VEGETATION REALISTIC VISUALIZATION IN VIRTUAL REALITY SCENARIOS
This paper focuses on optimizing realistic 3D vegetation models for standalone virtual reality experiences considering the visualization of the plant growth over time. Most of the existing solutions are focusing on personal computer based virtual reality utilization, which allows for more computational power, yet it restricts the movement of the user. Experiences meant for standalone virtual reality headsets mainly feature low-poly and unrealistic plant representations. Since plants, especially trees, can be rather complex structures it is important to find a balance between realistic depiction and performance. By achieving this it would allow the headset user to have better immersion in the virtual world and would reduce the risks of cybersickness. To test and find out the most appropriate solution we created a 3D model of an apple tree and its growth animation in SpeedTree software and then created a scene of varying numbers of trees using Unreal Engine 5.3 and tested framerate count using Meta Quest 2 and OVR Metrics tool. Tests were split into two big groups – static mesh tests, where we observed the changes in framerate before and after optimizing and skeletal mesh tests, where framerate changes were noted as the tree growth animation was running. The results of the tests showcased that it might be difficult to visualize growth process of the plants by using animated models, since their performance was way below the target rate for the headset, while using static meshes and swapping the models over time to simulate growth yielded a better framerate.
Authors
- Lauris Taube
- Arnis Cīrulis (ORCID: https://orcid.org/0000-0001-9577-0646)
- Laura Ozoliņa
Institutions
- Vidzeme University of Applied Sciences (LV)
Publication Details
- Journal
- Environment Technology Resources Proceedings of the International Scientific and Practical Conference
- Published
- 2025-06-08
- DOI
- https://doi.org/10.17770/etr2025vol2.8584
- Primary Topic
- 3D Modeling in Geospatial Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Vidzemes Augstskola