Projects

My work spans collaborative XR research, human-centered computing, applied machine learning, and interactive systems. The page is organized to show that progression: current PhD research first, followed by earlier research projects, human-centered applications, and systems/computer-vision work.

PhD Research

Gaze behavior of triads in varying task demands (Ongoing)

Developing a technical prototype on Samsung Galaxy XR using Photon Fusion to connect three players in a dynamic 360° environment. The prototype explores varying task-pressure levels through changes in flying direction. Additional technical and study details are not yet public.

Code: Private GitHub repository — available on request.

Gaze and Voice Communication in Collaborative MR (Under Review, IEEE VR 2027)

How does communication modality shape what collaborators notice, understand, and anticipate together?

I led an end-to-end investigation of collective situational awareness in asymmetric mixed reality, designing and implementing a networked HoloLens 2 experiment and refining the protocol through a 12-participant pilot before conducting an IRB-approved, counterbalanced study with 48 participants across 24 dyads.

The work extends SAGAT-based collective situation-awareness assessment with team-level measures of information coverage and accurate alignment, alongside correctness and response-alignment measures. I also built multimodal analysis pipelines integrating gaze, speech, SAGAT, workload, and performance data. The study found that Gaze+Voice improved task performance and shared awareness, while Gaze-only produced the strongest spatial gaze convergence, suggesting that performance, co-attention, and collective awareness capture distinct aspects of collaboration. The manuscript is under review at IEEE VR 2027.

Code: CogniLink on GitHub

Security Attacks in Remote Collaborative MR (IEEE VR 2025)

How do collaborators recognize and respond to attacks in shared immersive environments?

I designed and executed a mixed-methods security study on remote collaborative MR, iteratively refining the task and attack conditions across two pilot studies before conducting an IRB-approved main study with 20 participants across 10 dyads. I engineered four real-time attacks in a networked Unity/HoloLens 2 platform, including a novel click-redirection mechanism and MR-specific occlusion attacks.

By triangulating first-person recordings, behavioral observations, questionnaires, and interviews, the study uncovered gaps between users’ perceived and actual security behavior. The resulting design recommendations focus on immersive security indicators, interaction transparency, and user training. This work resulted in my first-author IEEE VR 2025 paper and was featured in Computerworld.

Code: MRattack on GitHub
Publication: IEEE VR 2025 Paper · Computerworld Feature
Presentations: IEEE VR 2025 — presentation recap · CCI Symposium 2025 — official presentation listing

Multimodal Privacy Indicators Under Situational Impairment (IEEE ISMAR Adjunct 2025)

How should XR systems communicate privacy-relevant information when visual cues alone are insufficient?

I collaborated on an iterative mixed-methods study of privacy indicators for situationally impaired XR bystanders. Four focus groups with 8 participants informed five visual and multimodal concepts, followed by a 7-participant evaluation of usability and contextual usefulness. The results showed limitations of visual-only cues under situational impairment and stronger preferences for multimodal indicators in privacy-sensitive scenarios. This work was published at IEEE ISMAR Adjunct 2025.

Publication: IEEE ISMAR Adjunct 2025 Paper · NSF Public Access Record

Research Projects

Cross-Verification of Revenue Using Remote Sensing Data: Code (GitHub) · Technical Report (International Growth Centre)

Human-Centered Applications

Hunehar Management System: Code (GitHub)

Systems & Computer Vision

Selected code: Airport Baggage Management System — GitHub