Pith. sign in

REVIEW 4 major objections 6 minor 2 cited by

Generative Inbetweening through Frame-wise Conditions-Driven Video Generation

T0 review · 4 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read This paper argues that injecting per-frame interpolated line and pose conditions into every denoising step removes the interpolation-path ambiguity destabilizing generative inbetweening, producing temporally stable videos even for large…

desk verdict A practical, honestly-limited method for stabilizing generative inbetweening via per-frame control conditions; deserves review but needs stronger temporal-stability evidence. read the letter →

arxiv 2412.11755 v1 pith:ZBQZ5AIA submitted 2024-12-16 cs.CV

Tianyi Zhu , Dongwei Ren , Qilong Wang , Xiaohe Wu , Wangmeng Zuo This is my paper · ORCID
classification cs.CV
keywords generativeinbetweeningvideoframeinterpolationframe-wiseconditionsdiffusionmodelstemporalstabilitylinematchingpose-guidedgenerationimage-to-video
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 argues that generative inbetweening — synthesizing intermediate frames between two key frames with a video diffusion model — is unstable because the interpolation path is ambiguous: the start and end frames condition only the first and last frames. To remove that ambiguity, the authors propose FCVG, which gives every intermediate frame its own explicit condition by linearly interpolating color-coded matched-line maps, optionally augmented with pose skeletons, between the two key frames. These frame-wise conditions are injected at every denoising step of a pretrained image-to-video model, aligning the forward and backward sampling paths so that a simple weighted fusion stays close to the video manifold. The paper reports that this yields temporally stable inbetweening under large motion gaps, with fewer sampling steps, no noise re-injection, and support for user-specified non-linear interpolation paths.

What carries the argument

The load-bearing object is the frame-wise condition sequence, a 'control path' of color-coded line-match images, optionally with pose skeletons, interpolated frame by frame between the two key frames and fed to the denoiser at every denoising step. The forward and backward condition sequences are time-flips of each other, so both sampling directions see the same intermediate guidance; the fusion is simply $z_t = \lambda \tilde z_t + (1-\lambda)\,\mathrm{flip}(\tilde z'_t)$ with linear weights $\lambda_i = 1 - (i-1)/(N-1)$. The conditions enter through a lightweight control branch that uses cross-normalization, and only the value and output projections of attention layers plus small residual blocks are fine-tuned, keeping the pretrained model largely frozen.

What would settle it

Feed the method a keyframe pair with large motion and repetitive texture, such as a patterned curtain or a crowd scene; if the line matcher yields few or wrong correspondences, intermediate frames should show abrupt jumps or ghosting. If no setting of the control weight removes those artifacts, the claim that frame-wise conditions confine simple fusion close to the manifold fails in exactly the large-gap regime the paper targets.

Watch

Extended reading notes

Core claim

The central claim is that a frame-wise 'control path' aligns the forward and backward denoising paths so closely that a plain weighted average of the two predictions stays close to the video manifold. The authors extract matched lines between start and end frames, render them as color-coded images, optionally add pose skeletons, and linearly interpolate these maps frame by frame to form the condition sequence. Each denoising step then conditions on the frame it is producing, not just on the start or end frame, and the backward path uses the same condition sequence flipped in time. The paper reports that this substantially reduces incoherent transitions and artifacts under large motion while keeping frame quality, and that the method works across natural scenes, human poses, camera motion, and animation and line art.

Load-bearing premise

The load-bearing premise is that lines and poses extracted from the two key frames, rendered as images and linearly interpolated, provide a faithful description of the motion path for every intermediate frame; when matches are incorrect or too sparse, the control path itself misleads the denoiser and artifacts persist.

Editorial extensions

If this is right

  • Generative inbetweening becomes stable at large motion gaps where optical-flow methods and prior diffusion-based fusion approaches produce incoherent transitions.
  • Noise re-injection is unnecessary and the number of denoising steps can be halved relative to prior time-reversal approaches, substantially reducing inference cost.
  • Users can specify non-linear interpolation paths, such as ease-in and ease-out motion, and the same frame-wise condition mechanism follows them.
  • The approach generalizes to animation and line art without fine-tuning on those data types, because the control path itself encodes the motion.
  • A single control weight setting works across diverse scenes, removing per-input tuning of motion-related hyper-parameters.

Reading between the lines

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

  • Editorial inference: The control-path idea is a general recipe: any per-frame signal that can be interpolated between keyframes, such as depth, edges, optical flow, or text prompts, could replace lines and poses in the same injection mechanism.
  • Editorial inference: If the line-condition bottleneck is real, replacing the hand-crafted matcher with a learned semantic correspondence module, or with user-specified correspondences, should extend the same stabilization to scenes with occlusion and repetitive texture.
  • Editorial inference: Because the paper's own metrics do not precisely measure temporal stability, a dedicated stability metric that checks path consistency under non-linear trajectories would make the claimed advantage directly measurable and comparable across methods.
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 / 6 minor

Summary. The paper proposes Frame-wise Conditions-driven Video Generation (FCVG) for generative inbetweening. Given two keyframes, FCVG extracts GlueStick line matches and DWPose pose skeletons, visualizes them as colored images, interpolates these visual conditions frame-by-frame (linearly or with user-specified non-linear paths), and injects them into Stable Video Diffusion via a ControlNeXt-style lightweight adapter. Inference uses bidirectional time-reversal denoising with a simple weighted average fusion of the two paths. The authors report quantitative comparisons on a collected dataset (524 clips from DAVIS, RealEstate10K, and Pexels) against FILM, DynamiCrafter, TRF, and GI, plus ablations, computational-efficiency measurements, and qualitative generalization to animation and line art. The central claim is that per-frame explicit conditions align the forward and backward denoising paths, thereby improving temporal stability especially for large motion gaps, while preserving frame quality.

Significance. If validated, the method would be a simple and practical addition to generative inbetweening: it requires only a small fine-tuned adapter, avoids noise reinjection, halves the number of denoising steps relative to prior generative inbetweening methods, and offers user control over the interpolation path. The paper is honest about its limitations and provides a project page and code link, which supports reproducibility. The main weakness is that the headline claim—temporal stability under large motion—is supported primarily by selected qualitative videos and by aggregate metrics that the authors themselves state cannot precisely measure temporal stability. The mechanism also depends on the reliability of line matching in exactly the large-motion regime where the paper acknowledges sparse matches cause artifacts. These gaps prevent the current evidence from fully establishing the central advantage.

major comments (4)
  1. [§4.1, Table 1] The paper's central claim is 'significantly enhanced temporal stability', but Table 1 reports no confidence intervals, no multiple-seed results, and no significance tests, and the text in §4.1 explicitly states that all used metrics 'are not capable of precisely evaluating temporal stability'. As a result, the quantitative evidence does not substantiate the central claim. I ask the authors to either add a dedicated temporal-coherence metric or human evaluation, or at minimum report means and variances over multiple inference seeds for LPIPS, FID, FVMD, and FVD so readers can judge whether the observed differences (e.g., FVMD 5607.2 vs. 5682.6 at gap 23) are meaningful.
  2. [§5, §3.2] The method's advantage is claimed to be largest for large motion gaps, but §5 states that when input frames differ significantly, matched lines may be sparse and artifacts cannot be fixed by adjusting gamma. This is precisely the regime the headline claim targets. The paper should include a stratified evaluation by match density or motion magnitude (e.g., bins based on number of GlueStick matches or optical-flow magnitude) and show that FCVG retains its advantage in the sparse-match regime. Without such a test, the core mechanism is only demonstrated where conditions are reliable, not where the paper claims it matters most.
  3. [§3.2, §3.3, Algorithm 1] The assertion that frame-wise conditions 'ensure that the forward and backward paths are relatively aligned during the denoising process' is presented without direct evidence. The final-frame metrics in Table 1 are an indirect test. I suggest a direct measurement of path alignment, such as the average latent distance between z_t from the forward path and flip(z_t) from the backward path during sampling, with and without frame-wise conditions. This would test the proposed mechanism more convincingly and help explain why simple weighted fusion in Eq. (5) suffices.
  4. [§4.5, Tables 3 and 4] The ablation results lack error bars, and some differences are small or inconsistent with the narrative. For example, in Table 3, removing the pose condition improves FVMD (5520.9 vs. 5607.2 for the full model), and in Table 4, gamma=2.0 gives FID 24.66 and FVMD 5726.9 while gamma=0.5 gives FID 23.80 and FVMD 5920.0, so no clear monotonic trend in the metrics supports the claim that the method is insensitive to gamma. Reporting multiple seeds with confidence intervals would make the component analysis more reliable.
minor comments (6)
  1. [§4.5] The phrase 'We conduct ablations to to discuss' contains a duplicated word 'to'.
  2. [§4.1] The word 'outdor' in 'indoor/outdor scenes' should be 'outdoor'.
  3. [Eq. (3) vs. Algorithm 1] Equation (3) writes the denoiser input as z_{t+1}, while Algorithm 1 at loop step t uses z_t. Please make the time indexing consistent between the equation, the algorithm, and Figure 4.
  4. [§3.3.1, Algorithm 1] The notation c_{1→N} is described as a concatenation of frame-wise conditions along the time dimension, but Algorithm 1 writes 'c_{1→N} = interpolate(c_1, c_N)' without showing the concatenation. Please clarify the exact tensor shape and how interpolation and concatenation compose.
  5. [§4.1] The test set is only one fifth of 524 clips, which is roughly 105 clips, and the split appears to be random. Please report the number of test clips and consider providing the dataset or split metadata so that the results can be reproduced.
  6. [§4.1] Fine-tuning uses 512×320 crops while inference is performed at 1024×576. A brief discussion of this resolution mismatch and its potential effect on the adapter would be helpful.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: FCVG's central claims are empirical and tested against external baselines; the only mild self-referential elements are the default control weight and an inspirational self-citation.

full rationale

The paper does not derive a prediction from a fitted parameter or from a self-citation chain. The core method is conditional generation: GlueStick and DWPose (external pretrained models) produce initial conditions, linear interpolation extends them frame-wise, and a lightweight ControlNeXt-style injection feeds them into Stable Video Diffusion. The claim in Sec. 3.2 that frame-wise conditions align forward and backward denoising paths is presented as a design motivation and then tested through ablations and comparisons with independent baselines (FILM, TRF, GI, DynamiCrafter) on external metrics (LPIPS, FID, FVD, FVMD, VBench). No equation reduces to its own input: Eq. 5 is a standard linear fusion of two denoiser outputs, and Eq. 6 is a condition-injection operation whose effect is empirically measured. The self-citation [61] is used only as inspiration for global matching in line-art settings, not as a load-bearing proof or uniqueness theorem. The authors' own Sec. 5 acknowledges failure cases when matches are sparse or incorrect, which further confirms that the method's benefits are not tautological. The mild score of 1 reflects only the choice of the default control weight gamma from the authors' own experiments and the presence of a non-load-bearing self-citation; neither makes the central result circular.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

The central claim rests on the reliability of pretrained condition extractors and on the assumption that linearly interpolated condition visualizations form a valid control path. These are engineering assumptions rather than derived facts, and the paper's own limitation section weakens the line-matching one. No new physical or mathematical entities are introduced.

free parameters (1)
  • Control weight gamma = 1.0 (default; 0.5 and 2.0 also tested)
    Balances the strength of frame-wise condition injection against SVD features. The default is chosen from the authors' own experiments, not derived from theory.
assumptions (4)
  • domain assumption Pretrained GlueStick line matching produces correct, dense correspondences between start and end frames across large motion gaps.
    The entire frame-wise condition construction in Sec 3.3.1 assumes these matches are reliable; Sec 5 documents failure cases.
  • ad hoc to paper Interpolating the rendered condition images linearly yields meaningful intermediate control signals that correspond to plausible intermediate frames.
    The method's core proposal in Sec 3.3.1; no theoretical or empirical proof that the interpolation path of condition images matches true scene motion.
  • ad hoc to paper Visualized matches and pose skeletons are sufficient conditions for the pretrained SVD to stay on-manifold during time-reversal fusion.
    Injection via ControlNeXt is assumed to guide the denoiser; only indirect ablations give support.
  • domain assumption Evaluation metrics FVD, FVMD, and VBench capture the relevant notion of temporal stability.
    Used in Table 1, yet the paper itself states that these metrics are not capable of precisely evaluating temporal stability.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Generative Inbetweening through Frame-wise Conditions-Driven Video Generation." pith.science (2026). https://pith.science/paper/ZBQZ5AIA

@misc{pith2026241211755,
  author       = {Pith},
  title        = {Pith review of: Generative Inbetweening through Frame-wise Conditions-Driven Video Generation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZBQZ5AIA}},
  note         = {Machine review of arXiv:2412.11755}
}
read the original abstract

Generative inbetweening aims to generate intermediate frame sequences by utilizing two key frames as input. Although remarkable progress has been made in video generation models, generative inbetweening still faces challenges in maintaining temporal stability due to the ambiguous interpolation path between two key frames. This issue becomes particularly severe when there is a large motion gap between input frames. In this paper, we propose a straightforward yet highly effective Frame-wise Conditions-driven Video Generation (FCVG) method that significantly enhances the temporal stability of interpolated video frames. Specifically, our FCVG provides an explicit condition for each frame, making it much easier to identify the interpolation path between two input frames and thus ensuring temporally stable production of visually plausible video frames. To achieve this, we suggest extracting matched lines from two input frames that can then be easily interpolated frame by frame, serving as frame-wise conditions seamlessly integrated into existing video generation models. In extensive evaluations covering diverse scenarios such as natural landscapes, complex human poses, camera movements and animations, existing methods often exhibit incoherent transitions across frames. In contrast, our FCVG demonstrates the capability to generate temporally stable videos using both linear and non-linear interpolation curves. Our project page and code are available at \url{https://fcvg-inbetween.github.io/}.

Figures

Figures reproduced from arXiv: 2412.11755 by the authors.

Figure 1
Figure 1. Video results of FILM [32], TRF [7], GI [44], and our FCVG. These results are presented as animated videos that can be viewed in Adobe PDF reader. More videos are provided in the project page. In pioneering video interpolation methods [1, 18, 24], the linear assumption is commonly adopted, which may not truly align with ground-truth temporal consistency but can lead to temporally stable videos for most scenes. As de… view at source ↗
Figure 3
Figure 3. The process for acquiring forward and backward frame [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figure 4
Figure 4. Inference of FCVG at time t. ambiguity in the interpolation path primarily arises from in￾sufficient conditions for intermediate frames, since two in￾put images only provide conditions for start and end frames. Therefore, in this work, we suggest offering an explicit con￾dition for each frame, which significantly alleviates the am￾biguity of the interpolation path. As shown in [PITH_FULL_IMAGE:figures/full_fig_p004… view at source ↗
Figures from the paper (7 more)
Figure 5
Figure 5. Figure 5: Overview of injecting frame-wise conditions into SVD. [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 6
Figure 6. Figure 6: Qualitative evaluation on diverse scenes, where our FCVG is superior in texture details and coherent intermediate motions. [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]
Figure 7
Figure 7. Figure 7: The effect of control weight γ. Red arrows indicate the movement directions. As the value of γ decreases, the diversity of intermediate motion increases, e.g., up-and-down swinging of the arm, while temporal stability is guaranteed for two cases. The videos are availab…
Figure 8
Figure 8. Figure 8: FCVG is able to handle linear and non-linear interpola [PITH_FULL_IMAGE:figures/full_fig_p007_8.png]
Figure 9
Figure 9. Figure 9: Inbetweening results on animations and linearts. FCVG exibits favorable performance without fine-tuning on these data types. [PITH_FULL_IMAGE:figures/full_fig_p008_9.png]
Figure 11
Figure 11. Figure 11: Failure cases. Incorrect matches and significant differ [PITH_FULL_IMAGE:figures/full_fig_p008_11.png]
Figure 10
Figure 10. Figure 10: Ablation study on condition components [PITH_FULL_IMAGE:figures/full_fig_p008_10.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

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

  1. DanceTogether! Identity-Preserving Multi-Person Interactive Video Generation

    cs.CV 2025-05 conditional novelty 6.0 of 10

    A diffusion model fuses per-person masks with pose keypoints to generate identity-preserving, two-person interactive videos from a single reference image, outperforming prior single-person-animation pipelines.

  2. Semantic Frame Interpolation

    cs.CV 2025-07 reject novelty 5.0 of 10

    The authors define Semantic Frame Interpolation, build a 300k-clip dataset and benchmark, and propose a Mixture-of-LoRA adaptation of Wan2.1 that improves temporal smoothness but does not preserve the given start and ...

Reference graph

Works this paper leans on

61 extracted references · 35 canonical work pages · cited by 2 Pith papers

  1. [1]

    A database and evaluation methodology for optical flow.IJCV, 92:1–31,

    Simon Baker, Daniel Scharstein, James P Lewis, Stefan Roth, Michael J Black, and Richard Szeliski. A database and evaluation methodology for optical flow.IJCV, 92:1–31,

    work page

  2. [2]

    Depth-aware video frame interpolation

    Wenbo Bao, Wei-Sheng Lai, Chao Ma, Xiaoyun Zhang, Zhiyong Gao, and Ming-Hsuan Yang. Depth-aware video frame interpolation. In CVPR, pages 3703–3712, 2019. 2

    work page 2019

  3. [3]

    Stable video diffusion: Scaling latent video diffusion models to large datasets

    Andreas Blattmann, Tim Dockhorn, Sumith Kulal, Daniel Mendelevitch, Maciej Kilian, Dominik Lorenz, Yam Levi, Zion English, Vikram V oleti, Adam Letts, et al. Stable video diffusion: Scaling latent video diffusion models to large datasets. arXiv preprint arXiv:2311.15127, 2023. 1, 3, 4

    arXiv 2023

  4. [4]

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

    Andreas Blattmann, Robin Rombach, Huan Ling, Tim Dock- horn, Seung Wook Kim, Sanja Fidler, and Karsten Kreis. Align your latents: High-resolution video synthesis with la- tent diffusion models. In CVPR, pages 22563–22575, 2023. 2

    2023

  5. [5]

    Improving the percep- tual quality of 2D animation interpolation

    Shuhong Chen and Matthias Zwicker. Improving the percep- tual quality of 2D animation interpolation. In ECCV, pages 271–287, 2022. 2

    work page 2022

  6. [6]

    Ldmvfi: Video frame interpolation with latent diffusion models

    Duolikun Danier, Fan Zhang, and David Bull. Ldmvfi: Video frame interpolation with latent diffusion models. In AAAI, pages 1472–1480, 2024. 1, 2, 3

    work page 2024

  7. [7]

    Ex- plorative inbetweening of time and space

    Haiwen Feng, Zheng Ding, Zhihao Xia, Simon Niklaus, Vic- toria Abrevaya, Michael J Black, and Xuaner Zhang. Ex- plorative inbetweening of time and space. In ECCV, pages 378–395, 2024. 1, 2, 3, 5, 6, 7

    work page 2024

  8. [8]

    Cameractrl: Enabling camera control for text-to-video generation

    Hao He, Yinghao Xu, Yuwei Guo, Gordon Wetzstein, Bo Dai, Hongsheng Li, and Ceyuan Yang. Cameractrl: Enabling camera control for text-to-video generation. arXiv preprint arXiv:2404.02101, 2024. 2

    arXiv 2024

Show all 61 references
  1. [9]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In CVPR, pages 770–778, 2016. 4

    2016

  2. [10]

    Gans trained by a two time-scale update rule converge to a local nash equilib- rium

    Martin Heusel, Hubert Ramsauer, Thomas Unterthiner, Bernhard Nessler, and Sepp Hochreiter. Gans trained by a two time-scale update rule converge to a local nash equilib- rium. NIPS, 30, 2017. 5

    2017

  3. [11]

    Denoising dif- fusion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising dif- fusion probabilistic models. NIPS, 33:6840–6851, 2020. 2, 3

    2020

  4. [12]

    Animate anyone: Consistent and controllable image- to-video synthesis for character animation

    Li Hu. Animate anyone: Consistent and controllable image- to-video synthesis for character animation. In CVPR, pages 8153–8163, 2024. 2

    2024

  5. [13]

    Iq-vfi: Implicit quadratic motion estima- tion for video frame interpolation

    Mengshun Hu, Kui Jiang, Zhihang Zhong, Zheng Wang, and Yinqiang Zheng. Iq-vfi: Implicit quadratic motion estima- tion for video frame interpolation. In CVPR, pages 6410– 6419, 2024. 2

    2024

  6. [14]

    Real-time intermediate flow estimation for video frame interpolation

    Zhewei Huang, Tianyuan Zhang, Wen Heng, Boxin Shi, and Shuchang Zhou. Real-time intermediate flow estimation for video frame interpolation. In ECCV, pages 624–642, 2022. 2

    2022

  7. [15]

    Vbench: Comprehensive bench- mark suite for video generative models

    Ziqi Huang, Yinan He, Jiashuo Yu, Fan Zhang, Chenyang Si, Yuming Jiang, Yuanhan Zhang, Tianxing Wu, Qingyang Jin, Nattapol Chanpaisit, et al. Vbench: Comprehensive bench- mark suite for video generative models. In CVPR, pages 21807–21818, 2024. 5

    2024

  8. [16]

    Motion-aware latent diffusion models for video frame interpolation

    Zhilin Huang, Yijie Yu, Ling Yang, Chujun Qin, Bing Zheng, Xiawu Zheng, Zikun Zhou, Yaowei Wang, and Wenming Yang. Motion-aware latent diffusion models for video frame interpolation. arXiv preprint arXiv:2404.13534, 2024. 2, 3

    2024 arXiv

  9. [17]

    Video interpolation with diffusion models

    Siddhant Jain, Daniel Watson, Eric Tabellion, Ben Poole, Janne Kontkanen, et al. Video interpolation with diffusion models. In CVPR, pages 7341–7351, 2024. 2

    2024

  10. [18]

    Super slomo: High quality estimation of multiple intermediate frames for video interpolation

    Huaizu Jiang, Deqing Sun, Varun Jampani, Ming-Hsuan Yang, Erik Learned-Miller, and Jan Kautz. Super slomo: High quality estimation of multiple intermediate frames for video interpolation. In CVPR, pages 9000–9008, 2018. 2

    2018

  11. [19]

    Kingma and Max Welling

    Diederik P. Kingma and Max Welling. Auto-encoding vari- ational bayes. In ICLR, 2014. 3

    2014

  12. [20]

    Ifrnet: Intermediate feature refine network for efficient frame interpolation

    Lingtong Kong, Boyuan Jiang, Donghao Luo, Wenqing Chu, Xiaoming Huang, Ying Tai, Chengjie Wang, and Jie Yang. Ifrnet: Intermediate feature refine network for efficient frame interpolation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , p...

    1969

  13. [21]

    Amt: All-pairs multi-field transforms for efficient frame interpolation

    Zhen Li, Zuo-Liang Zhu, Ling-Hao Han, Qibin Hou, Chun- Le Guo, and Ming-Ming Cheng. Amt: All-pairs multi-field transforms for efficient frame interpolation. In CVPR, pages 9801–9810, 2023. 1, 2

    2023

  14. [22]

    Sparse global matching for video frame interpolation with large motion

    Chunxu Liu, Guozhen Zhang, Rui Zhao, and Limin Wang. Sparse global matching for video frame interpolation with large motion. In CVPR, pages 19125–19134, 2024. 1, 2, 4, 7

    2024

  15. [23]

    Fr ´echet video motion distance: A metric for evaluating motion consistency in videos

    Jiahe Liu, Youran Qu, Qi Yan, Xiaohui Zeng, Lele Wang, and Renjie Liao. Fr ´echet video motion distance: A metric for evaluating motion consistency in videos. arXiv preprint arXiv:2407.16124, 2024. 5

    2024 arXiv

  16. [24]

    Video frame synthesis using deep voxel flow

    Ziwei Liu, Raymond A Yeh, Xiaoou Tang, Yiming Liu, and Aseem Agarwala. Video frame synthesis using deep voxel flow. In ICCV, pages 4463–4471, 2017. 2

    2017

  17. [25]

    T2i-adapter: Learning adapters to dig out more controllable ability for text-to-image diffusion models

    Chong Mou, Xintao Wang, Liangbin Xie, Yanze Wu, Jian Zhang, Zhongang Qi, and Ying Shan. T2i-adapter: Learning adapters to dig out more controllable ability for text-to-image diffusion models. In AAAI, pages 4296–4304, 2024. 2

    2024

  18. [26]

    Softmax splatting for video frame interpolation

    Simon Niklaus and Feng Liu. Softmax splatting for video frame interpolation. In CVPR, pages 5437–5446, 2020. 2

    2020

  19. [27]

    Video frame inter- polation via adaptive separable convolution

    Simon Niklaus, Long Mai, and Feng Liu. Video frame inter- polation via adaptive separable convolution. In ICCV, pages 261–270, 2017. 2

    2017

  20. [28]

    On aliased resizing and surprising subtleties in gan evaluation

    Gaurav Parmar, Richard Zhang, and Jun-Yan Zhu. On aliased resizing and surprising subtleties in gan evaluation. In CVPR, pages 11410–11420, 2022. 5

    2022

  21. [29]

    Gluestick: Robust image matching by stick- ing points and lines together

    R ´emi Pautrat, Iago Su ´arez, Yifan Yu, Marc Pollefeys, and Viktor Larsson. Gluestick: Robust image matching by stick- ing points and lines together. In ICCV, pages 9706–9716,

  22. [30]

    Controlnext: Powerful and effi- cient control for image and video generation

    Bohao Peng, Jian Wang, Yuechen Zhang, Wenbo Li, Ming- Chang Yang, and Jiaya Jia. Controlnext: Powerful and effi- cient control for image and video generation. arXiv preprint arXiv:2408.06070, 2024. 1, 3, 4 9

    2024 arXiv

  23. [31]

    The 2017 davis challenge on video object segmentation.arXiv preprint arXiv:1704.00675, 2017

    Jordi Pont-Tuset, Federico Perazzi, Sergi Caelles, Pablo Ar- bel´aez, Alex Sorkine-Hornung, and Luc Van Gool. The 2017 davis challenge on video object segmentation.arXiv preprint arXiv:1704.00675, 2017. 5

    2017 arXiv

  24. [32]

    Film: Frame interpo- lation for large motion

    Fitsum Reda, Janne Kontkanen, Eric Tabellion, Deqing Sun, Caroline Pantofaru, and Brian Curless. Film: Frame interpo- lation for large motion. In ECCV, pages 250–266, 2022. 2, 5, 6

    2022

  25. [33]

    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 CVPR, pages 10684– 10695, 2022. 2

    2022

  26. [34]

    U-net: Convolutional networks for biomedical image segmentation

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-net: Convolutional networks for biomedical image segmentation. In MICCAI, pages 234–241, 2015. 3

    2015

  27. [35]

    Progressive distillation for fast sampling of diffusion models

    Tim Salimans and Jonathan Ho. Progressive distillation for fast sampling of diffusion models. arXiv preprint arXiv:2202.00512, 2022. 3

    2022 arXiv

  28. [36]

    Bridging the gap: Sketch-aware interpola- tion network for high-quality animation sketch inbetween- ing

    Jiaming Shen, Kun Hu, Wei Bao, Chang Wen Chen, and Zhiyong Wang. Bridging the gap: Sketch-aware interpola- tion network for high-quality animation sketch inbetween- ing. In ACM MM, 2024. 2

    2024

  29. [37]

    Dreammover: Leveraging the prior of diffusion models for image interpolation with large motion

    Liao Shen, Tianqi Liu, Huiqiang Sun, Xinyi Ye, Baopu Li, Jianming Zhang, and Zhiguo Cao. Dreammover: Leveraging the prior of diffusion models for image interpolation with large motion. arXiv preprint arXiv:2409.09605, 2024. 1, 2

    2024 arXiv

  30. [38]

    Dragdiffusion: Harnessing diffusion models for interactive point-based image editing

    Yujun Shi, Chuhui Xue, Jun Hao Liew, Jiachun Pan, Han- shu Yan, Wenqing Zhang, Vincent YF Tan, and Song Bai. Dragdiffusion: Harnessing diffusion models for interactive point-based image editing. In CVPR, pages 8839–8849,

  31. [39]

    Xvfi: ex- treme video frame interpolation

    Hyeonjun Sim, Jihyong Oh, and Munchurl Kim. Xvfi: ex- treme video frame interpolation. In ICCV, pages 14489– 14498, 2021. 2

    2021

  32. [40]

    Deep animation video interpolation in the wild

    Li Siyao, Shiyu Zhao, Weijiang Yu, Wenxiu Sun, Dimitris Metaxas, Chen Change Loy, and Ziwei Liu. Deep animation video interpolation in the wild. In CVPR, pages 6587–6595,

  33. [41]

    Deep geometrized cartoon line inbe- tweening

    Li Siyao, Tianpei Gu, Weiye Xiao, Henghui Ding, Ziwei Liu, and Chen Change Loy. Deep geometrized cartoon line inbe- tweening. In ICCV, pages 7291–7300, 2023. 2

    2023

  34. [42]

    Fvd: A new metric for video generation

    Thomas Unterthiner, Sjoerd van Steenkiste, Karol Kurach, Rapha¨el Marinier, Marcin Michalski, and Sylvain Gelly. Fvd: A new metric for video generation. 2019. 5

    2019

  35. [43]

    Mcvd-masked conditional video diffusion for prediction, generation, and interpolation

    Vikram V oleti, Alexia Jolicoeur-Martineau, and Chris Pal. Mcvd-masked conditional video diffusion for prediction, generation, and interpolation. NIPS, 35:23371–23385, 2022. 2

    2022

  36. [44]

    Generative inbetweening: Adapt- ing image-to-video models for keyframe interpolation

    Xiaojuan Wang, Boyang Zhou, Brian Curless, Ira Kemelmacher-Shlizerman, Aleksander Holynski, and Steven M Seitz. Generative inbetweening: Adapt- ing image-to-video models for keyframe interpolation. arXiv:2408.15239, under review in ICLR , 2025. 1, 2, 3, 5, 6, 7

    2025 arXiv

  37. [45]

    Holistic-motion2d: Scalable whole-body human motion generation in 2d space

    Yuan Wang, Zhao Wang, Junhao Gong, Di Huang, Tong He, Wanli Ouyang, Jile Jiao, Xuetao Feng, Qi Dou, Shix- iang Tang, et al. Holistic-motion2d: Scalable whole-body human motion generation in 2d space. arXiv preprint arXiv:2406.11253, 2024. 8

    2024 arXiv

  38. [46]

    Motionctrl: A unified and flexible motion controller for video generation

    Zhouxia Wang, Ziyang Yuan, Xintao Wang, Yaowei Li, Tianshui Chen, Menghan Xia, Ping Luo, and Ying Shan. Motionctrl: A unified and flexible motion controller for video generation. In ACM SIGGRAPH, pages 1–11, 2024. 2

    2024

  39. [47]

    Perception-oriented video frame interpolation via asymmetric blending

    Guangyang Wu, Xin Tao, Changlin Li, Wenyi Wang, Xiao- hong Liu, and Qingqing Zheng. Perception-oriented video frame interpolation via asymmetric blending. In CVPR, pages 2753–2762, 2024. 2

    2024

  40. [48]

    Toon- crafter: Generative cartoon interpolation

    Jinbo Xing, Hanyuan Liu, Menghan Xia, Yong Zhang, Xintao Wang, Ying Shan, and Tien-Tsin Wong. Toon- crafter: Generative cartoon interpolation. arXiv preprint arXiv:2405.17933, 2024. 2

    2024 arXiv

  41. [49]

    Dynamicrafter: Animating open-domain images with video diffusion priors

    Jinbo Xing, Menghan Xia, Yong Zhang, Haoxin Chen, Wangbo Yu, Hanyuan Liu, Gongye Liu, Xintao Wang, Ying Shan, and Tien-Tsin Wong. Dynamicrafter: Animating open-domain images with video diffusion priors. In ECCV, pages 399–417, 2024. 2, 5, 6, 7

    2024

  42. [50]

    Vibidsam- pler: Enhancing video interpolation using bidirectional dif- fusion sampler

    Serin Yang, Taesung Kwon, and Jong Chul Ye. Vibidsam- pler: Enhancing video interpolation using bidirectional dif- fusion sampler. arXiv:2410.05651, under review in ICLR ,

    arXiv

  43. [51]

    Effec- tive whole-body pose estimation with two-stages distillation

    Zhendong Yang, Ailing Zeng, Chun Yuan, and Yu Li. Effec- tive whole-body pose estimation with two-stages distillation. In ICCV, pages 4210–4220, 2023. 5

    2023

  44. [52]

    Extracting motion and ap- pearance via inter-frame attention for efficient video frame interpolation

    Guozhen Zhang, Yuhan Zhu, Haonan Wang, Youxin Chen, Gangshan Wu, and Limin Wang. Extracting motion and ap- pearance via inter-frame attention for efficient video frame interpolation. In CVPR, pages 5682–5692, 2023. 2

    2023

  45. [53]

    Adding conditional control to text-to-image diffusion models

    Lvmin Zhang, Anyi Rao, and Maneesh Agrawala. Adding conditional control to text-to-image diffusion models. In ICCV, pages 3836–3847, 2023. 2

    2023

  46. [54]

    The unreasonable effectiveness of deep features as a perceptual metric

    Richard Zhang, Phillip Isola, Alexei A Efros, Eli Shechtman, and Oliver Wang. The unreasonable effectiveness of deep features as a perceptual metric. In CVPR, pages 586–595,

  47. [55]

    Mim- icmotion: High-quality human motion video generation with confidence-aware pose guidance

    Yuang Zhang, Jiaxi Gu, Li-Wen Wang, Han Wang, Junqi Cheng, Yuefeng Zhu, and Fangyuan Zou. Mim- icmotion: High-quality human motion video generation with confidence-aware pose guidance. arXiv preprint arXiv:2406.19680, 2024. 2

    2024 arXiv

  48. [56]

    Controlvideo: Training-free controllable text-to-video generation

    Yabo Zhang, Yuxiang Wei, Dongsheng Jiang, Xiaopeng Zhang, Wangmeng Zuo, and Qi Tian. Controlvideo: Training-free controllable text-to-video generation. In ICLR,

  49. [57]

    Clearer frames, anytime: Re- solving velocity ambiguity in video frame interpolation

    Zhihang Zhong, Gurunandan Krishnan, Xiao Sun, Yu Qiao, Sizhuo Ma, and Jian Wang. Clearer frames, anytime: Re- solving velocity ambiguity in video frame interpolation. In ECCV, pages 346–363, 2024. 2

    2024

  50. [58]

    Exploring motion ambiguity and alignment for high-quality video frame interpolation

    Kun Zhou, Wenbo Li, Xiaoguang Han, and Jiangbo Lu. Exploring motion ambiguity and alignment for high-quality video frame interpolation. In CVPR, pages 22169–22179,

  51. [59]

    Stereo magnification: Learning 10 view synthesis using multiplane images

    Tinghui Zhou, Richard Tucker, John Flynn, Graham Fyffe, and Noah Snavely. Stereo magnification: Learning 10 view synthesis using multiplane images. arXiv preprint arXiv:1805.09817, 2018. 5

    2018 arXiv

  52. [60]

    Champ: Controllable and consistent human image animation with 3d parametric guidance

    Shenhao Zhu, Junming Leo Chen, Zuozhuo Dai, Yinghui Xu, Xun Cao, Yao Yao, Hao Zhu, and Siyu Zhu. Champ: Controllable and consistent human image animation with 3d parametric guidance. arXiv preprint arXiv:2403.14781 ,

    arXiv

  53. [61]

    Thin-plate spline-based interpolation for animation line in- betweening

    Tianyi Zhu, Wei Shang, Dongwei Ren, and Wangmeng Zuo. Thin-plate spline-based interpolation for animation line in- betweening. arXiv preprint arXiv:2408.09131, 2024. 2, 4, 7 11

    2024 arXiv

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.