Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

From Reality to Virtual Worlds: The Role of Photogrammetry in Game Development

Santiago Berrezueta-Guzman, Andrei Koshelev, Stefan Wagner

TL;DR

The study evaluates RealityCapture, a GPU-accelerated photogrammetry tool, for VR game development by comparing it to manual low-poly modeling in Blender. It analyzes efficiency, reconstruction accuracy, and Unreal Engine integration, using a dataset of 1,550 images and a 10-participant user study. Key findings show photogrammetry delivers superior realism but higher mesh complexity and texture refinement needs, while manual modeling offers speed and real-time performance with lower visual fidelity. The work underscores a trade-off between visual fidelity and interactivity, recommending hybrid workflows and future AI/cloud-based enhancements to broaden accessibility. The results inform asset pipelines for VR and AR, highlighting how to balance realism, performance, and production time in game development.

Abstract

Photogrammetry is transforming digital content creation by enabling the rapid conversion of real-world objects into highly detailed 3D models. This paper evaluates the role of RealityCapture, a GPU-accelerated photogrammetry tool, in game development of Virtual Reality (VR). We assess its efficiency, reconstruction accuracy, and integration with Unreal Engine, comparing its advantages and limitations against traditional modeling workflows. Additionally, we examined user preferences between designed 3D assets and photogrammetry-generated models. The results revealed that while photogrammetry enhances realism and interactivity, users slightly preferred manually designed models for small, manipulable elements because of the level of detail. However, from a developer perspective, RealityCapture significantly reduces development time while maintaining geometric precision and photorealistic textures. Despite its reliance on high-performance hardware, its automation, scalability, and seamless integration with real-time rendering engines make it a valuable tool for game developers and VR creators. Future improvements in AI-driven optimization and cloud-based processing could enhance accessibility, broadening its applications in gaming, cultural heritage preservation, and simulation.

From Reality to Virtual Worlds: The Role of Photogrammetry in Game Development

TL;DR

The study evaluates RealityCapture, a GPU-accelerated photogrammetry tool, for VR game development by comparing it to manual low-poly modeling in Blender. It analyzes efficiency, reconstruction accuracy, and Unreal Engine integration, using a dataset of 1,550 images and a 10-participant user study. Key findings show photogrammetry delivers superior realism but higher mesh complexity and texture refinement needs, while manual modeling offers speed and real-time performance with lower visual fidelity. The work underscores a trade-off between visual fidelity and interactivity, recommending hybrid workflows and future AI/cloud-based enhancements to broaden accessibility. The results inform asset pipelines for VR and AR, highlighting how to balance realism, performance, and production time in game development.

Abstract

Photogrammetry is transforming digital content creation by enabling the rapid conversion of real-world objects into highly detailed 3D models. This paper evaluates the role of RealityCapture, a GPU-accelerated photogrammetry tool, in game development of Virtual Reality (VR). We assess its efficiency, reconstruction accuracy, and integration with Unreal Engine, comparing its advantages and limitations against traditional modeling workflows. Additionally, we examined user preferences between designed 3D assets and photogrammetry-generated models. The results revealed that while photogrammetry enhances realism and interactivity, users slightly preferred manually designed models for small, manipulable elements because of the level of detail. However, from a developer perspective, RealityCapture significantly reduces development time while maintaining geometric precision and photorealistic textures. Despite its reliance on high-performance hardware, its automation, scalability, and seamless integration with real-time rendering engines make it a valuable tool for game developers and VR creators. Future improvements in AI-driven optimization and cloud-based processing could enhance accessibility, broadening its applications in gaming, cultural heritage preservation, and simulation.

Paper Structure

This paper contains 16 sections, 3 figures, 1 table.

Table of Contents

  1. Introduction
  2. RealityCapture
  3. Technical Specifications
  4. Related Work
  5. Methodology
  6. Modeling Approaches
  7. Comparison Criteria
  8. User Testing and Preference Analysis
  9. Results
  10. Photogrammetry using RealityCapture
  11. Manual Low-Poly Modeling
  12. Integration and Performance Evaluation
  13. User Testing and Preference Analysis
  14. Discussion
  15. Recommendations
  16. ...and 1 more sections

Figures (3)

  • Figure 1: Workflow of image acquisition, RealityCapture-based model creation, the integration in Unreal Engine, and testing process.
  • Figure 2: Comparison among (Left) Original photograph of the object. (Center) Reconstructed 3D model using RealityCapture. (Right) The designed model (manually) in Blender.
  • Figure 3: Comparison of different 3D modeling approaches: (Left) Real-world photo of the original sculpture. (Center) RealityCapture-generated 3D model integrated into the virtual environment. (Right) A manually designed 3D model was created in Blender and placed in the same virtual setting.