arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2008.04200cs.CVcs.CL
收藏

描述要改变的内容:一种文本引导的无监督图像到图像翻译方法

Describe What to Change: A Text-guided Unsupervised Image-to-Image Translation Approach

University of Trento(特伦托大学);FBK(布鲁诺·凯斯勒基金会);The Chinese University of Hong Kong(香港中文大学);Tencent AI Lab(腾讯人工智能实验室);Inria Grenoble Rhône-Alpes(法国国家信息与自动化研究所格勒诺布尔罗讷-阿尔卑斯中心);Huawei Research(华为研究院)

Yahui Liu, Marco De Nadai, Deng Cai, Huayang Li, Xavier Alameda-Pineda, Nicu Sebe, Bruno Lepri

摘要:

通过人类书写的文本来操纵图像的视觉属性是一项非常具有挑战性的任务。一方面,模型必须在没有所需输出的基础事实的情况下学习操作。另一方面,模型必须处理自然语言固有的模糊性。以往的研究通常要求用户描述所需图像的所有特征,或者使用注释丰富的图像描述数据集。在这项工作中,我们提出了一种新的无监督的方法,基于图像到图像的翻译,通过命令般的句子,如“改变头发颜色为黑色”,改变给定图像的属性。与最先进的方法不同,我们的模型不需要人工注释的数据集,也不需要对所需图像的所有属性进行文本描述,而只需要修改那些属性。我们提出的模型从视觉属性中分离出图像内容,并在从内容和修改后的属性表示生成新图像之前,使用文本描述学习修改后者。因为文本可能本质上是模糊的(金发可能指金发的不同阴影,例如金色,冰冷,沙质),我们的方法生成同一翻译的多个随机版本。实验结果表明,该模型在CelebA和CUB两个大规模公共数据集上取得了较好的性能.我们相信,我们的方法将铺平道路的新途径的研究相结合的文本和语音命令与视觉属性。

英文摘要:

Manipulating visual attributes of images through human-written text is a very challenging task. On the one hand, models have to learn the manipulation without the ground truth of the desired output. On the other hand, models have to deal with the inherent ambiguity of natural language. Previous research usually requires either the user to describe all the characteristics of the desired image or to use richly-annotated image captioning datasets. In this work, we propose a novel unsupervised approach, based on image-to-image translation, that alters the attributes of a given image through a command-like sentence such as "change the hair color to black". Contrarily to state-of-the-art approaches, our model does not require a human-annotated dataset nor a textual description of all the attributes of the desired image, but only those that have to be modified. Our proposed model disentangles the image content from the visual attributes, and it learns to modify the latter using the textual description, before generating a new image from the content and the modified attribute representation. Because text might be inherently ambiguous (blond hair may refer to different shadows of blond, e.g. golden, icy, sandy), our method generates multiple stochastic versions of the same translation. Experiments show that the proposed model achieves promising performances on two large-scale public datasets: CelebA and CUB. We believe our approach will pave the way to new avenues of research combining textual and speech commands with visual attributes.

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。

↑