ZONE: Zero-Shot Instruction-Guided Local Editing
Autor: | Li, Shanglin, Zeng, Bohan, Feng, Yutang, Gao, Sicheng, Liu, Xuhui, Liu, Jiaming, Lin, Li, Tang, Xu, Hu, Yao, Liu, Jianzhuang, Zhang, Baochang |
---|---|
Rok vydání: | 2023 |
Předmět: | |
Druh dokumentu: | Working Paper |
Popis: | Recent advances in vision-language models like Stable Diffusion have shown remarkable power in creative image synthesis and editing.However, most existing text-to-image editing methods encounter two obstacles: First, the text prompt needs to be carefully crafted to achieve good results, which is not intuitive or user-friendly. Second, they are insensitive to local edits and can irreversibly affect non-edited regions, leaving obvious editing traces. To tackle these problems, we propose a Zero-shot instructiON-guided local image Editing approach, termed ZONE. We first convert the editing intent from the user-provided instruction (e.g., "make his tie blue") into specific image editing regions through InstructPix2Pix. We then propose a Region-IoU scheme for precise image layer extraction from an off-the-shelf segment model. We further develop an edge smoother based on FFT for seamless blending between the layer and the image.Our method allows for arbitrary manipulation of a specific region with a single instruction while preserving the rest. Extensive experiments demonstrate that our ZONE achieves remarkable local editing results and user-friendliness, outperforming state-of-the-art methods. Code is available at https://github.com/lsl001006/ZONE. Comment: Accepted at CVPR 2024 |
Databáze: | arXiv |
Externí odkaz: |