Dual-Schedule Inversion: Training- and Tuning-Free Inversion for Real Image 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_ | 1866915065241796608 |
|---|---|
| author | Huang, Jiancheng Huang, Yi Liu, Jianzhuang Zhou, Donghao Liu, Yifan Chen, Shifeng |
| author_facet | Huang, Jiancheng Huang, Yi Liu, Jianzhuang Zhou, Donghao Liu, Yifan Chen, Shifeng |
| contents | Text-conditional image editing is a practical AIGC task that has recently emerged with great commercial and academic value. For real image editing, most diffusion model-based methods use DDIM Inversion as the first stage before editing. However, DDIM Inversion often results in reconstruction failure, leading to unsatisfactory performance for downstream editing. To address this problem, we first analyze why the reconstruction via DDIM Inversion fails. We then propose a new inversion and sampling method named Dual-Schedule Inversion. We also design a classifier to adaptively combine Dual-Schedule Inversion with different editing methods for user-friendly image editing. Our work can achieve superior reconstruction and editing performance with the following advantages: 1) It can reconstruct real images perfectly without fine-tuning, and its reversibility is guaranteed mathematically. 2) The edited object/scene conforms to the semantics of the text prompt. 3) The unedited parts of the object/scene retain the original identity. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_11152 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Dual-Schedule Inversion: Training- and Tuning-Free Inversion for Real Image Editing Huang, Jiancheng Huang, Yi Liu, Jianzhuang Zhou, Donghao Liu, Yifan Chen, Shifeng Computer Vision and Pattern Recognition Text-conditional image editing is a practical AIGC task that has recently emerged with great commercial and academic value. For real image editing, most diffusion model-based methods use DDIM Inversion as the first stage before editing. However, DDIM Inversion often results in reconstruction failure, leading to unsatisfactory performance for downstream editing. To address this problem, we first analyze why the reconstruction via DDIM Inversion fails. We then propose a new inversion and sampling method named Dual-Schedule Inversion. We also design a classifier to adaptively combine Dual-Schedule Inversion with different editing methods for user-friendly image editing. Our work can achieve superior reconstruction and editing performance with the following advantages: 1) It can reconstruct real images perfectly without fine-tuning, and its reversibility is guaranteed mathematically. 2) The edited object/scene conforms to the semantics of the text prompt. 3) The unedited parts of the object/scene retain the original identity. |
| title | Dual-Schedule Inversion: Training- and Tuning-Free Inversion for Real Image Editing |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2412.11152 |