- Rating
- 3
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper proposes LETHE, a proactive image protection method designed to prevent unauthorized image-to-video (I2V) generation. The central observation is that subsequent video frames acquire information from the reference image through attention operations, where the diversity of the reference value vectors constrains the diversity of the information that can be retrieved.
Critical strengths
- Novel and well-motivated method: The paper provides an interesting perspective on I2V protection by targeting the reference-reading mechanism through value collapse and attention attraction. The theoretical analysis and mechanism experiments support the proposed design.
- Strong experimental performance: LETHE consistently outperforms existing methods across different I2V models and datasets. The additional robustness and transferability experiments further strengthen the empirical evaluation.
Critical weaknesses
- Gap between theoretical analysis and practical protection: The theoretical analysis focuses on value diversity within individual attention operations, but does not directly characterize the behavior of the complete I2V generation process. Therefore, the theoretical results provide motivation rather than a guarantee of protection effectiveness.
- High computational overhead: Protecting a single image requires approximately 7–8 minutes on H100 hardware. This limits practical scalability, and the paper lacks a direct comparison of protection-generation costs with existing methods.
- Limited generalization to challenging real-world images: The evaluation primarily focuses on carefully selected images with clear, frontal, and unobstructed faces. More experiments on occluded faces, non-frontal views, multiple people, and low-quality images are needed to establish practical robustness.
- Incomplete component ablation: The paper does not evaluate attention attraction independently, making it difficult to fully isolate the necessity and contribution of each proposed objective.
Questions for authors
See weakness and solve my concerns.
4 comments
Rebuttal by Authors (Part 2/2)
W3: Generalization to challenging real-world images (occluded faces, non-frontal views, multiple people, and low-quality images).
A3: Thank you for this constructive suggestion! We focused on clear frontal faces mainly because ArcFace (used for ISM) is most reliable on them, whereas by design, LETHE does not rely on any face-specific priors/component. Following your suggestion, we further build a challenging set of 100 images (25 for each of the four categories you mentioned). As ArcFace is less reliable here, we instead report two DINO-based metrics of VBench, i.e., I2V-Subject (similarity to the reference image) and Subject-Consistency (consistency across frames), together with QAF and QAV (on Wan2.1).
| Metric | Clean | I2VGuard | Anti-Prompt | Ours |
|---|---|---|---|---|
| I2V-Subject↓ | 0.966 | 0.922 | 0.924 | 0.843 |
| Subject-Cons.↓ | 0.945 | 0.934 | 0.889 | 0.834 |
| QAF↓ | 0.786 | 0.492 | 0.698 | 0.364 |
| QAV↓ | 0.836 | 0.559 | 0.761 | 0.448 |
From the table, LETHE remains the most effective on all four metrics (e.g., QAF: 0.364 vs. 0.492 of the second-best), and the same trend also holds within each of the four categories and all metrics. These results indicate that the effectiveness of LETHE extends beyond clear frontal faces. We will add these results in our revision and release these images under the same terms as a part of our benchmark's challenging subset. Thank you again for making our evaluation stronger!
W4: Additional ablation study on attention attraction objective.
A4: Thank you for this helpful suggestion! Following your suggestion, we further evaluate the variant with only (mean of three runs, Wan2.1), which completes Tab. 2 as follows:
| Dataset | Variant | ISM-A↓ | ISM-F↓ | QAF↓ | QAV↓ |
|---|---|---|---|---|---|
| CelebV-Text | Clean | 0.8173 | 0.8725 | 0.9123 | 0.9462 |
| only | 0.6514 | 0.7282 | 0.5590 | 0.6403 | |
| only | 0.5899 | 0.6827 | 0.5224 | 0.6009 | |
| 0.4280 | 0.5308 | 0.3970 | 0.4976 | ||
| Kinetics-700 | Clean | 0.7584 | 0.8242 | 0.7974 | 0.8381 |
| only | 0.6283 | 0.7164 | 0.5473 | 0.6293 | |
| only | 0.5084 | 0.6127 | 0.4600 | 0.5324 | |
| 0.4125 | 0.4957 | 0.3852 | 0.4745 |
From the table, we have two observations (note that all variants use the same budget of ). (1) Each objective alone is already effective. takes effect by directly collapsing the diversity of the reference values, while alone disturbs the attention distribution over the reference tokens and thus also degrades the generation, although generally less effectively than . (2) Their combination is notably stronger than devoting the entire budget to either one (e.g., ISM-A on CelebV-Text: 0.651/0.590 → 0.428), as draws subsequent frames to read more from the collapsed values and thus notably amplifies the effect of . This indicates that the two objectives are complementary, and both are necessary.
We will add these results to Tab. 2 in our revision. Thank you again for your insightful and constructive comments!
Thank you for your thorough responses and additional experiments. The revisions have addressed my concerns. I appreciate the authors' efforts in improving the manuscript. I have no further questions or concerns and will keep my positive score.
Thank You for Your Positive Feedback!
Dear Reviewer e68T, thank you so much for your positive feedback and for maintaining your positive score! It encourages us a lot! We also sincerely appreciate your prompt response and your constructive comments, which have helped us further improve our paper. We will ensure that the revised version includes all the new experiments (e.g., the complete ablation study and the evaluation on challenging images) and the corresponding discussions.
Rebuttal by Authors (Part 1/2)
Dear Reviewer e68T, we sincerely thank you for your precious time and valuable comments. We are deeply encouraged by your recognition of our novel and well-motivated method, and our strong performance. We hope the following responses address your concerns.
W1: Gap between theoretical analysis and practical protection.
A1: Thank you for this thoughtful comment! We fully agree that Prop. 1 characterizes a single attention operation rather than the complete generation process, and thus serves as a motivation rather than a guarantee. In fact, this is also why we did not rely on the bound alone in our paper (see "Empirical Validation" in Sec. 3 and the limitations in App. D.3). Besides, we also bridge this gap empirically from two aspects:
As such, we believe our theory provides a useful and empirically supported insight for designing the protection, although a formal end-to-end guarantee for the multi-layer and multi-step generation process remains challenging, which we leave as our future work. We will make this clearer in our revision.
W2: Computational overhead.
A2: Thank you for the practical comment! Following your suggestion, we measure the per-step time and peak GPU memory of all methods on both Wan2.1 and CogVideoX. For fairness, all methods are run on a single H100 GPU under identical settings (bf16 precision, block-wise gradient checkpointing, and 17/49 frames for Wan2.1/CogVideoX during optimization):
We will add these results and more discussion in our revision.