<div dir="ltr"><a href="https://cmu.zoom.us/j/98299626118?pwd=YW9jbE40eDcrOHM1RW4yNFJ5LzRpUT09">https://cmu.zoom.us/j/98299626118?pwd=YW9jbE40eDcrOHM1RW4yNFJ5LzRpUT09</a><br></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Mon, Dec 4, 2023 at 11:51 AM Brigitte Schmidt <<a href="mailto:vjschulz@andrew.cmu.edu">vjschulz@andrew.cmu.edu</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Hi Everyone,<br>
<br>
I’ll actually be in Pittsburgh for this talk!<br>
I know our room is Location: NSH 4201 but what is our zoom link?<br>
<br>
Thank you so much.<br>
Vera Schulz<br>
<br>
> On Dec 1, 2023, at 6:25 PM, Aaron Steinfeld <<a href="mailto:steinfeld@cmu.edu" target="_blank">steinfeld@cmu.edu</a>> wrote:<br>
> <br>
> <br>
> Hi everyone,<br>
> <br>
> Oscar will be presenting some work from our lab on Monday:<br>
> <br>
> "Independent Navigation and Interaction for Visually Impaired Users via AR, Computer Vision, and LLMs"<br>
> <br>
> Abstract<br>
> <br>
> Navigating unfamiliar indoor environments and interacting with objects are challenging for people who are blind or low-vision (pBLV). In this talk, I describe a functional prototype for smartphone-based, indoor navigation via a conversational assistant for pBLV. The prototype utilizes a cognitive architecture to integrate three different technologies: real-time object/people detection for obstacle avoidance<br>
> <br>
> and discovery; augmented-reality spatial anchors for high-precision localization and virtual content interaction; and large language models (LLMs) for information extraction, conversational interaction, and turn-by-turn navigation. We assessed the individual and collective impact of different technologies through a human performance analysis where task performance time for a user is measured. We conclude that conversation helps holistically integrate disparate technologies to deliver a better, more context-aware user experience while remarkably reducing task completion time.<br>
> <br>
> _______________________________________________<br>
> Accessibility-lunch mailing list<br>
> <a href="mailto:Accessibility-lunch@lists.andrew.cmu.edu" target="_blank">Accessibility-lunch@lists.andrew.cmu.edu</a><br>
> <a href="https://lists.andrew.cmu.edu/mailman/listinfo/accessibility-lunch" rel="noreferrer" target="_blank">https://lists.andrew.cmu.edu/mailman/listinfo/accessibility-lunch</a><br>
_______________________________________________<br>
Accessibility-lunch mailing list<br>
<a href="mailto:Accessibility-lunch@lists.andrew.cmu.edu" target="_blank">Accessibility-lunch@lists.andrew.cmu.edu</a><br>
<a href="https://lists.andrew.cmu.edu/mailman/listinfo/accessibility-lunch" rel="noreferrer" target="_blank">https://lists.andrew.cmu.edu/mailman/listinfo/accessibility-lunch</a><br>
</blockquote></div><br clear="all"><div><br></div><span class="gmail_signature_prefix">-- </span><br><div dir="ltr" class="gmail_signature"><div dir="ltr">==<br>Associate Professor<br>Human-Computer Interaction Institute<br>Language Technologies Institute<br>Carnegie Mellon University<br><a href="http://jeffreybigham.com" target="_blank">jeffreybigham.com</a></div></div>