Working together to improve image captioning for people who are blind

Microsoft continues to push the boundaries of technology to improve and positively influence all parts of society. In order to explore new approaches at the intersection of computer vision and accessibility, we’re announcing an 18-month partnership with The University of Texas at Austin for our new Microsoft Ability Initiative. This initiative is one of many endeavors at Microsoft in which researchers and product developers are coming together in a cross-company push to spur innovative and exciting new research and development in the area of accessible technologies. UT Austin stood out last year from a select number of universities with specialized experience invited to participate in the competitive process to identify an academic partner for the initiative.

We are thrilled to work closely with Principal investigator Professor Danna Gurari and Professor Kenneth R. Fleischmann, who are leading the team at UT Austin, which also includes several graduate students. “I love to see the excitement in so many of my students when they realize that they can use their skills to make a difference in the world, especially for people who are blind or with low vision,” said Professor Danna Gurari, at UT Austin.

At the end of this collaborative journey, we hope to create a public image captioning dataset that will ultimately generate new waves of innovative research and lead to new technologies for people who are blind or with low vision. Read more about this partnership and the Microsoft Ability Initiative via the Microsoft Research Blog.