Return to Selected Work


Research Pixelate2026



Project Members:
Joseph C. Kenny (PhD Student, Project Lead), Dr. Samuelle Bourgault, Philippe Maman, Soulaf Aburas
Prof. Dr. Daniela Mitterberger

Collaborators:
Populous Applied R&D

Photo & Video:
Samuelle Bourgault, Charlie Burt, & Keith Zhang

Event Information:
Date: April 2nd, 2026
Location: Populous Headquarters, Kansas City, MO
Description: The event included a 1-hour presentation with an overview of the Pixelate project, including its research context, system design, and approach to multi-user interaction on public displays. Further, it was followed by a 2-hour interactive demonstration session, during which attendees from all across North America engaged with the system directly in a hybrid online/in-person session with the system.

Publication (Poster) :
J. C. Kenny, S. Bourgault, P. Maman, S. Aburas, and D. Mitterberger, “Poster: Pixelate: An Event-Driven System for Scalable Distributed Multi-User Interaction on Public Displays,” in Proceedings of the 20th ACM International Conference on Distributed and Event-based Systems, Universidade de Lisboa Lisbon Portugal: ACM, Jun. 2026, pp. 175–177. doi: 10.1145/3809481.3816719.
Pixelate explores scalable multi-user interaction on public displays through a platform-independent collaborative drawing system. Developed as a distributed, event-driven application, the project examines how large numbers of co-located and remote users can interact with a persistent shared canvas across different devices, locations, and display configurations.

The system supports real-time synchronization, persistent shared state, and application-level concurrency control across web, Android, and iOS clients. Users select, modify, or erase pixels within a shared canvas, while transient locking and event-based reconciliation coordinate simultaneous interactions and resolve conflicts between distributed participants.

Pixelate was evaluated through a public deployment spanning at least four cities and three time zones, connecting 75 unique clients through personal devices, tablets, phones, laptops, and mixed-reality headsets. At peak use, the system supported 69 concurrent users and 48 simultaneous selections, demonstrating an approach to persistent, cross-platform collaboration on large-scale public interactive displays.

Pixelate application deployed across three device contexts. (A) Running the application on a mobile phone. (B) Running the application in a desktop environment. (C) Running the application through a head-mounted display.



 Interface and interaction modes of the application. (A) The primary user interface, including the shared pixel canvas,
interaction tools, color controls, action history, and interface legend. (B) Randomized interaction, which selects pixels according to a stochastic pattern. (C) Linear incremental selection, which extends selection step-by-step across adjacent pixels. (D) Free-form touch selection, allowing users to directly mark pixels through touch input. (E) Multi-user selection, showing simultaneous local and remote selections within the shared canvas.
Configuration and runtime event management of
the system data model. ScreenData stores persistent session
and pixel-grid state, UserCurrentSelection manages tran-
sient selection locks, and InteractionLog records append-
only interaction events used to synchronize updates across
connected clients.
Collaborative output diversity across the shared display. (A) Final display state showing the aggregated result of all user contributions. (B), (C), and (F) show outputs composed through contributions from multiple users. (D) and (E) show single-user contributions, illustrating individually produced forms within the broader collaborative canvas
Per-minute event counts over the 1-hour 18-minute session: 4,413 interaction events (green) and 9,402 selection events (red), showing initial peak activity followed by sustained engagement.



Index ©2047 All Rights Reserved