Adobe Research at UIST 2022

October 31, 2022

Tags: Conferences

A screenshot of CrossA11y’s interface. The top-left corner contains the video player. Below the video player are the audio and visual timelines. Both timelines are colored from gray to red indicating their accessibility level. The captions pane and the video description pane are two columns featuring detected inaccessible segments. Each accessibility issue provides a text field that users can type in description or captions, as well as a vertical side bar that is also colored according to its accessibility level.
In this UIST 2022 best paper, the authors present a system called CrossA11y that helps video authors to efficiently locate and address visual and auditory accessibility problems in their videos.

Adobe researchers have co-authored four full papers (with one best paper award), one poster, and one demo for this year’s ACM Symposium on User Interface Software and Technology (UIST). UIST is the premier forum for innovations in human-computer interfaces, bringing together diverse research areas including graphical & web user interfaces, tangible & ubiquitous computing, virtual & augmented reality, multimedia, new input & output devices, and computer-supported cooperative work.

For the first time since 2019, UIST is being held in person, and will take place in Bend, Oregon, from October 29 to November 2. For those who cannot attend in person, all talks will be livestreamed to YouTube in real time, and Discord will be available for questions and discussion.

In addition to co-authoring publications, Adobe researchers have also contributed to UIST 2022 in many other ways, including serving on the program committee, serving on the organizing committee, and reviewing papers. Several former Adobe Research interns also served on the program committee and organizing committee. As a Silver Sponsor, Adobe will also host a sponsor booth at the demo session on Monday evening, where researchers will demo Project Blink and Project Primrose.

Many of these publications are the result of research internships or other collaborations with university students and faculty. Please check out Adobe Research’s careers page to learn more about internships and full-time career opportunities.

Full papers

CrossA11y: Identifying Video Accessibility Issues via Cross-modal Grounding (best paper award!)
Xingyu “Bruce” Liu, Ruolin Wang, Dingzeyu Li, Xiang Anthony Chen, Amy Pavel

Project Primrose: Reflective Light-Diffuser Modules for Non-Emissive Flexible Display Systems
Christine Dierk, TJ Rhodes, Gavin Miller

RealityTalk: Real-time Speech-driven Augmented Presentation for AR Live Storytelling
Jian Liao, Adnan Karim, Shivesh Singh Jadon, Rubaiat Habib Kazi, Ryo Suzuki

Record Once, Post Everywhere: Automatic Shortening of Audio Stories for Social Media
Bryan Wang, Zeyu Jin, Gautham Mysore


iMarker: Instant and True-to-scale AR with Invisible Markers
Chang Xiao, Ryan Rossi, Eunyee Koh


Point Cloud Capture and Editing for AR Environmental Design
Zeyu Wang, Cuong Nguyen, Paul Asente, Julie Dorsey

Organizing committee

Alexa Siu, Diversity Co-chair

Program committee

Stephen DiVerdi
Ailie Fraser
Rubaiat Habib Kazi
Hijung Valentina Shin

Related Posts