Roomify: Spatially-Grounded Style Transformation for Immersive Virtual Environments Conference: Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems; April 13–17, 2026; Barcelona, Spain Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems (CHI ’26), April 13–17, 2026, Barcelona, Spain DOI: 10.1145/3772318.3791803 ISBN: 979-8-4007-2278-3/2026/04 CCS: Human-centered computing Interactive systems and tools CCS: Human-centered computing Virtual reality
Xueyang Wang, Qinxuan Cen, Weitao Bi, Yunxiang Ma, Xin Yi, Robert Xiao, Xinyi Fu, Hewu Li · 2026
Read Roomify: Spatially-Grounded Style Transformation for Immersive Virtual Environments Conference: Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems; April 13–17, 2026; Barcelona, Spain Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems (CHI ’26), April 13–17, 2026, Barcelona, Spain DOI: 10.1145/3772318.3791803 ISBN: 979-8-4007-2278-3/2026/04 CCS: Human-centered computing Interactive systems and tools CCS: Human-centered computing Virtual reality by Xueyang Wang, Qinxuan Cen, Weitao Bi, Yunxiang Ma, Xin Yi, Robert Xiao, Xinyi Fu, Hewu Li (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-04917.txt.
Introduction
Xueyang Wang and colleagues present Roomify, a system turning a user's physical room into a themed virtual environment while keeping its spatial layout and furniture functions intact. Virtual reality usually forces a choice between full immersion and awareness of one's surroundings. The pipeline treats rooms as spatial containers, combining on-site 3D scene understanding with style-aware generation and an authoring tool for editing and previews. A study with 18 VR users found presence up 63% against passthrough and 26% against fully virtual baselines, with spatial awareness retained.
Catalog record
| CREATOR | Xueyang Wang, Qinxuan Cen, Weitao Bi, Yunxiang Ma, Xin Yi, Robert Xiao, Xinyi Fu, Hewu Li |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 87530 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.04917
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Roomify: Spatially-Grounded Style Transformation for Immersive Virtual Environments Conference: Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems; April 13–17, 2026; Barcelona, Spain Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems (CHI ’26), April 13–17, 2026, Barcelona, Spain DOI: 10.1145/3772318.3791803 ISBN: 979-8-4007-2278-3/2026/04 CCS: Human-centered computing Interactive systems and tools CCS: Human-centered computing Virtual reality Xueyang Wang, Qinxuan Cen, Weitao Bi, Yunxiang Ma, Xin Yi, Robert Xiao, Xinyi Fu, Hewu Li Abstract . We present Roomify, a spatially-grounded transformation system that generates themed virtual environments anchored to users’ physical rooms while maintaining spatial structure and functional semantics. Current VR approaches face a fundamental trade-off: full immersion sacrifices spatial awareness, while passthrough solutions break presence. Roomify addresses this through spatially-grounded transformation—treating physical spaces as “spatial containers” that preserve key functional and geometric properties of furniture while enabling radical stylistic changes. Our pipeline combines in-situ 3D scene understanding, AI-driven spatial reasoning, and style-aware generation to create personalized virtual environments grounded in physical reality. We introduce a cross-reality authoring tool enabling fine-grained user control through MR editing and VR preview workflows. Two user studies validate our approach: one with 18 VR users demonstrates a 63% improvement in presence over passthrough and 26% over fully virtual baselines while maintaining spatial awareness; another with 8 design professionals confirms the system’s creative expressiveness (scene quality: 5.95/7; creativity support: 6.08/7) and professional workflow value across diverse environments. † † cc-license: by Figure 1. The Roomify user journey from physical room to themed virtual environment. Users begin in their real space (a), scan the room geometry (b), review spatial understanding results (c), specify their desired style through multimodal input (d), observe and adjust real-time generation in MR (e), and finally immerse themselves in the transformed environment in VR (f). The system preserves spatial layout and functional semantics of furniture while enabling radical stylistic transformation, allowing users to inhabit fantastical worlds grounded in their familiar physical space. 1. Introduction Virtual reality headsets are increasingly adopted in home environments for immersive entertainment and creative professional practices. However, this immersion comes at a fundamental cost: VR systems typically isolate users from their physical surroundings, creating spatial disorientation and safety risks ( 17 ; 71 ; 40 ) . Empirical studies demonstrate that users frequently breach safety boundaries even when visible warnings are present ( 68 ) , while research reveals that incorporating physical correspondence into VR experiences significantly improves both user experience and task performance ( 13 ; 8 ; 63 ) . For creative professionals, spatially-grounded immersive environments enhance understanding of spatial qualities like size, scale, and circulation ( 77 ; 66 ; 10 ) , with designers benefiting from MR-based on-site visualization and in-situ authoring tools ( 16 ; 44 ) . These phenomena highlight the significant role of spatial awareness in VR experiences: users must continuously know where their bodies and surrounding objects are while acting in a virtual scene. Studies confirm that spatial awareness is influenced by geometric features, semantic information, and spatial layouts—and when these sources are integrated, awareness is significantly enhanced ( 19 ; 6 ; 2 ) . For home entertainment, maintaining spatial awareness becomes inseparable from safety: virtual experiences must render furniture and boundaries legible as part of the spatial experience rather than as sporadic warnings ( 9 ) . Current approaches to bridging physical and virtual worlds face a fundamental design dilemma: a trade-off between making spatial structure continuously perceptible and preserving uninterrupted presence ( 9 ) . Traditional boundary systems like Guardian provide only 2D perimeter warnings ( 79 ) , failing to represent actual room geometry or furniture placement. Video passthrough solutions enable spatial awareness ( 80 ; 73 ; 26 ) but break immersion by constantly reminding users they are wearing a headset. Scene proxy methods that represent physical objects with virtual counterparts ( 46 ; 65 ; 63 ; 82 ) maintain visual immersion but struggle with flexibility—either transforming only walkable areas ( 65 ; 12 ) , retrieving from limited asset libraries ( 63 ; 41 ; 39 ) , or applying surface stylization without geometric freedom ( 82 ; 64 ) . We present Roomify, a spatially-grounded transformation system that addresses this trade-off by achieving balance across style diversity and consistency, spatial alignment, functional consistency, and user editability . Our approach generates themed virtual environments anchored to users’ physical rooms while maintaining spatial structure and functional semantics of furniture. Rather than claiming to preserve full physical affordances, Roomify maintains each object’s approximate volume, footprint, and major contact surfaces while keeping its high-level role (e.g., seating, storage, temperature regulation) consistent with the physical room. This combination supports spatial awareness and gross locomotion and body-support actions (e.g., walking around obstacles, sitting, leaning). The system implements an end-to-end pipeline integrating room scanning ( 48 ) , AI-driven spatial understanding ( 49 ) , style-aware prompt generation, and coordinated image/3D generation with intelligent scene assembly. To address registration errors while providing fine-grained creative control, we developed a cross-reality authoring tool leveraging complementary MR and VR modalities ( 71 ; 4 ) . Users manipulate wireframe scaffolds and provide multimodal style specifications in MR mode for spatially-grounded refinement, then seamlessly transition to VR mode for immersive preview. This dual-mode approach enables creators to maintain both spatial constraints and creative vision throughout the authoring process. Our evaluation through two user studies with 18 general VR users and 8 design professionals validates the system across diverse contexts—from themed entertainment (gaming, immersive movie viewing) to creative prototyping (interior design exploration, client visualization, media storyboarding). Roomify achieves 63% improvement in presence over passthrough and 26% over fully virtual baselines, while maintaining significantly better spatial awareness than fully virtual approaches. Design professionals rated scene quality at 5.95/7 compared to 4.50 for text-to-3D and 3.41 for AI re-texturing, with creativity support scores of 6.08/7. This work makes three primary contributions: • A spatially-grounded generation pipeline that balances stylistic flexibility with geometric preservation and functional consistency, enabling personalized room transformation while maintaining spatial alignment through AI-driven spatial understanding and style-aware 3D generation. • A cross-reality authoring tool combining MR’s spatial grounding with VR’s immersive preview, providing intuitive controls for iterative refinement of room-scale transformation workflows. • Comprehensive empirical validation through two user studies examining immersion, spatial awareness, and creative expressiveness, demonstrating significant improvements in presence while maintaining spatial alignment compared to existing approaches. 2. Related Work 2.1. Integrating Physical and Virtual Environments Virtual reality systems face a fundamental tension between achieving deep immersion and maintaining spatial awareness in physical environments ( 17 ; 71 ) . This challenge has driven researchers to explore various integration strategies. Boundary and Passthrough Systems. Commercial systems like Oculus Guardian display grid overlays when users approach play area limits ( 79 ) . However, users frequently breach boundaries during rapid movements, and these non-diegetic overlays disrupt immersion ( 68 ; 26 ) . Video passthrough solutions enable spatial awareness and object manipulation ( 80 ; 73 ; 26 ; 51 ; 17 ; 23 ) but fundamentally break immersion—a “dream collapsing” effect ( 43 ) . Selective approaches like RealityCheck ( 26 ) and RealityLens ( 73 ) blend physical objects into virtual scenes, but treat physical