Guardado en:
Detalles Bibliográficos
Autores principales: Hu, Yihan, Peng, Jianing, Lin, Yiheng, Liu, Ting, Qu, Xiaochao, Liu, Luoqi, Zhao, Yao, Wei, Yunchao
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:https://arxiv.org/abs/2503.16795
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866908276766015488
author Hu, Yihan
Peng, Jianing
Lin, Yiheng
Liu, Ting
Qu, Xiaochao
Liu, Luoqi
Zhao, Yao
Wei, Yunchao
author_facet Hu, Yihan
Peng, Jianing
Lin, Yiheng
Liu, Ting
Qu, Xiaochao
Liu, Luoqi
Zhao, Yao
Wei, Yunchao
contents This paper presents a novel approach to improving text-guided image editing using diffusion-based models. Text-guided image editing task poses key challenge of precisly locate and edit the target semantic, and previous methods fall shorts in this aspect. Our method introduces a Precise Semantic Localization strategy that leverages visual and textual self-attention to enhance the cross-attention map, which can serve as a regional cues to improve editing performance. Then we propose a Dual-Level Control mechanism for incorporating regional cues at both feature and latent levels, offering fine-grained control for more precise edits. To fully compare our methods with other DiT-based approaches, we construct the RW-800 benchmark, featuring high resolution images, long descriptive texts, real-world images, and a new text editing task. Experimental results on the popular PIE-Bench and RW-800 benchmarks demonstrate the superior performance of our approach in preserving background and providing accurate edits.
format Preprint
id arxiv_https___arxiv_org_abs_2503_16795
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DCEdit: Dual-Level Controlled Image Editing via Precisely Localized Semantics
Hu, Yihan
Peng, Jianing
Lin, Yiheng
Liu, Ting
Qu, Xiaochao
Liu, Luoqi
Zhao, Yao
Wei, Yunchao
Computer Vision and Pattern Recognition
This paper presents a novel approach to improving text-guided image editing using diffusion-based models. Text-guided image editing task poses key challenge of precisly locate and edit the target semantic, and previous methods fall shorts in this aspect. Our method introduces a Precise Semantic Localization strategy that leverages visual and textual self-attention to enhance the cross-attention map, which can serve as a regional cues to improve editing performance. Then we propose a Dual-Level Control mechanism for incorporating regional cues at both feature and latent levels, offering fine-grained control for more precise edits. To fully compare our methods with other DiT-based approaches, we construct the RW-800 benchmark, featuring high resolution images, long descriptive texts, real-world images, and a new text editing task. Experimental results on the popular PIE-Bench and RW-800 benchmarks demonstrate the superior performance of our approach in preserving background and providing accurate edits.
title DCEdit: Dual-Level Controlled Image Editing via Precisely Localized Semantics
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.16795