arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2404.05331cs.CV

Mask-ControlNet:借助额外掩码提示实现更高质量的图像生成

Mask-ControlNet: Higher-Quality Image Generation with An Additional Mask Prompt

Tsinghua University(清华大学);Megvii(旷视科技);Sun Yat-sen University(中山大学)

Zhiqi Huang, Huixin Xiong, Haoyu Wang, Longguang Wang, Zhiheng Li

AI总结:

本文提出 Mask-ControlNet 框架,通过大型视觉模型分割参考图像并引入物体掩码提示,增强扩散模型对复杂前景-背景关系的控制,在基准上提升图像质量与参考保真度。

摘要:

文本到图像生成已经取得了巨大进展,尤其是随着近期扩散模型的发展。由于文本无法提供物体外观等详细条件,参考图像通常被用于控制生成图像中的物体。然而,当前景与背景之间的关系较为复杂时,现有方法的准确性仍然有限。为解决这一问题,我们通过引入额外的掩码提示,开发了一个名为 Mask-ControlNet 的框架。具体而言,我们首先利用大型视觉模型获取掩码,以分割参考图像中感兴趣的物体。随后,物体图像被用作额外提示,以促使扩散模型在图像生成过程中更好地理解前景区域与背景区域之间的关系。实验表明,掩码提示增强了扩散模型的可控性,使其能够在获得更好图像质量的同时,对参考图像保持更高的保真度。与以往文本到图像生成方法的比较表明,我们的方法在基准数据集上具有更优异的定量和定性性能。

英文摘要:

Text-to-image generation has witnessed great progress, especially with the recent advancements in diffusion models. Since texts cannot provide detailed conditions like object appearance, reference images are usually leveraged for the control of objects in the generated images. However, existing methods still suffer limited accuracy when the relationship between the foreground and background is complicated. To address this issue, we develop a framework termed Mask-ControlNet by introducing an additional mask prompt. Specifically, we first employ large vision models to obtain masks to segment the objects of interest in the reference image. Then, the object images are employed as additional prompts to facilitate the diffusion model to better understand the relationship between foreground and background regions during image generation. Experiments show that the mask prompts enhance the controllability of the diffusion model to maintain higher fidelity to the reference image while achieving better image quality. Comparison with previous text-to-image generation methods demonstrates our method's superior quantitative and qualitative performance on the benchmark datasets.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑