Pith. sign in

REVIEW 5 major objections 4 minor 54 references

IP-FaceDiff: Identity-Preserving Facial Video Editing with Diffusion

T0 review · 5 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read This paper claims that facial video editing can preserve a person's identity and follow text prompts by injecting attention features from an identity-fine-tuned diffusion branch into an edit-fine-tuned branch, while cutting editing time…

desk verdict A plausible engineering pipeline for facial video editing with a real speed gain, but the identity-preservation table is internally implausible and the central claim lacks credible support. read the letter →

arxiv 2501.07530 v1 pith:P37HP3BF submitted 2025-01-13 cs.CV

Tharun Anand , Aryan Garg , Kaushik Mitra This is my paper · ORCID
classification cs.CV
keywords facialvideoeditingdiffusionmodelsidentitypreservationtext-guidedself-attentionfeatureinjectionkeyframedirectionalCLIPlossArcFace
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 a facial video editing framework that claims to preserve the subject's identity while following arbitrary text prompts, and to cut editing time by roughly 80%. It does so by fine-tuning two copies of a pre-trained text-to-image diffusion model: one copy learns to reconstruct faces faithfully using an ArcFace identity loss, the other learns localized, prompt-aligned edits using a directional CLIP loss. At edit time, self-attention features from the identity branch are injected into the editing branch, and edits are applied to keyframes and propagated to the rest of the video. The paper reports that this approach beats three established baselines on identity-retrieval metrics, temporal consistency, and human opinion scores, and works on in-the-wild videos.

What carries the argument

The load-bearing mechanism is cross-branch self-attention feature substitution during editing. Two U-Nets share the same pre-trained Stable Diffusion initialization but are fine-tuned independently: ε1 with an identity loss based on ArcFace, ε2 with a directional CLIP loss. For each frame, DDIM inversion produces noisy latents; ε1 generates self-attention features at every layer, and during editing these features replace the self-attention features of ε2 at corresponding layers. The edit direction therefore comes from the CLIP-fine-tuned branch while identity and head pose come from the ArcFace-fine-tuned branch. The same feature space also supports joint keyframe editing: only a subset of frames is edited, and the edits are propagated to the other frames via nearest-neighbor matching in the latent feature space, which provides temporal consistency and the reported speedup.

What would settle it

Take a held-out set of face videos, run the method, and at each substituted layer compute a distribution-shift metric (such as cosine distance or a feature-space distance) between the self-attention features produced by ε1 and ε2 for the same inverted latents. If the shift is large at layers that are nonetheless swapped, or if ablating the ArcFace fine-tuning of ε1 does not measurably reduce identity preservation, the central mechanism would be falsified.

Watch

Extended reading notes

Core claim

The central claim is that identity preservation and text-driven edit quality can be decoupled into two independently fine-tuned diffusion branches and recombined at inference without retraining per video. The identity branch ε1 is fine-tuned on face images to minimize both pixel reconstruction error and ArcFace cosine distance to the ground-truth identity; the editing branch ε2 is fine-tuned with a directional CLIP loss to make localized changes that follow the prompt. During editing, the self-attention features that ε1 computes from DDIM-inverted frames are substituted into the corresponding layers of ε2, so the editing branch receives the prompt direction while the identity branch supplies global facial structure. Joint keyframe editing propagates the edit across frames, yielding temporal consistency and an ~80% reduction in inference time compared with frame-wise baselines. The paper claims this consistently outperforms prior facial video editors across identity-retrieval metrics (CosFace, VGGFace, FaceNet), temporal loss, and mean opinion scores.

Load-bearing premise

The identity-preservation mechanism assumes that self-attention features computed by the ArcFace-fine-tuned branch stay compatible with the editing branch at every layer where they are injected, even though the two branches were fine-tuned independently; if their feature distributions drift apart, the swap could corrupt the edit or fail to hold identity.

Editorial extensions

If this is right

  • Facial video editing could become a single-pass, text-driven operation on one GPU, with no per-video optimization.
  • Identity preservation would transfer to unseen faces and in-the-wild footage because the identity branch is fixed after fine-tuning.
  • Multiple edits (for example, adding glasses, bangs, and a beard) could be applied in one forward pass without extra computational cost.
  • The two-branch attention-injection design may extend to other subject-critical video edits, such as preserving an object's or a body's consistency.

Reading between the lines

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

  • A natural extension would be to measure per-layer distribution shift between the two branches' self-attention features on held-out data; layers with large drift could be excluded or reweighted rather than blindly swapped.
  • The same decoupling may transfer to newer backbones such as SDXL or video diffusion models, provided both branches share architecture and the compatibility assumption holds.
  • The reported 80% speedup depends on how many keyframes are sampled; a systematic sweep of keyframe density against identity and temporal quality would reveal the practical operating range.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 4 minor

Summary. The paper proposes IP-FaceDiff, a facial video editing framework built on two fine-tuned Stable Diffusion 2.1 branches: one fine-tuned with a reconstruction plus ArcFace identity loss (ε1), and one fine-tuned with a directional CLIP loss for localized edits (ε2). During inference, the method performs DDIM inversion with ε1, extracts self-attention features, and injects them into ε2's self-attention layers while editing keyframes and propagating via the TokenFlow correspondence mechanism. The authors claim consistent identity preservation, faster inference (roughly 80% reduction), temporal consistency, and multi-attribute editing, and report quantitative comparisons against DVA, STIT, and LTFE.

Significance. If the empirical claims were reliable, the contribution would be valuable: leveraging pretrained T2I diffusion models for facial video editing with identity preservation and lower inference cost is a practically important goal. The proposed feature-injection idea is well motivated by PnP-style editing, and the paper includes a useful ablation of the two fine-tuning losses. However, the quantitative evidence—especially the identity-preservation table—is not credible as reported, and the evaluation is too thin (25 videos, no error bars, missing a directly relevant baseline) to support the abstract's claims of consistent superiority.

major comments (5)
  1. [Section 5.3, Table 1] The R@1 and MRR values for DVA and Ours (with ID guidance) are numerically identical across all three face-recognition backbones (VGGFace, CosFace, FaceNet): DVA reports 0.76/0.794 in all three columns and Ours reports 0.96/0.97 in all three columns. Because each backbone produces different embeddings, exact agreement across backbones is extremely unlikely on a 35,000-video retrieval database, which raises a serious question about whether three independent evaluations were actually performed. Since Table 1 is the main quantitative support for the paper's central identity-preservation claim, this is a load-bearing issue that must be resolved.
  2. [Section 5.2, Tables 1–3] The evaluation uses only 25 edited videos per method (from 50 original videos), with no error bars, confidence intervals, or significance tests. The reported improvements, such as the FaceNet cosine distance of 0.221 for Ours versus 0.256 for DVA, could be within noise at this sample size; the paper should provide variance estimates and a statistical comparison. Additionally, Section 5.5 and Table 3 report results for '20 ten-second videos', while Section 5.2 says the evaluation dataset is 25 edited videos; this inconsistency needs to be corrected.
  3. [Section 4.2, Eq. (5), Table 1] The identity-preservation mechanism is trained with an ArcFace cosine-distance loss (Eq. 5) and evaluated with cosine distances and retrieval metrics computed from face-recognition embeddings (Table 1). Although the evaluation uses CosFace, VGGFace, and FaceNet rather than ArcFace itself, all are face-recognition embedding spaces of the same family, so the reported identity scores are partially a fit to the training objective. The authors should discuss this circularity and include a human perceptual identity-judgment study or a metric not derived from face-recognition embeddings.
  4. [Section 4.1, Fig. 2] The central novelty is substituting ε1's self-attention features into ε2 during editing, but the manuscript provides no analysis of whether the independently fine-tuned branches' feature distributions remain aligned. If the distributions drift, the swap could either corrupt the edit or fail to transfer identity. At minimum, the paper should show a comparative visualization or a quantitative test (e.g., feature-distance histograms) demonstrating that the injected features are compatible at the selected layers.
  5. [Section 5.2, 5.9] The method is built directly on TokenFlow (Geyer et al. [14]), which is the relevant state-of-the-art for consistent video editing with T2I models, yet TokenFlow is neither included as a quantitative baseline nor evaluated on the facial-video benchmark. The paper states in Section 4.2 that TokenFlow 'suffers from this issue' but provides no numbers; adding this baseline is necessary to support the claim of outperforming existing techniques.
minor comments (4)
  1. [Throughout] Several typos and missing spaces appear, such as 'DV A' in Table 1 entries, 'modelϵ1' in Section 4.1, and 'networkϵ2' in Section 4.1; these should be corrected for readability.
  2. [Section 5.8] The text refers to 'pre-trained SDE models', but Stable Diffusion is a latent diffusion model, not an SDE in the sense used here; the terminology should be consistent ('pre-trained T2I models').
  3. [Section 5.5, Table 3] The MOS is reported on a 1–10 scale, but the caption does not explicitly state the scale, and no per-video variance or inter-rater agreement is provided; please clarify and add these details.
  4. [Section 5.1] The hyperparameters λ1=0.3 and λ2=0.7 are listed, but no sensitivity analysis or justification for these values is provided; the free parameters of the loss and training (30,000 iterations, batch size 16) are not ablated.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the method's identity preservation is a trained objective evaluated on held-out videos with external face-recognition embeddings.

full rationale

The paper does not derive any claimed result from an equivalent input. The identity-preservation branch epsilon_1 is fine-tuned with Eq. 5, which includes an ArcFace cosine-distance term, and identity preservation is then measured with cosine distance, R@1, and MRR from VGGFace, CosFace, and FaceNet on held-out CelebV-HQ videos. Although the training loss and evaluation metric belong to the same face-recognition cosine-distance family, the evaluation uses different embedding models and a different, held-out video dataset, so the reported identity preservation is a generalization result rather than the training objective renamed as a prediction. The keyframe propagation and self-attention substitution are explicitly borrowed from external works (TokenFlow, PNP), not from a self-citation chain, and the timing and temporal-consistency comparisons are against external baselines with independent measurements. The suspicious identical R@1/MRR values across backbones in Table 1 are a data-integrity and verification concern, not a circularity, and therefore do not change the circularity score. I find no load-bearing step in which a quantity is defined in terms of the target result or a fitted parameter is presented as a prediction.

Assumptions & free parameters 4 free parameters · 6 assumptions · 0 invented entities

The central claim rests on several borrowed assumptions, most notably cross-branch feature compatibility and the use of face-recognition embeddings both as a training signal and as an evaluator. The free parameters are mostly hand-chosen loss weights and training durations; no new physical or theoretical entities are introduced.

free parameters (4)
  • lambda_1 = 0.3
    Weight on the reconstruction term in the directional CLIP loss (Eq. 6), chosen by hand with no sensitivity analysis.
  • lambda_2 = 0.7
    Weight on the directional CLIP term in Eq. 6, chosen by hand with no sensitivity analysis.
  • identity_loss_weight = 1.0 (implicit)
    Both terms in the identity-preservation loss (Eq. 5) are summed without explicit weighting; the effective weight is 1.0 for L1 and ArcFace cosine distance.
  • fine-tuning iterations = 30,000
    Training duration for both epsilon_1 and epsilon_2; chosen empirically, no ablation on iteration count.
assumptions (6)
  • domain assumption Self-attention feature substitution from a separately fine-tuned branch preserves identity without retraining the combined system.
    Invoked in Sec 4.1 and Fig. 2; no evidence that the feature spaces of epsilon_1 and epsilon_2 remain aligned after independent fine-tuning.
  • domain assumption ArcFace embeddings are a valid identity measure for both training and evaluation.
    Used in Eq. 5 for training and indirectly validated by the retrieval metrics in Table 1; face-recognition embeddings are standard but not an exact proxy for perceived identity.
  • domain assumption Directional CLIP loss produces localized facial edits.
    Borrowed from StyleGAN-NADA and applied here to a diffusion U-Net without demonstrating locality beyond qualitative examples.
  • domain assumption CelebA-HQ fine-tuning transfers to in-the-wild video frames with pose and motion.
    Sec 5.1 fine-tunes on static 512x512 images; generalization to videos is claimed qualitatively in Sec 5.9.
  • domain assumption RAFT-based warping loss measures temporal consistency in edited videos.
    Eq. 7 assumes optical-flow warping residuals capture flicker; baselines may be penalized differently if their edits change motion in a way that is not a temporal artifact.
  • standard math DDIM inversion of each frame independently produces latents suitable for keyframe editing and propagation.
    Adopted from TokenFlow [14]; not re-derived or validated on this data.

how reviews work

0 comments
Cite this review

Pith. "Pith review of IP-FaceDiff: Identity-Preserving Facial Video Editing with Diffusion." pith.science (2026). https://pith.science/paper/P37HP3BF

@misc{pith2026250107530,
  author       = {Pith},
  title        = {Pith review of: IP-FaceDiff: Identity-Preserving Facial Video Editing with Diffusion},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/P37HP3BF}},
  note         = {Machine review of arXiv:2501.07530}
}
read the original abstract

Facial video editing has become increasingly important for content creators, enabling the manipulation of facial expressions and attributes. However, existing models encounter challenges such as poor editing quality, high computational costs and difficulties in preserving facial identity across diverse edits. Additionally, these models are often constrained to editing predefined facial attributes, limiting their flexibility to diverse editing prompts. To address these challenges, we propose a novel facial video editing framework that leverages the rich latent space of pre-trained text-to-image (T2I) diffusion models and fine-tune them specifically for facial video editing tasks. Our approach introduces a targeted fine-tuning scheme that enables high quality, localized, text-driven edits while ensuring identity preservation across video frames. Additionally, by using pre-trained T2I models during inference, our approach significantly reduces editing time by 80%, while maintaining temporal consistency throughout the video sequence. We evaluate the effectiveness of our approach through extensive testing across a wide range of challenging scenarios, including varying head poses, complex action sequences, and diverse facial expressions. Our method consistently outperforms existing techniques, demonstrating superior performance across a broad set of metrics and benchmarks.

Figures

Figures reproduced from arXiv: 2501.07530 by the authors.

Figure 1
Figure 1. Qualitative Facial Attribute Editing. Our method enables precise, localized facial edits whilst maintaining low computational latency. Through explicit identity preservation optimization, we ensure that global facial identity features are retained. Furthermore, our approach ensures temporal consistency and robust generalization across a wide range of facial video editing tasks. Abstract Facial video editing has beco… view at source ↗
Figure 2
Figure 2. Model Architecture. Left: Pre-trained T2I models ϵ1 and ϵ2 are fine-tuned independently with ArcFace loss and directional CLIP loss for identity-preservation and prompt-adhering localization, respectively. Right: Video frames are inverted with DDIM and then processed through ϵ1 to extract self-attention features at each timestep. Text-guided editing is applied to keyframes using ϵ2, guided by identity features from … view at source ↗
Figure 3
Figure 3. Strong Prompt-Adhering Multiple Editing. Left: Beyond local edits, our method manipulates facial expressions and age. Right: Our facial video editing method handles simultaneous edits to local facial features. dataset using captions generated by BLIP-2 [28] as refer￾ence prompts. These were paired with extensive set of edit￾ing prompts that covered both facial attributes (e.g., beard, glasses) and facial features (e… view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Editing Faces in the Wild. We successfully overcome a previous hurdle of out-of-domain adaptation for facial video editing methods. Recall at Rank 1 (R@1): Measures the proportion of queries where the original embedding is ranked first based on average Euclidean distan…
Figure 5
Figure 5. Figure 5: Ablation: fine-tuning ϵ1 and ϵ2 with Arc-Face and directional-Clip Loss for identity preservation and performing localized edits in facial videos. our method inherently enforces frame-to-frame coherence at the latent representation level. This helps in reducing per￾cep…
Figure 6
Figure 6. Figure 6: More Identity Preserving Localized Editing in the Wild. Left: Hair color change and accessory addition are performed from a randomly scraped online music video. Right: Facial hair and facial expressions of a consenting volunteer are edited using our method. This emphas…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

54 extracted references · 49 canonical work pages

  1. [14]

    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. In International Conference on Learning Represen- tations, 2023. 2, 3, 4, 6

    work page 2023

  2. [1]

    Im- age2stylegan: How to embed images into the stylegan latent space? 2019 IEEE/CVF International Conference on Com- puter Vision (ICCV), pages 4431–4440, 2019

    Rameen Abdal, Yipeng Qin, and Peter Wonka. Im- age2stylegan: How to embed images into the stylegan latent space? 2019 IEEE/CVF International Conference on Com- puter Vision (ICCV), pages 4431–4440, 2019. 2

    work page 2019

  3. [2]

    Im- age2stylegan++: How to edit the embedded images? In 2020 IEEE/CVF Conference on Computer Vision and Pat- tern Recognition (CVPR), pages 8293–8302, 2020

    Rameen Abdal, Yipeng Qin, and Peter Wonka. Im- age2stylegan++: How to edit the embedded images? In 2020 IEEE/CVF Conference on Computer Vision and Pat- tern Recognition (CVPR), pages 8293–8302, 2020. 2

    work page 2020

  4. [3]

    Seman- tic photo manipulation with a generative image prior

    David Bau, Hendrik Strobelt, William Peebles, Jonas Wulff, Bolei Zhou, Jun-Yan Zhu, and Antonio Torralba. Seman- tic photo manipulation with a generative image prior. ACM Transactions on Graphics, 38(4):1–11, July 2019. 2

    work page 2019

  5. [4]

    Tim Brooks, Aleksander Holynski, and Alexei A. Efros. In- structpix2pix: Learning to follow image editing instructions. 2023 IEEE/CVF Conference on Computer Vision and Pat- tern Recognition (CVPR), pages 18392–18402, 2022. 2

    work page 2023

  6. [5]

    Pix2video: Video editing using image diffusion

    Duygu Ceylan, Chun-Hao Paul Huang, and Niloy Jyoti Mi- tra. Pix2video: Video editing using image diffusion. 2023 IEEE/CVF International Conference on Computer Vision (ICCV), pages 23149–23160, 2023. 2

    work page 2023

  7. [6]

    Stablevideo: Text-driven consistency-aware diffusion video editing

    Wenhao Chai, Xun Guo, Gaoang Wang, and Yang Lu. Stablevideo: Text-driven consistency-aware diffusion video editing. 2023 IEEE/CVF International Conference on Com- puter Vision (ICCV), pages 22983–22993, 2023. 2

    work page 2023

  8. [7]

    Attend-and-excite: Attention-based se- mantic guidance for text-to-image diffusion models

    Hila Chefer, Yuval Alaluf, Yael Vinker, Lior Wolf, and Daniel Cohen-Or. Attend-and-excite: Attention-based se- mantic guidance for text-to-image diffusion models. ACM Transactions on Graphics (TOG), 42:1 – 10, 2023. 1

    work page 2023

Show all 54 references
  1. [8]

    Celebv- hq: A large-scale video facial attributes dataset

    Hao clip, Wayne Wu, Wentao clip, Liming Jiang, Siwei Tang, Li Zhang, Ziwei Liu, and Chen Change Loy. Celebv- hq: A large-scale video facial attributes dataset. In Com- puter Vision – ECCV 2022: 17th European Conference, Tel Aviv, Israel, October 23–27, 2022, Proceedings, Part ...

    2022

  2. [9]

    Arcface: Additive angular mar- gin loss for deep face recognition.IEEE Transactions on Pat- tern Analysis and Machine Intelligence , 44(10):5962–5979, Oct

    Jiankang Deng, Jia Guo, Jing Yang, Niannan Xue, Irene Kot- sia, and Stefanos Zafeiriou. Arcface: Additive angular mar- gin loss for deep face recognition.IEEE Transactions on Pat- tern Analysis and Machine Intelligence , 44(10):5962–5979, Oct. 2022. 2, 4, 6

    2022

  3. [10]

    Diffusion models beat gans on image synthesis

    Prafulla Dhariwal and Alexander Nichol. Diffusion models beat gans on image synthesis. In M. Ranzato, A. Beygelz- imer, Y . Dauphin, P.S. Liang, and J. Wortman Vaughan, ed- itors, Advances in Neural Information Processing Systems , volume 34, pages 8780–8794. Curran Associates...

    2021

  4. [11]

    Make-a-scene: Scene- based text-to-image generation with human priors

    Oran Gafni, Adam Polyak, Oron Ashual, Shelly Sheynin, Devi Parikh, and Yaniv Taigman. Make-a-scene: Scene- based text-to-image generation with human priors. In Computer Vision – ECCV 2022: 17th European Conference, Tel Aviv, Israel, October 23–27, 2022, Proceedings, Part ...

    2022

  5. [12]

    Bermano, Gal Chechik, and Daniel Cohen-Or

    Rinon Gal, Yuval Alaluf, Yuval Atzmon, Or Patashnik, Amit H. Bermano, Gal Chechik, and Daniel Cohen-Or. An image is worth one word: Personalizing text-to-image gen- eration using textual inversion. In International Conference on Learning Representations, 2022. 2

    2022

  6. [13]

    Bermano, Gal Chechik, and Daniel Cohen-Or

    Rinon Gal, Or Patashnik, Haggai Maron, Amit H. Bermano, Gal Chechik, and Daniel Cohen-Or. Stylegan-nada. ACM Transactions on Graphics (TOG), 41:1 – 13, 2021. 2, 4

    2021

  7. [15]

    Conditional generation from pre-trained diffusion models using denoiser representations

    Alexandros Graikos, Srikar Yellapragada, and Dimitris Samaras. Conditional generation from pre-trained diffusion models using denoiser representations. In 34th British Ma- chine Vision Conference 2023, BMVC 2023, Aberdeen, UK, November 20-24, 2023. BMV A, 2023. 2

    2023

  8. [16]

    Image process- ing using multi-code gan prior

    Jinjin Gu, Yujun Shen, and Bolei Zhou. Image process- ing using multi-code gan prior. 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 3009–3018, 2019. 2

    2020

  9. [17]

    Prompt-to-prompt image editing with cross attention control

    Amir Hertz, Ron Mokady, Jay Tenenbaum, Kfir Aberman, Yael Pritch, and Daniel Cohen-Or. Prompt-to-prompt image editing with cross attention control. In International Confer- ence on Learning Representations, 2022. 2

    2022

  10. [18]

    Denoising diffu- sion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffu- sion probabilistic models. Advances in Neural Information Processing Systems, 33:6840–6851, 2020. 1, 3

    2020

  11. [19]

    Classifier-free diffusion guidance, 2022

    Jonathan Ho and Tim Salimans. Classifier-free diffusion guidance, 2022. 1

    2022

  12. [20]

    LoRA: Low-rank adaptation of large language models

    Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen- clip, Yuanzhi Li, Shean Wang, Lu Wang, and Weiclip Chen. LoRA: Low-rank adaptation of large language models. InIn- ternational Conference on Learning Representations , 2022. 2

    2022

  13. [21]

    Text2scene: Text-driven indoor scene stylization with part- aware details

    Inwoo Hwang, Hyeonwoo Kim, and Young Min Kim. Text2scene: Text-driven indoor scene stylization with part- aware details. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 1890–1899, June 2023. 2

    2023

  14. [22]

    Spatial transformer networks

    Max Jaderberg, Karen Simonyan, Andrew Zisserman, and koray kavukcuoglu. Spatial transformer networks. In C. Cortes, N. Lawrence, D. Lee, M. Sugiyama, and R. Garnett, editors, Advances in Neural Information Processing Systems, volume 28. Curran Associates, Inc., 2015. 7

    2015

  15. [23]

    Identity-aware and shape-aware propagation of face editing in videos

    Yue-Ren Jiang, Shu-Yu Chen, Hongbo Fu, and Lin Gao. Identity-aware and shape-aware propagation of face editing in videos. IEEE Transactions on Visualization and Computer Graphics, 30(7):3444–3456, 2024. 5

    2024

  16. [24]

    A style-based generator architecture for generative adversarial networks

    Tero Karras, Samuli Laine, and Timo Aila. A style-based generator architecture for generative adversarial networks. 2019 IEEE/CVF Conference on Computer Vision and Pat- tern Recognition (CVPR), pages 4396–4405, 2018. 5

    2019

  17. [25]

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

    Levon Khachatryan, Andranik Movsisyan, Vahram Tade- vosyan, Roberto Henschel, Zhangyang Wang, Shant Navasardyan, and Humphrey Shi. Text2video-zero: Text- to-image diffusion models are zero-shot video generators. 2023 IEEE/CVF International Conference on Computer Vi- sion (ICCV...

    2023

  18. [26]

    Diffusion video autoencoders: Toward temporally consistent face video editing via disen- tangled video encoding

    Gyeongman Kim, Hajin Shim, Hyunsung Kim, Yunjey Choi, Junho Kim, and Eunho Yang. Diffusion video autoencoders: Toward temporally consistent face video editing via disen- tangled video encoding. 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , pages ...

    2023

  19. [27]

    Maskgan: Towards diverse and interactive facial image ma- nipulation

    Cheng-Han Lee, Ziwei Liu, Lingyun Wu, and Ping Luo. Maskgan: Towards diverse and interactive facial image ma- nipulation. In IEEE Conference on Computer Vision and Pat- tern Recognition (CVPR), 2020. 4

    2020

  20. [28]

    Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models

    Junnan Li, Dongxu Li, Silvio Savarese, and Steven Hoi. Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models. In In- ternational conference on machine learning , pages 19730– 19742. PMLR, 2023. 5

    2023

  21. [29]

    Deep learning face attributes in the wild

    Ziwei Liu, Ping Luo, Xiaogang Wang, and Xiaoou Tang. Deep learning face attributes in the wild. In Proceedings of International Conference on Computer Vision (ICCV) , De- cember 2015. 5

    2015

  22. [30]

    Dpm-solver: a fast ode solver for diffu- sion probabilistic model sampling in around 10 steps

    Cheng Lu, Yuhao Zhou, Fan Bao, Jianfei Chen, Chongxuan Li, and Jun Zhu. Dpm-solver: a fast ode solver for diffu- sion probabilistic model sampling in around 10 steps. In Proceedings of the 36th International Conference on Neu- ral Information Processing Systems , NIPS ’22, Red...

    2024

  23. [31]

    Introduction to Information Retrieval

    Christopher D Manning, Prabhakar Raghavan, and Hinrich Sch¨utze. Introduction to Information Retrieval. Cambridge University Press, 2008. 5

    2008

  24. [32]

    Sdedit: Guided image synthesis and editing with stochastic differential equa- tions

    Chenlin Meng, Yutong He, Yang Song, Jiaming Song, Jia- jun Wu, Jun-Yan clip, and Stefano Ermon. Sdedit: Guided image synthesis and editing with stochastic differential equa- tions. In International Conference on Learning Representa- tions, 2021. 1, 2

    2021

  25. [33]

    Dreamix: Video diffusion models are general video editors

    Eyal Molad, Eliahu Horwitz, Dani Valevski, Alex Rav Acha, Yossi Matias, Yael Pritch, Yaniv Leviathan, and Yedid Hoshen. Dreamix: Video diffusion models are general video editors. arXiv preprint arXiv:2302.01329, 2023. 2, 4

    2023 arXiv

  26. [34]

    Parkhi, Andrea Vedaldi, and Andrew Zisserman

    Omkar M. Parkhi, Andrea Vedaldi, and Andrew Zisserman. Deep face recognition. InProceedings of the British Machine Vision Conference (BMVC), pages 41.1–41.12. BMV A Press,

  27. [35]

    Sdxl: Improving latent diffusion models for high-resolution image synthesis

    Dustin Podell, Zion English, Kyle Lacey, Andreas Blattmann, Tim Dockhorn, Jonas M ¨uller, Joe Penna, and Robin Rombach. Sdxl: Improving latent diffusion models for high-resolution image synthesis. In International Con- ference on Learning Representations, 2023. 2, 3, 4

    2023

  28. [36]

    Fatezero: Fus- ing attentions for zero-shot text-based video editing

    Chenyang Qi, Xiaodong Cun, Yong Zhang, Chenyang Lei, Xintao Wang, Ying Shan, and Qifeng Chen. Fatezero: Fus- ing attentions for zero-shot text-based video editing. 2023 IEEE/CVF International Conference on Computer Vision (ICCV), pages 15886–15896, 2023. 2

    2023

  29. [37]

    Hierarchical text-conditional image gen- eration with clip latents

    Aditya Ramesh, Prafulla Dhariwal, Alex Nichol, Casey Chu, and Mark Chen. Hierarchical text-conditional image gen- eration with clip latents. arXiv preprint arXiv:2204.06125,

    arXiv

  30. [38]

    Blattmann, Dominik Lorenz, Patrick Esser, and Bj ¨orn Ommer

    Robin Rombach, A. Blattmann, Dominik Lorenz, Patrick Esser, and Bj ¨orn Ommer. High-resolution image synthesis with latent diffusion models. 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 10674–10685, 2021. 1

    2022

  31. [39]

    U-net: Convolutional networks for biomedical image segmentation

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-net: Convolutional networks for biomedical image segmentation. In Nassir Navab, Joachim Hornegger, William M. Wells, and Alejandro F. Frangi, editors, Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015...

    2015

  32. [40]

    Sara Mahdavi, Raphael Gontijo-Lopes, Tim Salimans, Jonathan Ho, David J Fleet, and Mohammad Norouzi

    Chitwan Saharia, William Chan, Saurabh Saxena, Lala Lit, Jay Whang, Emily Denton, Seyed Kamyar Seyed Ghasemipour, Burcu Karagol Ayan, S. Sara Mahdavi, Raphael Gontijo-Lopes, Tim Salimans, Jonathan Ho, David J Fleet, and Mohammad Norouzi. Photorealistic text-to-image diffusion ...

    2024

  33. [41]

    Facenet: A unified embedding for face recognition and clus- tering

    Florian Schroff, Dmitry Kalenichenko, and James Philbin. Facenet: A unified embedding for face recognition and clus- tering. In Proceedings of the IEEE Conference on Com- puter Vision and Pattern Recognition (CVPR) , pages 815– 823, 2015. 5

    2015

  34. [42]

    Laion-5b: an open large-scale dataset for training next generation image-text models

    Christoph Schuhmann, Romain Beaumont, Richard Vencu, Cade Gordon, Ross Wightman, Mehdi Cherti, Theo Coombes, Aarush Katta, Clayton Mullis, Mitchell Worts- man, Patrick Schramowski, Srivatsa Kundurthy, Katherine Crowson, Ludwig Schmidt, Robert Kaczmarczyk, and Jenia Jitsev. Lai...

    2024

  35. [43]

    Denois- ing diffusion implicit models

    Jiaming Song, Chenlin Meng, and Stefano Ermon. Denois- ing diffusion implicit models. In International Conference on Learning Representations, 2020. 1, 3

    2020

  36. [44]

    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 , page 402–419, Berlin, Heidelberg, 2020. Springer-Verlag. 7

    2020

  37. [45]

    Plug-and-play diffusion features for text-driven image-to-image translation

    Narek Tumanyan, Michal Geyer, Shai Bagon, and Tali Dekel. Plug-and-play diffusion features for text-driven image-to-image translation. 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 1921–1930, 2022. 2

    2023

  38. [46]

    Stitch it in time: Gan-based facial editing of real videos

    Rotem Tzaban, Ron Mokady, Rinon Gal, Amit Bermano, and Daniel Cohen-Or. Stitch it in time: Gan-based facial editing of real videos. In SIGGRAPH Asia 2022 Conference Papers, SA ’22, New York, NY , USA, 2022. Association for Computing Machinery. 2, 5, 6, 7, 8

    2022

  39. [47]

    The trec-8 question answering track re- port

    Ellen M V oorhees. The trec-8 question answering track re- port. In Proceedings of the 8th Text Retrieval Conference (TREC-8). National Institute of Standards and Technology (NIST), 1999. 5

    1999

  40. [48]

    Wang, Yitong Wang, Zheng Zhou, Xing Ji, Zhifeng Li, Dihong Gong, Jin Zhou, and Wei Liu

    H. Wang, Yitong Wang, Zheng Zhou, Xing Ji, Zhifeng Li, Dihong Gong, Jin Zhou, and Wei Liu. Cosface: Large mar- gin cosine loss for deep face recognition. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 5265–5274, 2018. 5, 6

    2018

  41. [49]

    Tune-a-video: One-shot tuning of im- age 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 im- age diffusion models for text-to-video generation. 2023 IEEE/CVF International Conference on Computer Vision (ICCV), pag...

    2023

  42. [50]

    Tedigan: Text-guided diverse face image generation and ma- nipulation

    Weihao Xia, Yujiu Yang, Jing-Hao Xue, and Baoyuan Wu. Tedigan: Text-guided diverse face image generation and ma- nipulation. In IEEE Conference on Computer Vision and Pat- tern Recognition (CVPR), 2021. 4

    2021

  43. [51]

    Wong, and Ping Luo

    Yangyang Xu, Shengfeng He, Kwan-Yee K. Wong, and Ping Luo. Rigid: Recurrent gan inversion and editing of real face videos. 2023 IEEE/CVF International Conference on Com- puter Vision (ICCV), pages 13645–13655, 2023. 5

    2023

  44. [52]

    A latent transformer for disentangled face editing in im- ages and videos

    Xu Yao, Alasdair Newson, Yann Gousseau, and Pierre Hel- lier. A latent transformer for disentangled face editing in im- ages and videos. 2021 IEEE/CVF International Conference on Computer Vision (ICCV), pages 13769–13778, 2021. 2, 5, 6, 7

    2021

  45. [53]

    Fed-nerf: Achieve high 3d consistency and temporal coherence for face video editing on dynamic nerf

    Hao Zhang, Yu-Wing Tai, and Chi-Keung Tang. Fed-nerf: Achieve high 3d consistency and temporal coherence for face video editing on dynamic nerf. ArXiv, abs/2401.02616,

    arXiv

  46. [54]

    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 (ICCV), pages 3836–3847, October 2023. 2

    2023

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.