Game development projects often evolve over time as new technologies, collaborators, and research opportunities emerge. This paper looks at the evolution of Anonymized Game, which is an educational game that helps to teach the concept of recursion. Over the past years, the project expanded from being a small desktop game to a multi-institutional research platform that collects data on how the game is being played. In addition to technical development, the project incorporated interdisciplinary collaboration with students in the music and art departments, while maintaining a mentoring-based development model that emphasized student involvement. As the project expanded to support desktop, tablet, and headset augmented reality platforms, the development team faced several challenges, including coordinating work across institutions and time zones, managing code branches and testing environments, and adapting to rapidly evolving technologies. This paper discusses these challenges and describes lessons learned from the development process. In particular, we highlight strategies such as platform-dependent scripting, iterative planning using Scrum-inspired practices, and the development of tooling to support faster testing and evaluation cycles. These insights provide practical guidance for long-term, student-driven game development projects that aim to support both educational games and research initiatives.

This poster reports a follow-up study of the Computer Graphics Wizard Academy (CGWA) that evaluates its second iteration, introducing three story modes with learner autonomy to stay in one mode or switch at any time. This study examines which modes students prefer and how switching unfolds by combining web analytics (mode choice, time-in-mode, and switching patterns) with post-trial surveys administered after two trials. Surveys include paired-comparison and two validated measures: the User Engagement Scale-Short Form (UES-SF) and the Transportation Scale-Short Form (TS-SF). The goal is to identify story-mode designs that sustain engagement while maintaining clarity in a narrative-driven computer graphics learning experience.

Teaching dynamic computing concepts, such as the quicksort's partitioning step and linked list pointer updates, can be challenging for students to visualize and interact with in real time. Augmented reality (AR) can make these processes tangible by overlaying immediate feedback on the physical objects, but commercial optical see-through head-mounted displays (OST-HMDs) remain costly and closed-source, limiting the adoption within typical classrooms. To address this, we adapted a low-cost OST-HMD open-source project (Project Northstar) for computing education. This in-house OST-HMD uses 3D-printed parts and commodity components, dual micro-displays, and tracking sensors. We documented our experience and provided a practical and repeatable workflow for mechanical assembly and optical calibration, including a troubleshooting guide for commonly faced problems such as how to calibrate the optics for a stable virtual–real alignment. The current prototype operates as a secondary display driven by a host laptop, and software for interactive activities is in active development. Once the software is completed, we will pilot AR activities that walk students through quicksort partitioning and step-by-step linked list pointer updates. We will evaluate learning and usability using pre- and post-quizzes, task-completion times, and a post-task usability survey. This poster contributes (1) a reproducible bill of materials with transparent cost breakdown, (2) 3D-printable models and wiring diagrams, and (3) a calibration workflow with representative results. We aim to provide an open-source replicable OST-HMD build to lower financial and technical barriers to AR adoption in computer science (CS) courses.

Traditional final exams in introductory computer science courses often rely on one-size-fits-all testing, which may not support individual learning needs. Research on mastery learning suggests that students learn more effectively when focusing on weak areas and actively constructing knowledge. This poster presents a teaching-based final exam that promotes individualized learning. Students are assigned a challenging topic related to their weaknesses and tasked with preparing a mini-lecture, designing a quiz, and creating an exam-level problem with a rubric.

Abstract Screened Poisson Surface Reconstruction creates 2D surfaces from sets of oriented points in 3D (and can be extended to co‐dimension one surfaces in arbitrary dimensions). In this work we generalize the technique to manifolds of co‐dimension larger than one. The reconstruction problem consists of finding a vector‐valued function whose zero set approximates the input points. We argue that the right extension of screened Poisson Surface Reconstruction is based on exterior products: the orientation of the point samples is encoded as the exterior product of the local normal frame. The goal is to find a set of scalar functions such that the exterior product of their gradients matches the exterior products prescribed by the input points. We show that this setup reduces to the standard formulation for co‐dimension 1, and leads to more challenging multi‐quadratic optimization problems in higher co‐dimension. We explicitly treat the case of co‐dimension 2, i.e., curves in 3D and 2D surfaces in 4D. We show that the resulting bi‐quadratic problem can be relaxed to a set of quadratic problems in two variables and that the solution can be made effective and efficient by leveraging a hierarchical approach.

Abstract We describe a novel approach that addresses the problem of establishing correspondences between non‐rigidly deformed shapes by performing the registration over the unit sphere. In a pre‐processing step, each shape is conformally parametrized over the sphere, centered to remove Möbius inversion ambiguity, and authalically evolved to expand regions that are excessively compressed by the conformal parametrization. Then, for each pair of shapes, we perform fast SO(3) correlation to find the optimal rotational alignment and refine the registration using optical flow. We evaluate our approach on the TOSCA dataset, demonstrating that our approach compares favorably to state‐of‐the‐art methods.