Cross-Modal Learning via Pairwise Constraints

Ran He, Man Zhang, Liang Wang +2

In multimedia applications, the text and image components in a web document form a pairwise constraint that potentially indicates the same semantic concept. This paper studies cros…