Context-specific approaches for image caption generation

Automatic image captioning is a challenging Artificial Intelligence task. Given an image file, the objective is to automatically generate a natural language description of its salient visual content. This talk describes professor mason’s research modeling correlations between image and text features, which are learned from examples of captioned images found on the Internet. She will present results from applications in social media and online shopping websites.

Please join Professor Mason on Tuesday, September 30, 2014 for the first Faculty Lunch Talk of the Fall semester.  12:30 pm in PDR

 

Tagged