UniEdit: A Unified Tuning-Free Framework for Video Motion and Appearance Editing
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914743223058432 |
|---|---|
| author | Bai, Jianhong He, Tianyu Wang, Yuchi Guo, Junliang Hu, Haoji Liu, Zuozhu Bian, Jiang |
| author_facet | Bai, Jianhong He, Tianyu Wang, Yuchi Guo, Junliang Hu, Haoji Liu, Zuozhu Bian, Jiang |
| contents | Recent advances in text-guided video editing have showcased promising results in appearance editing (e.g., stylization). However, video motion editing in the temporal dimension (e.g., from eating to waving), which distinguishes video editing from image editing, is underexplored. In this work, we present UniEdit, a tuning-free framework that supports both video motion and appearance editing by harnessing the power of a pre-trained text-to-video generator within an inversion-then-generation framework. To realize motion editing while preserving source video content, based on the insights that temporal and spatial self-attention layers encode inter-frame and intra-frame dependency respectively, we introduce auxiliary motion-reference and reconstruction branches to produce text-guided motion and source features respectively. The obtained features are then injected into the main editing path via temporal and spatial self-attention layers. Extensive experiments demonstrate that UniEdit covers video motion editing and various appearance editing scenarios, and surpasses the state-of-the-art methods. Our code will be publicly available. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_13185 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | UniEdit: A Unified Tuning-Free Framework for Video Motion and Appearance Editing Bai, Jianhong He, Tianyu Wang, Yuchi Guo, Junliang Hu, Haoji Liu, Zuozhu Bian, Jiang Computer Vision and Pattern Recognition Recent advances in text-guided video editing have showcased promising results in appearance editing (e.g., stylization). However, video motion editing in the temporal dimension (e.g., from eating to waving), which distinguishes video editing from image editing, is underexplored. In this work, we present UniEdit, a tuning-free framework that supports both video motion and appearance editing by harnessing the power of a pre-trained text-to-video generator within an inversion-then-generation framework. To realize motion editing while preserving source video content, based on the insights that temporal and spatial self-attention layers encode inter-frame and intra-frame dependency respectively, we introduce auxiliary motion-reference and reconstruction branches to produce text-guided motion and source features respectively. The obtained features are then injected into the main editing path via temporal and spatial self-attention layers. Extensive experiments demonstrate that UniEdit covers video motion editing and various appearance editing scenarios, and surpasses the state-of-the-art methods. Our code will be publicly available. |
| title | UniEdit: A Unified Tuning-Free Framework for Video Motion and Appearance Editing |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2402.13185 |