Browsing by Subject "People with visual impairments"
Now showing 1 - 2 of 2
- Results Per Page
- Sort Options
Item Designing algorithms that assist people to ask visual questions(2019-08-16) Wang, Yanan, M.S. in Information Studies; Gurari, DannaVisual question answering services can help people with visual impairments answer their visual questions by supporting them to submit an image and a question. However, people with visual impairments may not know exactly what is captured in their images, and their questions may not be explicit and specific enough to be answered accurately. In the project, we designed several algorithms to assist people by identifying when to instruct them to take a higher quality image, ask a valid question, or clarify the object of interest. This can help people to make the visual question they asked answerable with a single, unambiguous answer. The algorithms are evaluated based on accuracy and demonstrated to be effective. Finally, we designed a system to interact with users with our algorithms and will implement it for a user study in future workItem Image captioning algorithms for images taken by people with visual impairments(2019-07-08) Zhang, Meng, M.S. in Information Studies; Gurari, DannaPeople with visual impairments regularly encounter the challenge that their visual impairments expose them to a time-consuming, or even impossible, task: what content is presented in an image without assistance. One method to address this problem is image captioning with machine learning. With the help of image captioning algorithms together with artificial intelligence speech system, people who are blind can instantly learn what is in an image, since such systems can automatically generate text captions. In this work, we analyze the new VizWiz dataset and compare it to the MSCOCO dataset, which is widely used for evaluating the performance of image captioning algorithms. We also implement and evaluate two state-of-the-art image caption models with accuracy, runtime, and resource analysis. Hopefully, our research will help the improvement of image captioning algorithms which focus on fulfilling the everyday needs of people with visual impairments