Beyond Editing Pairs: Fine-Grained Instructional Image Editing via Multi-Scale Learnable Regions
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916757673869312 |
|---|---|
| author | Ma, Chenrui Xiao, Xi Wang, Tianyang Shen, Yanning |
| author_facet | Ma, Chenrui Xiao, Xi Wang, Tianyang Shen, Yanning |
| contents | Current text-driven image editing methods typically follow one of two directions: relying on large-scale, high-quality editing pair datasets to improve editing precision and diversity, or exploring alternative dataset-free techniques. However, constructing large-scale editing datasets requires carefully designed pipelines, is time-consuming, and often results in unrealistic samples or unwanted artifacts. Meanwhile, dataset-free methods may suffer from limited instruction comprehension and restricted editing capabilities. Faced with these challenges, the present work develops a novel paradigm for instruction-driven image editing that leverages widely available and enormous text-image pairs, instead of relying on editing pair datasets. Our approach introduces a multi-scale learnable region to localize and guide the editing process. By treating the alignment between images and their textual descriptions as supervision and learning to generate task-specific editing regions, our method achieves high-fidelity, precise, and instruction-consistent image editing. Extensive experiments demonstrate that the proposed approach attains state-of-the-art performance across various tasks and benchmarks, while exhibiting strong adaptability to various types of generative models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_19352 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Beyond Editing Pairs: Fine-Grained Instructional Image Editing via Multi-Scale Learnable Regions Ma, Chenrui Xiao, Xi Wang, Tianyang Shen, Yanning Computer Vision and Pattern Recognition Current text-driven image editing methods typically follow one of two directions: relying on large-scale, high-quality editing pair datasets to improve editing precision and diversity, or exploring alternative dataset-free techniques. However, constructing large-scale editing datasets requires carefully designed pipelines, is time-consuming, and often results in unrealistic samples or unwanted artifacts. Meanwhile, dataset-free methods may suffer from limited instruction comprehension and restricted editing capabilities. Faced with these challenges, the present work develops a novel paradigm for instruction-driven image editing that leverages widely available and enormous text-image pairs, instead of relying on editing pair datasets. Our approach introduces a multi-scale learnable region to localize and guide the editing process. By treating the alignment between images and their textual descriptions as supervision and learning to generate task-specific editing regions, our method achieves high-fidelity, precise, and instruction-consistent image editing. Extensive experiments demonstrate that the proposed approach attains state-of-the-art performance across various tasks and benchmarks, while exhibiting strong adaptability to various types of generative models. |
| title | Beyond Editing Pairs: Fine-Grained Instructional Image Editing via Multi-Scale Learnable Regions |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2505.19352 |