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.
@article{arxiv.2503.16795,
title = {DCEdit: Dual-Level Controlled Image Editing via Precisely Localized Semantics},
author = {Yihan Hu and Jianing Peng and Yiheng Lin and Ting Liu and Xiaochao Qu and Luoqi Liu and Yao Zhao and Yunchao Wei},
journal= {arXiv preprint arXiv:2503.16795},
year = {2025}
}