Torralba & team to present work on image meaning at IEEE Computer Vision & Pattern Recognition conference

May 26, 2011

Antonio Torralba, the Esther and Harold E. Edgerton Associate Professor of Electrical Engineering and Computer Science, has teamed with Aude Oliva, associate professor of cognitive science and graduate students Phillip John Isola (Brain and Cognitive Sciences) and Jianxiong Xiao (EECS) to develop the first reliable method to determine human reaction to images. Based on these findings the team has developed an algorithm that can rank images according to memorability. Besides being useful to graphic designers, photo editors, or anyone trying to decide which of their vacation photos to post on Facebook, future photographic equipment including phone cameras and more could help guide photographers in the future in making appropriate subject choices.

The paper entitled "What makes an image memorable?" will be presented by the team at the IEEE Conference on Computer Vision and Pattern Recognition in Colorado Springs, June 20-25.

Read more about this work in the MIT News Office May 24 article.