Which AI Models Create Accurate Alt Text for Picture Books?

Date & Time

Thursday, March 21, 2024 - 1:20 PM PDT  

Location

Platinum 1  

Description

The accessibility of children’s picture books has often been approached from the perspective of children who are blind and have low-vision, but what about blind and low-vision parents? Parents, especially those with sighted children, want access to the images in children’s books to engage their children fully in the story. These images are rarely described in print-braille or EPUB digital book formats, raising the question of whether image-to-text AI models may help bridge the gap. Can existing AI models describe the rich illustrations found in children’s books? In this session, we will share the results of a research study in which we tested five of the most promising AI models against 20 representative images from various popular children’s picture books. Join us to learn about the current state of image accessibility in children’s picture books, which AI models show the most promise for generating reliable image descriptions and what resources exist for blind parents seeking more accessible reading experiences with their children.

This Presentation Link is provided by the Presenter(s) and not hosted by the Center on Disabilities at CSUN. The Center on Disabilities has confirmed, as of April 4, 2024, content linked is relevant to the presentation, but has not been reviewed for accessibility nor will the Center on Disabilities attempt to remediate any accessibility issues in the linked content. Please contact the Presenter(s) with any accessibility concerns.

 

Audience

• Higher Education
• Information & Communications Technology
• Disability Specific
• Research & Development

Audience Level

Intermediate  

Session Summary (Abstract)

Join us to learn about the current state of image accessibility in children’s picture books, which AI models generate accurate alt text for picture books and what resources exist for blind parents seeking more accessible reading experiences.  

Primary Topic

Artificial Intelligence (AI) & Machine Learning (ML)  

Secondary Topics

• Blind/Low Vision
• Digital Accessibility
• Research

 

Session Type

Journal Track  

Presenters

• Quan (Monica) Zhou
  UC Irvine
• Nathan Samarasena
  UC Irvine
• Kyle Han
  UC Irvine
• Nataly López
  UC Irvine
• Stacy Branham
  UC Irvine

Add to Outlook/iCal (How it works)

Back to Session List