CS547 Human-Computer Interaction Seminar (Seminar on People, Computers, and Design)
Fridays 11:30am-12:30pm PT · Gates B3 · Open to the public
Amy Pavel University of Texas, Austin Making Videos Accessible
January 7, 2022
Video is becoming a primary medium for communication across education, work, social media, and entertainment. While videos are engaging and informative, they lack the familiar and useful affordances of text for browsing, skimming, and consuming information. This severely limits who can interact with video content and how they can interact with it, and means that much of the information in videos is not accessible to everyone. But, how can we make videos accessible for all users? In this talk, I'll share my work creating interactive Human-AI systems to improve the non-visual accessibility of videos during three key stages: (1) the planning and capture stage, (2) the post-production stage, and (3) the search and consumption stage. First, I'll discuss core challenges of finding information in videos from interviews of people with domain expertise and/or visual impairments. Then, I will present new systems that leverage AI, and the results of technical and user evaluations that demonstrate system efficacy. I will conclude by discussing how hybrid HCI-AI work will make digital communication more effective and accessible in the future, and how new interactions can help us to realize the full potential of recent AI/ML advances. |
|