Author(s)

Han Li, Liang Bai, Songyang Lao, Zhihong Deng, Qiuyu Ren, Can Luo

Corresponding Author

Han Li

ABSTRACT

With the popularization and development of the Internet and digital media, the demand for cross-media retrieval based on images and texts has also increased. This paper first introduces the image-text retrieval and deep learning, and then gives the basic steps of cross retrieval. On this basis, the paper analyzes the key points and difficulties of image-text retrieval based on deep learning at present, and further discuss its research direction based on the existing work. Finally, the future development trend of image-text retrieval is discussed.

KEYWORDS

Image-text retrieval; deep learning; key and difficult points; research direction