Knowledge Commons of Institute of Automation,CAS
Data-efficient image captioning of fine art paintings via virtual-real semantic alignment training | |
Lu Yue1,2; Guo Chao1,2; Dai Xingyuan1,2; Wang Fei-yue1 | |
发表期刊 | Neurocomputing |
2022 | |
页码 | 163–180 |
摘要 | The image captioning of fine art paintings aims at generating a sentence to describe the painting content. Compared with photographic images, there are few annotated data of clear and precise content descriptions for paintings. Besides, painting images usually have abstract expressions, making it hard to extract their representative features. In this paper, we propose a virtual-real semantic alignment training process to address these challenges in painting captioning. To provide sufficient training data, we generate a virtual painting captioning dataset by applying style transfer to a large-scale photographic image captioning dataset and maintaining their annotations. To tackle the difficulty of abstract expressions, we employ a semantic alignment loss between photographic image features and virtual painting features to guide the training of the painting feature extractor. We evaluate our method in two data-hungry scenarios where only a few or no annotated painting data for training. According to the evaluation results on a public painting captioning dataset and our annotated painting captioning dataset, our model achieves significant improvements and higher data efficiency than the baselines in the two data-hungry scenarios on all datasets. |
关键词 | 平行艺术 绘画图像标注 |
收录类别 | SCI |
文献类型 | 期刊论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/48745 |
专题 | 多模态人工智能系统全国重点实验室_平行智能技术与系统团队 |
通讯作者 | Wang Fei-yue |
作者单位 | 1.中国科学院自动化研究所 2.中国科学院大学 |
第一作者单位 | 中国科学院自动化研究所 |
通讯作者单位 | 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | Lu Yue,Guo Chao,Dai Xingyuan,et al. Data-efficient image captioning of fine art paintings via virtual-real semantic alignment training[J]. Neurocomputing,2022:163–180. |
APA | Lu Yue,Guo Chao,Dai Xingyuan,&Wang Fei-yue.(2022).Data-efficient image captioning of fine art paintings via virtual-real semantic alignment training.Neurocomputing,163–180. |
MLA | Lu Yue,et al."Data-efficient image captioning of fine art paintings via virtual-real semantic alignment training".Neurocomputing (2022):163–180. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
1-s2.0-S092523122200(6207KB) | 期刊论文 | 作者接受稿 | 开放获取 | CC BY-NC-SA | 浏览 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论