Computer Vision News - December 2018

Lydia, can I ask you about your work? I do crowdsourcing. One of my main goals is to decompose hard problems so that individuals can do them better or they're easier for groups to break up. Many of those problems are about visual communication, how visual images can convey a meaning. This is used a lot in print media. Often, if you have a news article, you want an image to go at the top that will attract attention. The cover of the Economist does it a lot. They have a magazine with lots and lots of words printed inside it. If you don’t see the cover and get interested in the main story, you may not get to all those words in the middle. Visual communication tries to send a message like a headline or a public service announcement. It requires some creativity, but also there’s some grounding in what it has to do so there’s an actual purpose to it that we can evaluate in its meaning. The first time I heard about your work was at CVPR2018, at Adriana Kovashka ’ s workshop about understanding visual advertisements, and I was impressed by your presentation. What is the connection between your work and computer vision? Lydia Chilton Lydia Chilton is an assistant professor at Columbia University in New York where she works in the Computer Science Department. Her research is in human-computer interaction (HCI), crowdsourcing, and computational design. More interviews with women scientists 32 Women in Computer Vision Computer Vision News Women in Science “When you are interpreting symbols, first associations are better than logic”