Pith. sign in

REVIEW 5 major objections 4 minor 31 references

FloAt: Flow Warping of Self-Attention for Clothing Animation Generation

T0 review · 5 major / 4 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 A training-free diffusion method animates clothing by warping self-attention along normal-map flow.

desk verdict A plausible training-free recipe for clothing cinemagraphs that warps self-attention with normal-map flow; the core idea is new, but the evaluation overclaims because the metrics reward what the method is built to do. read the letter →

arxiv 2411.15028 v1 pith:AMGDGT4L submitted 2024-11-22 cs.CV

Swasti Shreya Mishra , Kuldeep Kulkarni , Duygu Ceylan , Balaji Vasan Srinivasan This is my paper · ORCID
classification cs.CV
keywords clothinganimationcinemagraphgenerationself-attentionwarpingnormalmapflowopticalControlNettraining-freediffusiontext-conditionedvideo
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 aims to show that clothing cinemagraphs can be generated from a text prompt plus a sequence of surface normal maps without any additional training. Its central move is to compute the optical flow between consecutive normal maps and use it to warp the self-attention map of the previous generated frame, blending the warped map with the current frame's map at each denoising step. The paper argues that this makes garment motion more natural, keeps high-frequency textures such as stripes and prints intact, and suppresses background flicker because the background has zero flow and therefore inherits the first frame's attention. It reports that the method outperforms the considered baselines on normal-map fidelity metrics and in a user study, while noting that standard perceptual metrics are not sensitive to subtle clothing motion.

What carries the argument

The load-bearing mechanism is flow-warped self-attention. After cross-frame self-attention feature injection from an anchor frame and the previous frame, the attention map of the last self-attention layer in the third up-block is blended with the previous frame's attention map warped by RAFT optical flow computed on the normal maps, and then corrected by a binary flow mask. In regions where the normal-map flow is below threshold, the mask is zero, so the corrected attention map is exactly the previous frame's map; by induction, background attention stays identical to frame 0, which the paper identifies as the source of background suppression.

What would settle it

Run FloAtControlNet on a constant normal-map sequence, so the computed flow is zero everywhere. The method's mask-correction step should then force every frame's attention map to equal the first frame's, making the generated video exactly static; if any frame moves or flickers, the background-suppression mechanism is incomplete.

Watch

Extended reading notes

Core claim

In the paper's framing, the discovery is that the animation of a garment is already contained in the flow field of its normal maps, and injecting this flow into the self-attention layers of a normal-map-conditioned diffusion model transfers that motion to the RGB output. For each frame and denoising step, the self-attention map is recomputed as $\widehat{A}^i_t = \alpha A^i_t + (1-\alpha)\,\mathrm{warp}(A^{i-1}_t, f^i_c)$, then corrected with a binary mask derived by thresholding the normal-map flow so that no-flow regions reuse the previous frame's attention. Because background normals do not move, the mask forces background attention to remain equal to that of the first frame, effectively freezing the background. The paper presents this as a training-free way to obtain temporally coherent clothing animation, including high-frequency textured garments where a GAN-based predecessor fails.

Load-bearing premise

The method assumes that the movement field estimated from the normal-map sequence, computed by treating normal maps as ordinary images, faithfully describes how the fabric really moves; if that estimate is wrong, warping attention along it will distort the generated clothing.

Editorial extensions

If this is right

  • If the central claim holds, clothing cinemagraphs can be produced from a text prompt plus a simulated or predicted normal-map sequence without fine-tuning any diffusion model.
  • High-frequency textures such as stripes, leopard print, tie-dye, and lace can be animated naturally, a case where the GAN-based predecessor is reported to fail.
  • Because background attention is pinned to the first frame wherever flow is zero, the method suppresses background flicker at the cost of freezing all zero-flow regions.
  • Standard perceptual metrics such as LPIPS, SSIM, and CLIP under-report quality differences for subtle clothing motion, so normal-map conditioning metrics and user studies become the informative comparisons.

Reading between the lines

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

  • Not stated in the paper: the same flow-warping recipe should transfer to other ControlNet condition modalities whose input sequences carry geometric motion, such as depth or edge maps, giving a general training-free recipe for condition-driven animation.
  • The background-freezing guarantee is conditional on the input normal maps having zero flow in the background; supplying normal maps with camera motion would violate the mask assumption and would test whether the method distinguishes garment motion from scene motion.
  • An extension implicit in the applications discussion: chaining a single-image normal-map predictor with FloAtControlNet would turn a still photograph of a person into a clothing cinemagraph, since the paper already demonstrates that predicted normal-map sequences can drive the method.
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 / 4 minor

Summary. The paper proposes FloAtControlNet, a training-free method for generating clothing cinemagraphs from a text prompt and a sequence of normal maps. The method uses a normal-map-conditioned ControlNet and injects optical flow computed on the normal maps (via RAFT) into the self-attention maps: each frame's self-attention is recomputed as a linear combination of its own attention and the flow-warped attention from the previous frame (Eq. 2), and a binary flow mask forces zero-flow regions to reuse the previous frame's attention (Eqs. 3-4). The authors claim this improves clothing motion naturalness and suppresses background flicker, and they report quantitative comparisons on N-RMSE/N-PSNR/F-RMSE/F-PSNR, Self-LPIPS/SSIM, ablations, and a user study, concluding that the method beats all baselines.

Significance. If the empirical claims are supported, the method is a simple, training-free alternative for controllable clothing animation that exploits normal-map flow to guide self-attention warping. The ablation demonstrating the individual contributions of mask correction and flow warping is useful, and the qualitative results show plausible outputs with reduced background flicker. However, the evaluation relies on metrics that reward the exact properties the method is designed to enforce, the reported gains over the closest baseline are small and lack significance testing, and a key hyperparameter (the mask threshold) is not disclosed. The significance of the claimed superiority is therefore not yet established, though the core idea is coherent and potentially valuable.

major comments (5)
  1. [Section 4.3, User Study] The metrics N-RMSE, N-PSNR, F-RMSE, and F-PSNR are computed as agreement between the input normal maps/flows and normal maps/flows estimated from the generated RGB frames. The method directly injects the input normal-map flow into the self-attention maps (Eq. 2) and uses the flow mask to freeze zero-flow background regions (Eq. 3). These metrics therefore reward exactly the property the method is engineered to maximize; they do not independently establish that the animation is more natural or of higher visual quality. Moreover, the numerical advantage over the closest baseline, Rerender-A-Video Adapt, is small (N-RMSE 19.396 vs. 19.456; F-RMSE 54.887 vs. 56.873) and no error bars or significance tests are provided, even though the 2990 (normal, prompt) pairs are generated from only 230 normal-map sequences and are thus highly correlated. The claim that the method "beats all baselines" is not supported by the experiments as presented.
  2. [Section 4.3] The user study asks participants which generated video's clothing-region animation "best resembles" the input normal-map animation. This is the same fidelity criterion that Eqs. 2-3 optimize, so the study does not assess naturalness or visual quality independently. The study uses only 10 examples, 67 users, and no statistical test is reported; the observed preference counts (41.3% vs. 31.3% and 27.3%) could easily arise from chance. Please add a preference test that asks about plausibility or quality without showing the normal-map input, and report confidence intervals or a significance test.
  3. [Section 3.4] The background-suppression effect is enforced by construction. In background regions where the optical flow is zero, the mask M_fc is zero, so Eq. 3 reduces to A^i_tcor = A^{i-1}_t; by induction, the self-attention in zero-flow regions is frozen across all frames. The reported reduction in background flicker is therefore an expected consequence of the design rather than an empirical discovery. The paper should state this explicitly and frame the qualitative background comparisons as verification of the implementation, not as evidence of a novel capability.
  4. [Section 3.4] The threshold in Eq. 4 is never specified in the paper or the supplementary material, which is a reproducibility gap. Additionally, the mixing weight alpha is selected by an ablation (Figure 6) using the same Self-LPIPS and Self-SSIM metrics that are later used to claim superiority in Table 3, creating a selection-on-the-evaluation-metric concern. Please provide the exact threshold, report sensitivity to both alpha and the threshold, and avoid choosing hyperparameters on the same metrics used for the main comparison.
  5. [Section 5] The paper acknowledges that erroneous normal maps can cause severe imperfections, but it does not quantify robustness to normal-map noise or error. Since the flow is computed on the normal maps and then used to warp attention, the method's core assumption is that the normal-map flow is a faithful proxy for true garment motion. Please include experiments with corrupted or perturbed normal maps to show when the method degrades and to what extent.
minor comments (4)
  1. [Section 4.3] There are inconsistent spellings: "FloAtControlNet" appears as "FloatControlNet" and "Rerender-A-Video" as "Renderer-A-Video" in the same section. Please standardize the method and baseline names throughout.
  2. [Table 2] In Table 2, CycleNet Reshading reports Self-LPIPS of 0.002, which is lower (better) than FloAtControlNet's 0.025. The text explains this as due to static outputs, but the abstract's broad claim that the method "beats all baselines" should be qualified, since CycleNet is not included in the ablation table where "beats all baselines" is claimed for Self-LPIPS/SSIM.
  3. [Algorithm 1] The loop over s = 0 to 319 is presumably the channel dimension of the self-attention output (1x64x64x320), but this is not stated. Please clarify the notation so that the algorithm is self-contained.
  4. [Section 3.4] The binary mask in Eq. 4 uses a hard threshold with no smoothing or dilation, which may create spatial discontinuities in the attention maps. A brief discussion of the threshold choice and its sensitivity would help readers understand the practical behavior.

Circularity Check

3 steps flagged · score 5.0 of 10

Evaluation metrics and user-study question reward the same normal-map flow that Eqs. 2-3 inject, and background suppression is enforced by Eq. 3 by construction; the central method itself is not circular.

  1. self definitional [Section 3.4, 'Background Suppression' (Eq. 3 and following paragraph)]
    "given that Mfc is obtained by thresholding f i c, it is zero in the background regions. Therefore, for the background region, Ai tcor = Ai−1 t , for all i. Thus by induction, we have ˆAi tcor = A0 t for all regions in the background. This ensures that background noise is nearly filtered out in the generated RGB sequence."

    The advertised 'suppression of background artifacts' is not an empirical finding but an algebraic consequence of Eq. 3: any spatial location with sub-threshold flow is forcibly assigned the previous frame's self-attention map, so by induction the zero-flow background attention is constant across the whole sequence. The qualitative and quantitative claims about alleviating background flicker therefore cannot fail at the level of the manipulated attention layer; they are true by the definition of the method.

  2. fitted input called prediction [Section 4.2 'Metrics for Normal Conditioning' vs Section 3.4, Eqs. 2-3]
    "First, we estimate the normal maps from the generated video sequence using an off-the-shelf normal estimator [28] and obtain the root of mean squared error (RMSE), and peak signal-to-noise ratio (PSNR) between the input sequence of normal maps and the estimated sequence of normal maps and call them N-RMSE and N-PSNR respectively. Further, we compute the flow on both the input and estimated normal maps using [24] and use them to compute the RMSE (F-RMSE) and PSNR (F-PSNR) between them."

    Eq. 2 directly injects the input normal-map flow f i c into the self-attention maps, and Eq. 3 uses the mask derived from that same flow to freeze background attention. The F-RMSE/F-PSNR metrics then measure agreement between the flow of the input normal maps (the signal already used for warping) and the flow of normals estimated from the generated RGB frames, while N-RMSE/N-PSNR measure agreement with the same input normal maps. The Table 1 superiority claim is therefore substantially a self-consistency check that rewards the method for reproducing the exact conditioning signal it was engineered from, not an independent test of animation naturalness.

1 more flagged steps
  1. other [Section 4.3 'User Study']
    "In this study, we show the input sequence of normal maps (Input Video) and ask the following question to the user: In which of the following generated videos does the animation in the “clothing region” best resemble the “clothing region” animation shown in the Input Video?"

    The user-study question operationalizes quality as resemblance to the input normal-map animation, which is precisely the signal that Eqs. 2-3 use as the warping target. FloAtControlNet is constructed to transfer that normal-map flow into the self-attention maps, so the question largely measures adherence to the method's own design objective rather than an independent preference for natural or artifact-free animation. The reported 41.3% preference is thus a partially self-referential evaluation of the central claim.

full rationale

The core algorithmic step, flow-warped self-attention in Eq. 2, is a heuristic manipulation and is not itself circular: it could plausibly hurt generation, and the qualitative examples and ablations provide some independent evidence. However, several load-bearing evaluation claims are self-referential. Eq. 3 forces zero-flow background attention to be constant by induction, so the advertised background-flicker suppression is an algebraic property of the method rather than a discovered empirical benefit. The quantitative normal/flow metrics and the user-study question both reward fidelity to the input normal-map sequence, which is the same signal that Eqs. 2-3 inject, so the 'beats all baselines' claims in Table 1 and in the user study are partly measuring design compliance. The alpha hyperparameter is also selected using the same Self-LPIPS/Self-SSIM metrics that are later reported for the final comparison, adding a mild selection-on-evaluation bias. These issues make the evaluation partially circular, but they do not reduce the entire method to its inputs: the flow-warping mechanism itself has independent content and is benchmarked against external baselines. Hence the score is 5 rather than higher.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

The method is heuristic and rests on two modeling assumptions about normal-map flow and self-attention warping. It introduces no new physical entities, but it relies on two hand-chosen parameters, alpha and the mask threshold, that affect the reported metrics.

free parameters (2)
  • alpha (mixing weight) = 0.4
    Linear combination weight in Eq. 2; chosen by the ablation in Fig. 6 that maximizes Self-LPIPS and Self-SSIM, the same metrics used for the final visual-quality comparison.
  • flow threshold for mask M_fc = unspecified
    Eq. 4 defines a binary mask by thresholding the flow, but the threshold value is never reported. It controls which regions are treated as background and is set ad hoc.
assumptions (3)
  • domain assumption RAFT optical flow on normal maps is a valid proxy for garment motion in RGB space.
    Sections 3.3 and 3.4 use normal maps as RGB input to RAFT and treat the resulting flow as the motion that should warp self-attention features. If normals do not track semantic folds and wrinkles, the warp misaligns features.
  • domain assumption Self-attention maps of the last layer of the 3rd ConvUpBlock encode spatial layout that can be warped by bilinear flow.
    Eq. 2 warps these maps; the supplementary ablation shows this specific layer is chosen because earlier layers have no discernible effect. This is an empirical modeling choice, not a derived fact.
  • domain assumption The pretrained ControlNet normal-conditioned generator produces frames consistent with input normal maps.
    Section 3.1 relies on ControlNet's normal map conditioning to synthesize RGB frames; the method adds no training to correct conditioning errors.

how reviews work

0 comments
Cite this review

Pith. "Pith review of FloAt: Flow Warping of Self-Attention for Clothing Animation Generation." pith.science (2026). https://pith.science/paper/AMGDGT4L

@misc{pith2026241115028,
  author       = {Pith},
  title        = {Pith review of: FloAt: Flow Warping of Self-Attention for Clothing Animation Generation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/AMGDGT4L}},
  note         = {Machine review of arXiv:2411.15028}
}
read the original abstract

We propose a diffusion model-based approach, FloAtControlNet to generate cinemagraphs composed of animations of human clothing. We focus on human clothing like dresses, skirts and pants. The input to our model is a text prompt depicting the type of clothing and the texture of clothing like leopard, striped, or plain, and a sequence of normal maps that capture the underlying animation that we desire in the output. The backbone of our method is a normal-map conditioned ControlNet which is operated in a training-free regime. The key observation is that the underlying animation is embedded in the flow of the normal maps. We utilize the flow thus obtained to manipulate the self-attention maps of appropriate layers. Specifically, the self-attention maps of a particular layer and frame are recomputed as a linear combination of itself and the self-attention maps of the same layer and the previous frame, warped by the flow on the normal maps of the two frames. We show that manipulating the self-attention maps greatly enhances the quality of the clothing animation, making it look more natural as well as suppressing the background artifacts. Through extensive experiments, we show that the method proposed beats all baselines both qualitatively in terms of visual results and user study. Specifically, our method is able to alleviate the background flickering that exists in other diffusion model-based baselines that we consider. In addition, we show that our method beats all baselines in terms of RMSE and PSNR computed using the input normal map sequences and the normal map sequences obtained from the output RGB frames. Further, we show that well-established evaluation metrics like LPIPS, SSIM, and CLIP scores that are generally for visual quality are not necessarily suitable for capturing the subtle motions in human clothing animations.

Figures

Figures reproduced from arXiv: 2411.15028 by the authors.

Figure 1
Figure 1. We introduce a method for human clothing generation given a text prompt and a sequence of normal maps [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Overview of FloAtControlNet. Given a text prompt and an input sequence of normal maps, we first compute the flow over the sequence of normal maps using the RAFT [24] model and threshold it as mentioned to get a binary mask. We sequentially input the normal maps and text prompt into the normal-conditioned ControlNet [1]. Next, we perform self-attention feature injection inspired by Pix2Video [8] to ensure temporal co… view at source ↗
Figure 3
Figure 3. Self-attention visualisation. For a given FeatInControlNet generated sequence, we take the first PCA component of the self-attention map for frames 2, 3 and 4, at the last layer of the 3rd ConvUpBlock of the U-Net and plot its heatmap at the final denoising step. We mark 3 spatial regions of the self-attention maps, depicted as dotted bounding boxes (1, 2) and an ellipse (3) and its corresponding region in the gener… view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Qualitative results for our different methods and Rerender-A-Video Adapt [ [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]
Figure 5
Figure 5. Figure 5: Qualitative results for our different methods and CycleNet Reshading [ [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]
Figure 6
Figure 6. Figure 6: We report the min-max normalized inverse Self-LPIPS and Self-SSIM scores with the specified [PITH_FULL_IMAGE:figures/full_fig_p011_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

31 extracted references · 11 canonical work pages

  1. [1]

    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 Proceedings of the IEEE/CVF International Conference on Computer Vision , pages 3836–3847, 2023

    2023

  2. [2]

    Curless, Steven M

    Aleksander Holynski, Brian L. Curless, Steven M. Seitz, and Richard Szeliski. Animating pictures with eulerian motion fields. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pages 5810–5819, June 2021

    work page 2021

  3. [3]

    Controllable animation of fluid elements in still images

    Aniruddha Mahapatra and Kuldeep Kulkarni. Controllable animation of fluid elements in still images. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 3667–3676, 2022

    work page 2022

  4. [4]

    Text-Guided Synthesis of Eulerian Cinemagraphs

    Aniruddha Mahapatra, Aliaksandr Siarohin, Hsin-Ying Lee, Sergey Tulyakov, and Jun-Yan Zhu. Synthesizing artistic cinemagraphs from text. arXiv preprint arXiv:2307.03190, 2023

    work page Pith review arXiv 2023

  5. [5]

    Generative image dynamics.arXiv preprint arXiv:2309.07906, 2023

    Zhengqi Li, Richard Tucker, Noah Snavely, and Aleksander Holynski. Generative image dynamics.arXiv preprint arXiv:2309.07906, 2023

    arXiv 2023

  6. [6]

    Endless loops: detecting and animating periodic patterns in still images

    Tavi Halperin, Hanit Hakim, Orestis Vantzos, Gershon Hochman, Netai Benaim, Lior Sassy, Michael Kupchik, Ofir Bibi, and Ohad Fried. Endless loops: detecting and animating periodic patterns in still images. ACM Transactions on Graphics (TOG), 40(4):1–12, 2021

    work page 2021

  7. [7]

    Blowing in the wind: Cyclenet for human cinemagraphs from still images

    Hugo Bertiche, Niloy J Mitra, Kuldeep Kulkarni, Chun-Hao P Huang, Tuanfeng Y Wang, Meysam Madadi, Sergio Escalera, and Duygu Ceylan. Blowing in the wind: Cyclenet for human cinemagraphs from still images. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 459–468, 2023

    work page 2023

  8. [8]

    Pix2video: Video editing using image diffusion

    Duygu Ceylan, Chun-Hao P Huang, and Niloy J Mitra. Pix2video: Video editing using image diffusion. In Proceedings of the IEEE/CVF International Conference on Computer Vision , pages 23206–23217, 2023

    2023

Show all 31 references
  1. [9]

    Latentwarp: Consistent diffusion latents for zero-shot video-to-video translation

    Yuxiang Bao, Di Qiu, Guoliang Kang, Baochang Zhang, Bo Jin, Kaiye Wang, and Pengfei Yan. Latentwarp: Consistent diffusion latents for zero-shot video-to-video translation. arXiv preprint arXiv:2311.00353, 2023. 14 A PREPRINT - NOVEMBER 25, 2024

    2023 arXiv

  2. [10]

    Generative adversarial nets

    Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. Generative adversarial nets. Advances in neural information processing systems , 27, 2014

    2014

  3. [11]

    Simulating fluids in real-world still images

    Siming Fan, Jingtan Piao, Chen Qian, Hongsheng Li, and Kwan-Yee Lin. Simulating fluids in real-world still images. In Proceedings of the IEEE/CVF International Conference on Computer Vision , pages 15922–15931, 2023

    2023

  4. [12]

    Animating pictures with stochastic motion textures

    Yung-Yu Chuang, Dan B Goldman, Ke Colin Zheng, Brian Curless, David H Salesin, and Richard Szeliski. Animating pictures with stochastic motion textures. In ACM SIGGRAPH 2005 Papers, pages 853–860. 2005

    2005

  5. [13]

    Denoising diffusion probabilistic models

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

    2020

  6. [14]

    Denoising diffusion implicit models

    Jiaming Song, Chenlin Meng, and Stefano Ermon. Denoising diffusion implicit models. arXiv preprint arXiv:2010.02502, 2020

    2010 arXiv

  7. [15]

    High-resolution image synthesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Björn Ommer. High-resolution image synthesis with latent diffusion models. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 10684–10695, 2022

    2022

  8. [16]

    Tune-a-video: One-shot tuning of image diffusion models for text-to-video generation

    Jay Zhangjie Wu, Yixiao Ge, Xintao Wang, Weixian Lei, Yuchao Gu, Wynne Hsu, Ying Shan, Xiaohu Qie, and Mike Zheng Shou. Tune-a-video: One-shot tuning of image diffusion models for text-to-video generation. arXiv e-prints, pages arXiv–2212, 2022

    2022

  9. [17]

    Rerender a video: Zero-shot text-guided video-to-video translation

    Shuai Yang, Yifan Zhou, Ziwei Liu, and Chen Change Loy. Rerender a video: Zero-shot text-guided video-to-video translation. arXiv preprint arXiv:2306.07954, 2023

    2023 arXiv

  10. [18]

    Video-p2p: Video editing with cross-attention control

    Shaoteng Liu, Yuechen Zhang, Wenbo Li, Zhe Lin, and Jiaya Jia. Video-p2p: Video editing with cross-attention control. arXiv preprint arXiv:2303.04761, 2023

    2023 arXiv

  11. [19]

    Fatezero: Fusing attentions for zero-shot text-based video editing

    Chenyang Qi, Xiaodong Cun, Yong Zhang, Chenyang Lei, Xintao Wang, Ying Shan, and Qifeng Chen. Fatezero: Fusing attentions for zero-shot text-based video editing. arXiv preprint arXiv:2303.09535, 2023

    2023 arXiv

  12. [20]

    Text2video-zero: Text-to-image diffusion models are zero-shot video generators

    Levon Khachatryan, Andranik Movsisyan, Vahram Tadevosyan, Roberto Henschel, Zhangyang Wang, Shant Navasardyan, and Humphrey Shi. Text2video-zero: Text-to-image diffusion models are zero-shot video generators. arXiv preprint arXiv:2303.13439, 2023

    2023 arXiv

  13. [21]

    Tokenflow: Consistent diffusion features for consistent video editing

    Michal Geyer, Omer Bar-Tal, Shai Bagon, and Tali Dekel. Tokenflow: Consistent diffusion features for consistent video editing. arXiv preprint arXiv:2307.10373, 2023

    2023 arXiv

  14. [22]

    Flatten: optical flow-guided attention for consistent text-to-video editing

    Yuren Cong, Mengmeng Xu, Christian Simon, Shoufa Chen, Jiawei Ren, Yanping Xie, Juan-Manuel Perez-Rua, Bodo Rosenhahn, Tao Xiang, and Sen He. Flatten: optical flow-guided attention for consistent text-to-video editing. arXiv preprint arXiv:2310.05922, 2023

    arXiv

  15. [23]

    Lamp: Learn a motion pattern for few-shot-based video generation

    Ruiqi Wu, Liangyu Chen, Tong Yang, Chunle Guo, Chongyi Li, and Xiangyu Zhang. Lamp: Learn a motion pattern for few-shot-based video generation. arXiv preprint arXiv:2310.10769, 2023

    2023 arXiv

  16. [24]

    Raft: Recurrent all-pairs field transforms for optical flow

    Zachary Teed and Jia Deng. Raft: Recurrent all-pairs field transforms for optical flow. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part II 16 , pages 402–419. Springer, 2020

    2020

  17. [25]

    Principal components analysis (pca)

    Andrzej Ma´ckiewicz and Waldemar Ratajczak. Principal components analysis (pca). Computers & Geosciences, 19(3):303–342, 1993

    1993

  18. [26]

    Cloth3d: clothed 3d humans

    Hugo Bertiche, Meysam Madadi, and Sergio Escalera. Cloth3d: clothed 3d humans. In European Conference on Computer Vision, pages 344–359. Springer, 2020

    2020

  19. [27]

    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. arXiv preprint arXiv:2305.13077, 2023

    2023 arXiv

  20. [28]

    Estimating and exploiting the aleatoric uncertainty in surface normal estimation

    Gwangbin Bae, Ignas Budvytis, and Roberto Cipolla. Estimating and exploiting the aleatoric uncertainty in surface normal estimation. In Proceedings of the IEEE/CVF International Conference on Computer Vision , pages 13137–13146, 2021

    2021

  21. [29]

    Image quality assessment: from error visibility to structural similarity

    Zhou Wang, Alan C Bovik, Hamid R Sheikh, and Eero P Simoncelli. Image quality assessment: from error visibility to structural similarity. IEEE transactions on image processing, 13(4):600–612, 2004

    2004

  22. [30]

    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 Proceedings of the IEEE conference on computer vision and pattern recognition, pages 586–595, 2018

    2018

  23. [31]

    Imagic: Text-based real image editing with diffusion models

    Bahjat Kawar, Shiran Zada, Oran Lang, Omer Tov, Huiwen Chang, Tali Dekel, Inbar Mosseri, and Michal Irani. Imagic: Text-based real image editing with diffusion models. In Conference on Computer Vision and Pattern Recognition 2023, 2023. 15

    2023

Pith tools

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