.IEEE Transactions on Evolutionary Computation, 23(4):556–571, 2019

    Zhenkun Wang, Yew-Soon Ong, Jianyong Sun, Abhishek Gupta, and Qingfu Zhang. A generator for multiobjective test problems with difficult-to-approximate pareto front boundaries.IEEE Transactions on Evolutionary Computation, 23(4):556–571, 2019

    2019

  36. [44]

    Consistencies and contradictions of performance metrics in multiobjective optimiza- tion.IEEE Transactions on Cybernetics, 44(12):2391–2404, 2014

    Siwei Jiang, Yew-Soon Ong, Jie Zhang, and Liang Feng. Consistencies and contradictions of performance metrics in multiobjective optimiza- tion.IEEE Transactions on Cybernetics, 44(12):2391–2404, 2014

    2014

  37. [45]

    Prompt evolution for generative ai: A classifier- guided approach

    Melvin Wong, Yew-Soon Ong, Abhishek Gupta, Kavitesh Kumar Bali, and Caishun Chen. Prompt evolution for generative ai: A classifier- guided approach. In2023 IEEE Conference on Artificial Intelligence (CAI), pages 226–229, 2023

    2023

  38. [46]

    Motion in-betweening via two- stage transformers.ACM Transactions on Graphics (TOG), 41(6):1–16, 2022

    Jia Qin, Youyi Zheng, and Kun Zhou. Motion in-betweening via two- stage transformers.ACM Transactions on Graphics (TOG), 41(6):1–16, 2022

    2022

  39. [47]

    Action-conditioned 3d human motion synthesis with transformer vae

    Mathis Petrovich, Michael J Black, and G ¨ul Varol. Action-conditioned 3d human motion synthesis with transformer vae. InProceedings of the IEEE/CVF International Conference on Computer Vision, pages 10985– 10995, 2021

    2021

  40. [48]

    Babel: Bodies, action and behavior with english labels

    Abhinanda R Punnakkal, Arjun Chandrasekaran, Nikos Athanasiou, Alejandra Quiros-Ramirez, and Michael J Black. Babel: Bodies, action and behavior with english labels. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 722– 731, 2021

    2021

  41. [49]

    Ntu rgb+ d 120: A large-scale benchmark for 3d human activity understanding.IEEE transactions on pattern analysis and machine intelligence, 42(10):2684–2701, 2019

    Jun Liu, Amir Shahroudy, Mauricio Perez, Gang Wang, Ling-Yu Duan, and Alex C Kot. Ntu rgb+ d 120: A large-scale benchmark for 3d human activity understanding.IEEE transactions on pattern analysis and machine intelligence, 42(10):2684–2701, 2019

    2019

  42. [50]

    Grab: A dataset of whole-body human grasping of objects

    Omid Taheri, Nima Ghorbani, Michael J Black, and Dimitrios Tzionas. Grab: A dataset of whole-body human grasping of objects. InComputer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part IV 16, pages 581–600. Springer, 2020

    2020

  43. [51]

    Vibe: Video inference for human body pose and shape estimation

    Muhammed Kocabas, Nikos Athanasiou, and Michael J Black. Vibe: Video inference for human body pose and shape estimation. InPro- ceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 5253–5263, 2020

    2020

  44. [52]

    A fast and elitist multiobjective genetic algorithm: Nsga-ii.IEEE transactions on evolutionary computation, 6(2):182–197, 2002

    Kalyanmoy Deb, Amrit Pratap, Sameer Agarwal, and TAMT Meyarivan. A fast and elitist multiobjective genetic algorithm: Nsga-ii.IEEE transactions on evolutionary computation, 6(2):182–197, 2002

    2002

  45. [53]

    Conditional motion in-betweening.Pattern Recognition, 132:108894, 2022

    Jihoon Kim, Taehyun Byun, Seungyoun Shin, Jungdam Won, and Sungjoon Choi. Conditional motion in-betweening.Pattern Recognition, 132:108894, 2022

    2022

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.