Leveraging unpaired out-of-domain data for image captioning
Pattern Recognition Letters(2020)
摘要
•A novel method using out-of-domain data for image captioning has been proposed.•The proposed model just need unpaired image data and text data in pre-training stage.•We proposed new variants of concepts-decoder framework in image captioning field.•Our proposed method can promote the baseline model significantly.
更多查看译文
关键词
41A05,41A10,65D05,65D17
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络