Guardado en:
| Autores principales: | , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2511.18131 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866908670679318528 |
|---|---|
| author | Li, Xiaofan Sun, Yanpeng Wu, Chenming Duan, Fan Wang, YuAn Bo, Weihao Zhang, Yumeng Liang, Dingkang |
| author_facet | Li, Xiaofan Sun, Yanpeng Wu, Chenming Duan, Fan Wang, YuAn Bo, Weihao Zhang, Yumeng Liang, Dingkang |
| contents | We observe that recent advances in multimodal foundation models have propelled instruction-driven image generation and editing into a genuinely cross-modal, cooperative regime. Nevertheless, state-of-the-art editing pipelines remain costly: beyond training large diffusion/flow models, they require curating massive high-quality triplets of \{instruction, source image, edited image\} to cover diverse user intents. Moreover, the fidelity of visual replacements hinges on how precisely the instruction references the target semantics. We revisit this challenge through the lens of temporal modeling: if video can be regarded as a full temporal process, then image editing can be seen as a degenerate temporal process. This perspective allows us to transfer single-frame evolution priors from video pre-training, enabling a highly data-efficient fine-tuning regime. Empirically, our approach matches the performance of leading open-source baselines while using only about one percent of the supervision demanded by mainstream editing models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_18131 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Video4Edit: Viewing Image Editing as a Degenerate Temporal Process Li, Xiaofan Sun, Yanpeng Wu, Chenming Duan, Fan Wang, YuAn Bo, Weihao Zhang, Yumeng Liang, Dingkang Computer Vision and Pattern Recognition We observe that recent advances in multimodal foundation models have propelled instruction-driven image generation and editing into a genuinely cross-modal, cooperative regime. Nevertheless, state-of-the-art editing pipelines remain costly: beyond training large diffusion/flow models, they require curating massive high-quality triplets of \{instruction, source image, edited image\} to cover diverse user intents. Moreover, the fidelity of visual replacements hinges on how precisely the instruction references the target semantics. We revisit this challenge through the lens of temporal modeling: if video can be regarded as a full temporal process, then image editing can be seen as a degenerate temporal process. This perspective allows us to transfer single-frame evolution priors from video pre-training, enabling a highly data-efficient fine-tuning regime. Empirically, our approach matches the performance of leading open-source baselines while using only about one percent of the supervision demanded by mainstream editing models. |
| title | Video4Edit: Viewing Image Editing as a Degenerate Temporal Process |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2511.18131 |