arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1909.07083cs.CVcs.CLcs.LG
收藏

可控的文本到图像生成

Controllable Text-to-Image Generation

University of Oxford(牛津大学)

Bowen Li, Xiaojuan Qi, Thomas Lukasiewicz, Philip H. S. Torr

摘要:

在本文中,我们提出了一种新的可控的文本到图像的生成对抗网络(ControlGAN),它可以有效地合成高质量的图像,并根据自然语言描述控制部分图像生成。为了实现这一目标,我们引入了一个词级空间和通道的注意力驱动的生成器,可以解开不同的视觉属性,并允许模型专注于生成和操纵对应于最相关的单词的子区域。此外,提出了一个字级的监督反馈提供细粒度的监督反馈,通过相关的话与图像区域,促进训练一个有效的生成器,能够操纵特定的视觉属性,而不影响其他内容的生成。此外,感知损失被用来减少图像生成中的随机性,并鼓励生成器操纵修改文本中所需的特定属性。在基准数据集上的大量实验表明,我们的方法优于现有的最先进的,并且能够有效地操纵合成图像使用自然语言描述。代码可在https://github.com/mrlibw/ControlGAN上获得。

英文摘要:

In this paper, we propose a novel controllable text-to-image generative adversarial network (ControlGAN), which can effectively synthesise high-quality images and also control parts of the image generation according to natural language descriptions. To achieve this, we introduce a word-level spatial and channel-wise attention-driven generator that can disentangle different visual attributes, and allow the model to focus on generating and manipulating subregions corresponding to the most relevant words. Also, a word-level discriminator is proposed to provide fine-grained supervisory feedback by correlating words with image regions, facilitating training an effective generator which is able to manipulate specific visual attributes without affecting the generation of other content. Furthermore, perceptual loss is adopted to reduce the randomness involved in the image generation, and to encourage the generator to manipulate specific attributes required in the modified text. Extensive experiments on benchmark datasets demonstrate that our method outperforms existing state of the art, and is able to effectively manipulate synthetic images using natural language descriptions. Code is available at https://github.com/mrlibw/ControlGAN.

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。

↑