Research map / Video / Video generation
Are Image-to-Video Models Good Zero-Shot Image Editors?
Zechuan Zhang, Zhenyuan Chen, Zongxin Yang, Yi Yang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add chain-of-thought prompt enhancement (ablation) | IF-Edit · TEdBench | CLIP-T: 0.59 -> 0.65 | Improved |
| add temporal latent dropout (K=3 vs K=1) (ablation) | IF-Edit · TEdBench | inference time: 21s -> 12s | Improved |
| add temporal latent dropout (K=3 vs K=1) (ablation) | IF-Edit · TEdBench | CLIP-T: 0.65 -> 0.65 | No clear effect |
| increase dropout stride from K=3 to K=4 (ablation) | IF-Edit · TEdBench | CLIP-I: 0.96 -> 0.92 | Hurt |
| add self-consistent post-refinement (ablation) | IF-Edit · TEdBench | sharpness: 840 -> 983 | Improved |
| add self-consistent post-refinement (ablation) | IF-Edit · TEdBench | CLIP-T: 0.63 -> 0.65 | Improved |
| replace self-consistent refinement with VLM filter (ablation) | IF-Edit · TEdBench | inference time: 12s -> 37s | Hurt |
| replace self-consistent refinement with VLM filter (ablation) | IF-Edit · TEdBench | sharpness: 983 -> 895 | Hurt |
| IF-Edit (video prior) vs prior editors | IF-Edit · TEdBench | CLIP-I: 0.89 -> 0.96 | Improved |
| IF-Edit vs open-source editors | IF-Edit · RISEBench | overall accuracy: 8.9 -> 11.1 | Improved |
| IF-Edit vs specialized image editors | IF-Edit · ImgEdit | overall score: 3.71 -> 2.19 | Hurt |
Closest papers by meaning
- Learning Action and Reasoning-Centric Image Editing from Videos and Simulation (NeurIPS 2024 · similarity 0.91)
- CCEdit: Creative and Controllable Video Editing via Diffusion Models (CVPR 2024 · similarity 0.89)
- Bernini: Latent Semantic Planning for Video Diffusion (NeurIPS 2026 · similarity 0.89)
- VINCIE-NExT: Unlocking Video Editing from Images via In-Context Modeling (NeurIPS 2026 · similarity 0.89)
- MaskINT: Video Editing via Interpolative Non-autoregressive Masked Transformers (CVPR 2024 · similarity 0.89)
- FADE: Frequency-Aware Diffusion Model Factorization for Video Editing (CVPR 2025 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
