EditCtrl: Disentangled Local and Global Control for Real-Time Generative Video Editing
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915903940067328 |
|---|---|
| author | Litman, Yehonathan Liu, Shikun Seyb, Dario Milef, Nicholas Zhou, Yang Marshall, Carl Tulsiani, Shubham Leak, Caleb |
| author_facet | Litman, Yehonathan Liu, Shikun Seyb, Dario Milef, Nicholas Zhou, Yang Marshall, Carl Tulsiani, Shubham Leak, Caleb |
| contents | High-fidelity generative video editing has seen significant quality improvements by leveraging pre-trained video foundation models. However, their computational cost is a major bottleneck, as they are often designed to inefficiently process the full video context regardless of the inpainting mask's size, even for sparse, localized edits. In this paper, we introduce EditCtrl, an efficient video inpainting control framework that focuses computation only where it is needed. Our approach features a novel local video context module that operates solely on masked tokens, yielding a computational cost proportional to the edit size. This local-first generation is then guided by a lightweight temporal global context embedder that ensures video-wide context consistency with minimal overhead. Not only is EditCtrl 10 times more compute efficient than state-of-the-art generative editing methods, it even improves editing quality compared to methods designed with full-attention. Finally, we showcase how EditCtrl unlocks new capabilities, including multi-region editing with text prompts and autoregressive content propagation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_15031 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | EditCtrl: Disentangled Local and Global Control for Real-Time Generative Video Editing Litman, Yehonathan Liu, Shikun Seyb, Dario Milef, Nicholas Zhou, Yang Marshall, Carl Tulsiani, Shubham Leak, Caleb Computer Vision and Pattern Recognition High-fidelity generative video editing has seen significant quality improvements by leveraging pre-trained video foundation models. However, their computational cost is a major bottleneck, as they are often designed to inefficiently process the full video context regardless of the inpainting mask's size, even for sparse, localized edits. In this paper, we introduce EditCtrl, an efficient video inpainting control framework that focuses computation only where it is needed. Our approach features a novel local video context module that operates solely on masked tokens, yielding a computational cost proportional to the edit size. This local-first generation is then guided by a lightweight temporal global context embedder that ensures video-wide context consistency with minimal overhead. Not only is EditCtrl 10 times more compute efficient than state-of-the-art generative editing methods, it even improves editing quality compared to methods designed with full-attention. Finally, we showcase how EditCtrl unlocks new capabilities, including multi-region editing with text prompts and autoregressive content propagation. |
| title | EditCtrl: Disentangled Local and Global Control for Real-Time Generative Video Editing |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2602.15031 |