Pith. sign in

REVIEW 4 major objections 5 minor 1 cited by

Discrete to Continuous: Generating Smooth Transition Poses from Sign Language Observation

T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read Randomly masking frames in continuous sign videos turns transition generation into a supervised problem, and a conditional diffusion model learns to fill the gaps smoothly.

desk verdict Plausible diffusion-based transition generator, but the evaluation measures masked-frame reconstruction, not the claimed discrete-to-continuous stitching task. read the letter →

arxiv 2411.16810 v1 pith:KXQLKU5B submitted 2024-11-25 cs.CV

Shengeng Tang , Jiayi He , Lechao Cheng , Jingjing Wu , Dan Guo , Richang Hong This is my paper · ORCID
classification cs.CV
keywords signlanguageproductiontransitionposegenerationconditionaldiffusionmodelrandommaskingsynthesisvideo
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 proposes Sign-D2C, a framework that generates the missing transition poses needed to stitch isolated word-level signs into continuous sign language sentences. The key trick is to convert the unsupervised task of transition generation into a supervised one: during training, random segments of long continuous signing videos are masked out, and a conditional diffusion model learns to reconstruct those masked frames from the surrounding visible frames. At inference, missing frames between two discrete sign segments are initialized by linear interpolation between the boundary poses and then refined by the same diffusion model. On sentence-level benchmarks the method reports semantic accuracy close to ground truth (BLEU-1 of 69.64% on USTC-CSL100) and far lower mismatch (DTW) than the two compared generation baselines, and it shows qualitative stitching of isolated signs on USTC-SLR500.

What carries the argument

The load-bearing mechanism is a conditional diffusion model in latent space, paired with two training and inference strategies. A pre-trained encoder and decoder handle pose sequences, while a denoiser with self-attention and cross-attention layers removes noise conditionally on observed frames. The random masking strategy simulates the absence of transitions by erasing random frame blocks during training, turning the unsupervised stitching problem into supervised reconstruction. The linear interpolating padding strategy initializes the missing frames between boundary poses at inference, giving the diffusion process a stable starting point. Together these components let the model generate temporally coherent transition poses that preserve surrounding sign context.

What would settle it

One concrete check would be to assemble a test set of continuous sentences for which ground-truth videos exist, artificially split each sentence into two isolated segments at a cut point inside a transition, run Sign-D2C to reconnect them, and measure the mismatches (DTW, MPJPE, and human fluency ratings) between the generated bridging frames and the actual frames at that cut; if the error is no better than simple interpolation for transitions that cross signer or recording-session boundaries, the masking assumption would fail to transfer to realistic concatenation.

Watch

Extended reading notes

Core claim

The central discovery is that transition poses between signs are learnable in a fully self-supervised way from continuous sign video alone, without any manually labeled transition data. By randomly masking a contiguous block of frames in a long video and conditioning the denoiser on the remaining observed frames, the model learns smooth, contextually appropriate bridging motion purely from reconstruction. During inference, the same model fills the gap between two discrete sign segments, using linear interpolation between the boundary frames as a warm start that the iterative denoising process then refines. The paper demonstrates this on PHOENIX14T and USTC-CSL100, where the generated transition frames match the semantics and smoothness of the original continuous sequences, and on USTC-SLR500, where it qualitatively joins isolated vocabulary signs into a continuous signing video.

Load-bearing premise

The load-bearing premise is that randomly masking frames from continuous signing videos produces training and evaluation conditions that faithfully mimic the real transition task of concatenating two isolated word-level signs.

Editorial extensions

If this is right

  • If the framework works as claimed, word-level sign dictionaries become building blocks for sentence-level video synthesis without collecting new paired sentence data.
  • The masking-based supervision recipe could apply to other motion-continuity tasks, such as action stitching or gesture-to-gesture transitions, where only long continuous recordings are available.
  • The near-ground-truth semantic accuracy on USTC-CSL100 suggests the generated transitions preserve enough information for downstream sign-language translation after reconstruction via the NSLT evaluation pipeline.
  • The linear interpolation warm start reduces the difficulty of the diffusion refinement, so shorter denoising schedules may be sufficient for practical stitching.

Reading between the lines

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

  • A likely extension not explored in the paper is to make the masking ratio and mask location adaptive to pose velocity, so transitions are simulated where real coarticulation is hardest, which may improve generalization to variable-length transitions.
  • If the transition model is conditioned on gloss or text embeddings of the two signs being joined, it could produce transitions that are semantically aware of the words' meanings rather than purely pose-driven.
  • The BLEU and DTW numbers on sentence-level datasets are measured on masked-frame reconstruction; a stitched sequence of separately recorded vocabulary clips would be a sterner test because it introduces signer and recording variations, so the real-world gains over baselines remain to be quantified.
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. Sign-D2C is a conditional diffusion framework intended to generate smooth transition frames between discrete sign-language pose segments so that isolated signs can be assembled into continuous signing. During training, the authors randomly mask segments of continuous signing videos and train the model to reconstruct the masked frames from the remaining observed frames. At inference, a linearly interpolated padding initializes the missing gap, which the diffusion model then refines. The paper reports ablations of masking ratio and padding strategy, comparisons against G2P-DDM and VQ-GCDM on PHOENIX14T and USTC-CSL100, and qualitative examples on USTC-SLR500.

Significance. If the proposed method actually delivered smooth, meaning-preserving transitions between independently recorded word-level signs, it would address a genuine bottleneck in sign-language production and synthesis, where continuous sentence-level data are scarce. The idea of converting the unsupervised transition-generation problem into a supervised masked-reconstruction task is appealing, and the intended use of a conditional diffusion model is reasonable. The paper also provides ablations on masking ratios and padding strategies, and the inclusion of three datasets is appropriate. However, the current experimental design does not quantitatively evaluate the stated discrete-to-continuous task, and the numerical results therefore cannot substantiate the central claim.

major comments (4)
  1. [Sec. 3.2 and Tables 1–5] The training and all quantitative evaluations operate on a single continuous signing video, masking frames and reconstructing them from surrounding frames of the same video. The paper's stated task (Sec. 1 and the inference phase in Fig. 1) is instead to synthesize transitions between discrete, separately recorded sign segments. Because observed and masked frames in the same continuous sequence share signer, camera, and grammatical context, performance on the masked-reconstruction benchmarks does not measure generalization to the discrete-to-continuous setting. The only experiment on that setting, Sec. 4.4 and Fig. 4 for USTC-SLR500, reports no quantitative metrics. The headline numbers, e.g., BLEU-1 of 69.64 and DTW of 1.92 in Table 5, thus support only conditional inpainting within continuous videos, not the paper's central claim of generating smooth transitions between discrete segments. This must be addressed with either a quantitative evaluation of the discrete-to-continuous scenario or a substantial reframing of the paper's claims.
  2. [Sec. 4.3 and Tables 3–5] The comparison against G2P-DDM and VQ-GCDM is not informative for the transition-generation task. These baselines are gloss-to-pose generators that synthesize an entire pose sequence from a gloss input, rather than observation-conditioned in-betweeners that must respect the surrounding pose observations. Their poor BLEU and high DTW may simply reflect that they are solving a different, harder problem under the authors' adaptation. The paper should compare against observation-conditioned inpainting baselines, including the trivial linear interpolation baseline and a frame-interpolation method, and should report how the gloss-to-pose baselines were adapted to the masked-setting protocol.
  3. [Sec. 3.3, Eq. (8)] The formula labeled "linear interpolation padding" does not implement linear interpolation. The iterate f(i) = sum_{k=0}^{i} 1/(k+1) produces cumulative harmonic weights, so the intermediate poses pc(i) are spaced by decreasing increments rather than uniformly between ps and pe. As written, the equation contradicts the surrounding text and the "Linear Padding" experimental results in Table 2. The authors must correct the equation or clarify the actual initialization used in inference.
  4. [Sec. 4.2, Table 2 and Sec. 4.1] Because only 10 of 30 frames are predicted in the main 20-observed/10-predicted setting, the reported DTW and MPJPE over the full sequence may be dominated by the observed frames, which are identical to ground truth. A simple copy-observed baseline would likely already achieve very low DTW. The paper should report coherence metrics computed only on the masked (predicted) frames and should include a no-generation baseline such as direct linear interpolation on the same metrics, to assess the incremental value of the diffusion refinement.
minor comments (5)
  1. [Sec. 4.2, Table 2] The text states that the "w/o Padding" approach yields a TEST DTW of 10.36, but the table reports 11.60; the numbers are inconsistent.
  2. [Sec. 4.3] The text says "As shown in Table 4" when referring to the 20-observed/10-predicted comparison, but Table 3 is the 20/10 table; the citation appears to be wrong.
  3. [Sec. 4.3] In the USTC-CSL100 paragraph, the text mentions an "FID score of 3.21" for the 20/10 setting, while Table 5 reports FID 0.31 for Ours; the text likely has a decimal error.
  4. [Sec. 1] The contribution list spells the dataset "UCST-CSL100" while the rest of the paper uses "USTC-CSL100".
  5. [Sec. 3.3, Eqs. (1) and (4)] The noise-adding notation is inconsistent: Eq. (1) uses sqrt(1 - alpha_t) while Eq. (4) uses sqrt(1 - alpha_t^2), with no definition of alpha_t in either place.

Circularity Check

1 steps flagged · score 6.0 of 10

The quantitative headline is masked-frame reconstruction, not discrete-to-continuous transition generation; the evaluation target is the training objective by construction.

  1. fitted input called prediction [Sec. 3.2 (Core Idea) and Sec. 4.1 (Experimental Settings); Tables 3–5]
    "Given a complete continuous sign language sequence X = {x1, x2, · · ·, xn}, we simulate missing transitions by randomly masking portions of X, creating gaps that mimic real-world transition absences. ... This setup frames the task as predicting Xmask based on Xobs, effectively generating supervised training data for transition prediction. ... Since generating continuous sign language videos from discrete fragments is inherently an unsupervised task, there are no standardized evaluation metrics. ..."

    The paper's training step defines 'transitions' as the randomly masked frames Xmask inside a continuous sign video and trains Sign-D2C to predict Xmask from the surrounding Xobs (Eqs. 3–6, 10). The quantitative evaluation (Tables 3–5) then uses the same masked-gap protocol (10/20 and 20/10 frames) and scores BLEU/ROUGE/WER/DTW/FID/MPJPE against those same Xmask ground-truth frames. The evaluated 'prediction' is therefore the exact training objective by construction; the model is never quantitatively tested on the paper's stated task of bridging separately recorded discrete sign segments. The one real discrete-segment experiment (USTC-SLR500, Sec. 4.4) reports only qualitative frames (Fig.

full rationale

The central circularity is that the paper converts 'transition generation' into a random-masking reconstruction task and then uses that same reconstruction task as its only quantitative evaluation. Section 3.2 explicitly defines the supervised target as masked frames Xmask in continuous videos; Section 4.1 states there are no standardized metrics and therefore designs experiments with the same masking strategies (10/20 and 20/10); Tables 3–5 report reconstruction quality against the masked ground truth. This is not an independent test of the discrete-to-continuous assembly claimed in Sec. 1 and Fig. 1. The USTC-SLR500 experiment, which actually concatenates two isolated signs, is presented only as visual examples (Fig. 4), with no BLEU/DTW/MPJPE numbers. I did not find load-bearing self-citation: the authors' own VQ-GCDM [43] is used only as a comparison baseline, and the diffusion/interpolation machinery is not justified by a self-cited uniqueness theorem. The flaw is therefore a self-defined evaluation rather than a derivation that reduces to its inputs; but because the headline claim is supported only by this self-referential proxy, the paper warrants a partial-circularity score of 6.

Assumptions & free parameters 4 free parameters · 3 assumptions · 0 invented entities

The method depends on a few free design choices (masking ratio, transition length, noise schedule) and on the key domain assumption that masked in-video gaps represent real cross-sign transitions. No new physical or model entities are introduced beyond the network architecture itself.

free parameters (4)
  • Masking ratio r = 0.3 (selected on DEV)
    Three values (0.1, 0.3, 0.5) were tried and r=0.3 was chosen based on validation BLEU and DTW, as stated in Section 4.2.
  • Transition window lengths = 10 or 20 frames
    The paper states in Section 4.1 that transitions typically span 10 to 20 frames, but no data analysis is shown to support this; the masking experiments fix these two lengths.
  • Diffusion noise schedule alpha_t = Unspecified
    The forward process in Eq. 4 uses an alpha_t schedule, but the actual schedule values or functional form are not given, so the model can only be reproduced by guessing.
  • Latent dimension W and architecture sizes = Unspecified
    Section 3.3 defines latent features x in R^{F x W} and attention layers, but W, number of heads, hidden sizes, and layer counts are not specified.
assumptions (3)
  • domain assumption Masking frames from continuous sign videos is a valid proxy for missing transitions between isolated signs.
    This assumption underlies the entire training scheme (Section 3.2) and the quantitative evaluation. No experiments validate that masked in-video gaps match cross-video concatenation statistics.
  • domain assumption Sign transitions are approximately 10 to 20 frames long.
    Stated in Section 4.1 ('observed that transitions typically span 10 to 20 frames'), with no supporting measurement or citation; this fixes the prediction lengths for all experiments.
  • domain assumption 2D-to-3D skeleton poses from OpenPose and a skeleton correction model fully capture the semantics of sign language for translation back to text.
    The evaluation uses NSLT on generated poses to compute BLEU/ROUGE/WER; this assumes pose-only representation is sufficient for the downstream translator and that pose errors translate linearly into semantic errors, which is not established.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Discrete to Continuous: Generating Smooth Transition Poses from Sign Language Observation." pith.science (2026). https://pith.science/paper/KXQLKU5B

@misc{pith2026241116810,
  author       = {Pith},
  title        = {Pith review of: Discrete to Continuous: Generating Smooth Transition Poses from Sign Language Observation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KXQLKU5B}},
  note         = {Machine review of arXiv:2411.16810}
}
read the original abstract

Generating continuous sign language videos from discrete segments is challenging due to the need for smooth transitions that preserve natural flow and meaning. Traditional approaches that simply concatenate isolated signs often result in abrupt transitions, disrupting video coherence. To address this, we propose a novel framework, Sign-D2C, that employs a conditional diffusion model to synthesize contextually smooth transition frames, enabling the seamless construction of continuous sign language sequences. Our approach transforms the unsupervised problem of transition frame generation into a supervised training task by simulating the absence of transition frames through random masking of segments in long-duration sign videos. The model learns to predict these masked frames by denoising Gaussian noise, conditioned on the surrounding sign observations, allowing it to handle complex, unstructured transitions. During inference, we apply a linearly interpolating padding strategy that initializes missing frames through interpolation between boundary frames, providing a stable foundation for iterative refinement by the diffusion model. Extensive experiments on the PHOENIX14T, USTC-CSL100, and USTC-SLR500 datasets demonstrate the effectiveness of our method in producing continuous, natural sign language videos.

Figures

Figures reproduced from arXiv: 2411.16810 by the authors.

Figure 1
Figure 1. Task and key steps. Our work aims to generate continu [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Overview. The proposed framework for generating continuous sign language videos with smooth transitions between discrete [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Visualization examples of generating 10-frame transition pose under 20-frame observations on PHOENIX14T. We compare our [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Visualization examples on the USTC-SLR500, demonstrate the generation of transition poses according to two discrete sign [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

  1. Using Sign Language Production as Data Augmentation to enhance Sign Language Translation

    cs.CL 2025-06 conditional novelty 5.0 of 10

    Adding synthetic sign-language data produced by stitching, a GAN, or Gaussian splatting to the training set improves sign-language translation, with the largest gains for skeleton-pose models.

Reference graph

Works this paper leans on

53 extracted references · 50 canonical work pages · cited by 1 Pith paper

  1. [1]

    Development of arabic sign language dictionary using 3d avatar technologies

    Ahmed H Aliwy and A Alethary Ahmed. Development of arabic sign language dictionary using 3d avatar technologies. Indonesian Journal of Electrical Engineering and Computer Science, 21(1):609–616, 2021. 1

    work page 2021

  2. [2]

    Neural sign language trans- lation

    Necati Cihan Camgoz, Simon Hadfield, Oscar Koller, Her- mann Ney, and Richard Bowden. Neural sign language trans- lation. In Computer Vision and Pattern Recognition, pages 7784–7793, 2018. 1, 6

    work page 2018

  3. [3]

    Multi-channel transformers for multi- articulatory sign language translation

    Necati Cihan Camgoz, Oscar Koller, Simon Hadfield, and Richard Bowden. Multi-channel transformers for multi- articulatory sign language translation. In European Confer- ence on Computer Vision, pages 301–319, 2020. 1, 2

    work page 2020

  4. [4]

    Realtime multi-person 2d pose estimation using part affinity fields

    Zhe Cao, Tomas Simon, Shih-En Wei, and Yaser Sheikh. Realtime multi-person 2d pose estimation using part affinity fields. In Computer Vision and Pattern Recognition , pages 7291–7299, 2017. 6

    work page 2017

  5. [5]

    The devisign large vocabulary of chinese sign language database and base- line evaluations

    Xiujuan Chai, Hanjie Wang, and Xilin Chen. The devisign large vocabulary of chinese sign language database and base- line evaluations. In Technical report, Key Lab of Intelli- gent Information Processing of CAS. Institute of Computing Technology, 2014. 1

    work page 2014

  6. [6]

    Querying a sign lan- guage dictionary with videos using dense vector search

    Mathieu De Coster and Joni Dambre. Querying a sign lan- guage dictionary with videos using dense vector search. In 2023 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops, pages 1–5. IEEE, 2023. 1

    work page 2023

  7. [7]

    Benchmark databases for video- based automatic sign language recognition

    Philippe Dreuw, Carol Neidle, Vassilis Athitsos, Stan Sclaroff, and Hermann Ney. Benchmark databases for video- based automatic sign language recognition. In International Conference on Language Resources and Evaluation, 2008. 1

    work page 2008

  8. [8]

    Cross-modal neural sign language translation

    Amanda Cardoso Duarte. Cross-modal neural sign language translation. In ACM International Conference on Multime- dia, pages 1650–1654, 2019. 1

    work page 2019

Show all 53 references
  1. [9]

    Llms are good sign language translators

    Jia Gong, Lin Geng Foo, Yixuan He, Hossein Rahmani, and Jun Liu. Llms are good sign language translators. In Com- puter Vision and Pattern Recognition , pages 18362–18372,

  2. [10]

    Online early-late fusion based on adaptive hmm for sign language recognition

    Dan Guo, Wengang Zhou, Houqiang Li, and Meng Wang. Online early-late fusion based on adaptive hmm for sign language recognition. ACM Transactions on Multimedia Computing, Communications, and Applications, pages 1–18,

  3. [11]

    Hierarchical lstm for sign language translation

    Dan Guo, Wengang Zhou, Houqiang Li, and Meng Wang. Hierarchical lstm for sign language translation. In AAAI Conference on Artificial Intelligence , pages 6845–6852,

  4. [12]

    Connectionist temporal modeling of video and language: A joint model for translation and sign labeling

    Dan Guo, Shengeng Tang, and Meng Wang. Connectionist temporal modeling of video and language: A joint model for translation and sign labeling. In International Joint Confer- ence on Artificial Intelligence, pages 751–757, 2019. 2

    2019

  5. [13]

    Sign language recognition

    Dan Guo, Shengeng Tang, Richang Hong, and Meng Wang. Sign language recognition. Multimedia for Accessible Hu- man Computer Interfaces, pages 23–59, 2021. 1

    2021

  6. [14]

    Denoising diffusion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffusion probabilistic models. Conference and Workshop on Neural Information Processing Systems , 33:6840–6851,

  7. [15]

    Collaborative multilingual continuous sign language recog- nition: A unified framework

    Hezhen Hu, Junfu Pu, Wengang Zhou, and Houqiang Li. Collaborative multilingual continuous sign language recog- nition: A unified framework. IEEE Transactions on Multi- media, pages 7559–7570, 2022. 1, 2

    2022

  8. [16]

    Signbert+: Hand-model-aware self-supervised pre- training for sign language understanding

    Hezhen Hu, Weichao Zhao, Wengang Zhou, and Houqiang Li. Signbert+: Hand-model-aware self-supervised pre- training for sign language understanding. IEEE Transactions on Pattern Analysis and Machine Intelligence, pages 11221– 11239, 2023. 1, 2

    2023

  9. [17]

    Attention-based 3d-cnns for large-vocabulary sign language recognition

    Jie Huang, Wengang Zhou, Houqiang Li, and Weiping Li. Attention-based 3d-cnns for large-vocabulary sign language recognition. IEEE Transactions on Circuits and Systems for Video Technology, pages 2822–2832, 2018. 6

    2018

  10. [18]

    Video-based sign language recognition without temporal segmentation

    Jie Huang, Wengang Zhou, Qilin Zhang, Houqiang Li, and Weiping Li. Video-based sign language recognition without temporal segmentation. In AAAI Conference on Artificial In- telligence, pages 2257–2264, 2018. 1, 6

    2018

  11. [19]

    To- wards fast and high-quality sign language production

    Wencan Huang, Wenwen Pan, Zhou Zhao, and Qi Tian. To- wards fast and high-quality sign language production. In ACM International Conference on Multimedia, pages 3172– 3181, 2021. 6

    2021

  12. [20]

    Developing a bilingual mobile dictionary for in- dian sign language and gathering users experience with sign- dict

    Jestin Joy, Kannan Balakrishnan, and Sreeraj Madha- vankutty. Developing a bilingual mobile dictionary for in- dian sign language and gathering users experience with sign- dict. Assistive Technology, 2020. 1

    2020

  13. [21]

    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 International Con- ference on Computer Vision, pages 2151–2162, 2023. 3

    2023

  14. [22]

    Quantitative survey of the state of the art in sign language recognition

    Oscar Koller. Quantitative survey of the state of the art in sign language recognition. ArXiv Preprint ArXiv:2008.09918, pages ArXiv–2008, 2020. 1, 2

    2008 arXiv

  15. [23]

    Grid diffusion models for text-to-video generation

    Taegyeong Lee, Soyeong Kwon, and Taehwan Kim. Grid diffusion models for text-to-video generation. In Computer Vision and Pattern Recognition, pages 8734–8743, 2024. 2

    2024

  16. [24]

    Word-level deep sign language recognition from video: A new large-scale dataset and methods comparison

    Dongxu Li, Cristian Rodriguez, Xin Yu, and Hongdong Li. Word-level deep sign language recognition from video: A new large-scale dataset and methods comparison. In IEEE Winter Conference on Applications of Computer Vi- sion, pages 1459–1469, 2020. 1

    2020

  17. [25]

    Tspnet: Hier- archical feature learning via temporal semantic pyramid for sign language translation

    Dongxu Li, Chenchen Xu, Xin Yu, Kaihao Zhang, Benjamin Swift, Hanna Suominen, and Hongdong Li. Tspnet: Hier- archical feature learning via temporal semantic pyramid for sign language translation. Neural Information Processing Systems, pages 12034–12045, 2020. 2

    2020

  18. [26]

    Movideo: Motion-aware video generation with diffusion model

    Jingyun Liang, Yuchen Fan, Kai Zhang, Radu Timofte, Luc Van Gool, and Rakesh Ranjan. Movideo: Motion-aware video generation with diffusion model. In European Con- ference on Computer Vision , pages 56–74. Springer, 2025. 3

    2025

  19. [27]

    Rouge: A package for automatic evaluation of summaries

    Chin-Yew Lin. Rouge: A package for automatic evaluation of summaries. In Text Summarization Branches Out, pages 74–81, 2004. 6

    2004

  20. [28]

    Translating sign language videos to talking faces

    Seshadri Mazumder, Rudrabha Mukhopadhyay, Vinay P Namboodiri, and CV Jawahar. Translating sign language videos to talking faces. In Indian Conference on Computer Vision, Graphics and Image Processing , pages 1–10, 2021. 2 9

    2021

  21. [29]

    An auto- mated technique for real-time production of lifelike anima- tions of american sign language

    John McDonald, Rosalee Wolfe, Jerry Schnepp, Julie Hochgesang, Diana Gorman Jamrozik, Marie Stumbo, Lar- wan Berke, Melissa Bialek, and Farah Thomas. An auto- mated technique for real-time production of lifelike anima- tions of american sign language. Universal Access in the I...

    2016

  22. [30]

    Dds2m: Self-supervised denoising diffusion spatio- spectral model for hyperspectral image restoration

    Yuchun Miao, Lefei Zhang, Liangpei Zhang, and Dacheng Tao. Dds2m: Self-supervised denoising diffusion spatio- spectral model for hyperspectral image restoration. In In- ternational Conference on Computer Vision , pages 12086– 12096, 2023. 3

    2023

  23. [31]

    Dit-3d: Exploring plain diffusion transformers for 3d shape generation

    Shentong Mo, Enze Xie, Ruihang Chu, Lanqing Hong, Matthias Niessner, and Zhenguo Li. Dit-3d: Exploring plain diffusion transformers for 3d shape generation. Neural In- formation Processing Systems, 36:67960–67971, 2023. 2

    2023

  24. [32]

    Conditional image-to-video genera- tion with latent flow diffusion models

    Haomiao Ni, Changhao Shi, Kai Li, Sharon X Huang, and Martin Renqiang Min. Conditional image-to-video genera- tion with latent flow diffusion models. In Computer Vision and Pattern Recognition, pages 18444–18455, 2023. 3

    2023

  25. [33]

    Bleu: a method for automatic evaluation of machine translation

    Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. Bleu: a method for automatic evaluation of machine translation. In Association for Computational Linguistics , pages 311–318, 2002. 6

    2002

  26. [34]

    High-resolution image syn- thesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bj¨orn Ommer. High-resolution image syn- thesis with latent diffusion models. In Computer Vision and Pattern Recognition, pages 10684–10695, 2022. 2

    2022

  27. [35]

    Progressive transformers for end-to-end sign language pro- duction

    Ben Saunders, Necati Cihan Camgoz, and Richard Bowden. Progressive transformers for end-to-end sign language pro- duction. In European Conference on Computer Vision, pages 687–705, 2020. 2, 6

    2020

  28. [36]

    Mixed signals: Sign language production via a mixture of motion primitives

    Ben Saunders, Necati Cihan Camgoz, and Richard Bowden. Mixed signals: Sign language production via a mixture of motion primitives. In International Conference on Computer Vision, pages 1919–1929, 2021. 6

    1919

  29. [37]

    A study of sign language coarticulation

    J ´er´emie Segouat. A study of sign language coarticulation. ACM Sigaccess Accessibility and Computing , pages 31–38,

  30. [38]

    Diffusion-based 3d human pose estimation with multi- hypothesis aggregation

    Wenkang Shan, Zhenhua Liu, Xinfeng Zhang, Zhao Wang, Kai Han, Shanshe Wang, Siwei Ma, and Wen Gao. Diffusion-based 3d human pose estimation with multi- hypothesis aggregation. In International Conference on Computer Vision, pages 14761–14771, 2023. 3

    2023

  31. [39]

    Deep unsupervised learning using nonequilibrium thermodynamics

    Jascha Sohl-Dickstein, Eric Weiss, Niru Maheswaranathan, and Surya Ganguli. Deep unsupervised learning using nonequilibrium thermodynamics. In International Confer- ence on Machine Learning, pages 2256–2265, 2015. 2

    2015

  32. [40]

    Text2sign: Towards sign language produc- tion using neural machine translation and generative adver- sarial networks

    Stephanie Stoll, Necati Cihan Camgoz, Simon Hadfield, and Richard Bowden. Text2sign: Towards sign language produc- tion using neural machine translation and generative adver- sarial networks. International Journal of Computer Vision , pages 891–908, 2020. 2

    2020

  33. [41]

    Sign language generation system based on indian sign language grammar

    Sugandhi, Parteek Kumar, and Sanmeet Kaur. Sign language generation system based on indian sign language grammar. ACM Transactions on Asian and Low-Resource Language Information Processing, 19(4):1–26, 2020. 1

    2020

  34. [42]

    Graph-based multimodal sequential embedding for sign lan- guage translation

    Shengeng Tang, Dan Guo, Richang Hong, and Meng Wang. Graph-based multimodal sequential embedding for sign lan- guage translation. IEEE Transactions on Multimedia , 24: 4433–4445, 2022. 1

    2022

  35. [43]

    Gloss-driven conditional diffusion mod- els for sign language production

    Shengeng Tang, Feng Xue, Jingjing Wu, Shuo Wang, and Richang Hong. Gloss-driven conditional diffusion mod- els for sign language production. ACM Transactions on Multimedia Computing, Communications, and Applications,

  36. [44]

    Including facial expressions in contextual embed- dings for sign language generation

    Carla Viegas, Mert Inan, Lorna Quandt, and Malihe Alikhani. Including facial expressions in contextual embed- dings for sign language generation. In Joint Conference on Lexical and Computational Semantics, pages 1–10, 2023. 6

    2023

  37. [45]

    Diffir: Efficient diffusion model for image restoration

    Bin Xia, Yulun Zhang, Shiyin Wang, Yitong Wang, Xing- long Wu, Yapeng Tian, Wenming Yang, and Luc Van Gool. Diffir: Efficient diffusion model for image restoration. In In- ternational Conference on Computer Vision , pages 13095– 13105, 2023. 3

    2023

  38. [46]

    Boxdiff: Text-to-image synthesis with training-free box-constrained diffusion

    Jinheng Xie, Yuexiang Li, Yawen Huang, Haozhe Liu, Wen- tian Zhang, Yefeng Zheng, and Mike Zheng Shou. Boxdiff: Text-to-image synthesis with training-free box-constrained diffusion. In International Conference on Computer Vision, pages 7452–7461, 2023. 3

    2023

  39. [47]

    G2p-ddm: Generating sign pose se- quence from gloss sequence with discrete diffusion model

    Pan Xie, Qipeng Zhang, Peng Taiying, Hao Tang, Yao Du, and Zexian Li. G2p-ddm: Generating sign pose se- quence from gloss sequence with discrete diffusion model. In AAAI Conference on Artificial Intelligence , pages 6234– 6242, 2024. 2, 3, 7, 8

    2024

  40. [48]

    Fresco: Spatial-temporal correspondence for zero-shot video translation

    Shuai Yang, Yifan Zhou, Ziwei Liu, and Chen Change Loy. Fresco: Spatial-temporal correspondence for zero-shot video translation. In Computer Vision and Pattern Recognition , pages 8703–8712, 2024. 2

    2024

  41. [49]

    Diff-retinex: Rethinking low-light image enhancement with a generative diffusion model

    Xunpeng Yi, Han Xu, Hao Zhang, Linfeng Tang, and Jiayi Ma. Diff-retinex: Rethinking low-light image enhancement with a generative diffusion model. In International Confer- ence on Computer Vision, pages 12302–12311, 2023. 3

    2023

  42. [50]

    Neural sign language synthe- sis: Words are our glosses

    Jan Zelinka and Jakub Kanis. Neural sign language synthe- sis: Words are our glosses. In Winter Conference on Appli- cations of Computer Vision , pages 3395–3403, 2020. 1, 2, 6

    2020

  43. [51]

    Highly fluent sign language synthesis based on variable motion frame interpolation

    Ni Zeng, Yiqiang Chen, Yang Gu, Dongdong Liu, and Yun- bing Xing. Highly fluent sign language synthesis based on variable motion frame interpolation. In 2020 IEEE Interna- tional Conference on Systems, Man, and Cybernetics, pages 1772–1777. IEEE, 2020. 1

    2020

  44. [52]

    Chinese sign language recognition with adap- tive hmm

    Jihai Zhang, Wengang Zhou, Chao Xie, Junfu Pu, and Houqiang Li. Chinese sign language recognition with adap- tive hmm. In IEEE International Conference on Multimedia and Expo, pages 1–6. IEEE, 2016. 1

    2016

  45. [53]

    Best: Bert pre-training for sign language recognition with coupling tokenization

    Weichao Zhao, Hezhen Hu, Wengang Zhou, Jiaxin Shi, and Houqiang Li. Best: Bert pre-training for sign language recognition with coupling tokenization. In AAAI Conference on Artificial Intelligence, pages 3597–3605, 2023. 1, 2 10

    2023

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.