Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of...
Transcript of Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of...
![Page 1: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/1.jpg)
AutoBiomes
Roland Fischer, Philipp Dittmann, René Weller, Gabriel Zachmann
University of Bremen, Germany
CGI, 20-23 October 2020, Geneva
Procedural Generation of Multi-Biome Landscapes
![Page 2: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/2.jpg)
The Good Dinosaur [Disney/Pixar]
Motivation
Introduction Previous Work Overview Details Results Conclusion 1
![Page 3: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/3.jpg)
Ghost Recon: Wildlands [Ubisoft]The Good Dinosaur [Disney/Pixar]
Motivation
Introduction Previous Work Overview Details Results Conclusion 1
![Page 4: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/4.jpg)
Ghost Recon: Wildlands [Ubisoft]The Good Dinosaur [Disney/Pixar]
Motivation
Introduction Previous Work Overview Details Results Conclusion 1
VaMEx-VTB [University of Bremen]
![Page 5: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/5.jpg)
Motivation
• Huge landscapes as combination of different biomes
Introduction Previous Work Overview Details Results Conclusion 2
Minecraft
![Page 6: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/6.jpg)
Motivation
• Huge landscapes as combination of different biomes
• Populating the terrain with objects
Introduction Previous Work Overview Details Results Conclusion 2
Minecraft
![Page 7: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/7.jpg)
Motivation
• Huge landscapes as combination of different biomes
• Populating the terrain with objects
• Manual creation is not an option
Introduction Previous Work Overview Details Results Conclusion 2
Minecraft
![Page 8: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/8.jpg)
Motivation
• Huge landscapes as combination of different biomes
• Populating the terrain with objects
• Manual creation is not an option
• Procedural terrain generation (PTG)
Introduction Previous Work Overview Details Results Conclusion 2
Minecraft
![Page 9: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/9.jpg)
Motivation
• Huge landscapes as combination of different biomes
• Populating the terrain with objects
• Manual creation is not an option
• Procedural terrain generation (PTG)
• Much researched, still open challenges
Introduction Previous Work Overview Details Results Conclusion 2
Realism Speed
Usability
Flexibility
Minecraft
![Page 10: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/10.jpg)
Related Work
• Main PTG approaches:
• Synthetic, e.g. noise
• Fast, unintuitive, hard to get realistic results [Thorimbert18]
Introduction Previous Work Overview Details Results Conclusion 3
[Thorimbert18]
![Page 11: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/11.jpg)
Related Work
• Main PTG approaches:
• Synthetic, e.g. noise
• Fast, unintuitive, hard to get realistic results [Thorimbert18]
• Physics-based, e.g. erosion, fluid simulation
• Complex, realistic results, slow [Stam03, Jákó11, Ihmsen14]
Introduction Previous Work Overview Details Results Conclusion 3
[Thorimbert18]
[Jákó11]
![Page 12: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/12.jpg)
Related Work
• Main PTG approaches:
• Synthetic, e.g. noise
• Fast, unintuitive, hard to get realistic results [Thorimbert18]
• Physics-based, e.g. erosion, fluid simulation
• Complex, realistic results, slow [Stam03, Jákó11, Ihmsen14]
• Example-based, e.g. image synthesis, DEMs, neural networks
• Realistic, good usability, inflexible [Zhou07, Beckham17, Wulff-Jensen18]
Introduction Previous Work Overview Details Results Conclusion 3
[Thorimbert18]
[Jákó11]
[Zhou07]
![Page 13: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/13.jpg)
Related Work
• Main PTG approaches:
• Synthetic, e.g. noise
• Fast, unintuitive, hard to get realistic results [Thorimbert18]
• Physics-based, e.g. erosion, fluid simulation
• Complex, realistic results, slow [Stam03, Jákó11, Ihmsen14]
• Example-based, e.g. image synthesis, DEMs, neural networks
• Realistic, good usability, inflexible [Zhou07, Beckham17, Wulff-Jensen18]
• Also valid for commercial tools (e.g. World Creator, World Machine, Terragen)
Introduction Previous Work Overview Details Results Conclusion 3
[Thorimbert18]
[Jákó11]
[Zhou07]
![Page 14: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/14.jpg)
Our Contributions
• Novel PTG system combining advantages of 3 approaches
Introduction Previous Work Overview Details Results Conclusion 4
![Page 15: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/15.jpg)
Our Contributions
• Novel PTG system combining advantages of 3 approaches
• Effective generation of vast, plausible-looking terrains
Introduction Previous Work Overview Details Results Conclusion 4
![Page 16: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/16.jpg)
Our Contributions
• Novel PTG system combining advantages of 3 approaches
• Effective generation of vast, plausible-looking terrains
• Multi-biome landscapes
Introduction Previous Work Overview Details Results Conclusion 4
![Page 17: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/17.jpg)
Our Contributions
• Novel PTG system combining advantages of 3 approaches
• Effective generation of vast, plausible-looking terrains
• Multi-biome landscapes
• Dense, complex asset distribution
Introduction Previous Work Overview Details Results Conclusion 4
![Page 18: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/18.jpg)
Our Contributions
• Novel PTG system combining advantages of 3 approaches
• Effective generation of vast, plausible-looking terrains
• Multi-biome landscapes
• Dense, complex asset distribution
• Easy-to-use, iterative workflow
Introduction Previous Work Overview Details Results Conclusion 4
![Page 19: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/19.jpg)
Our Contributions
• Novel PTG system combining advantages of 3 approaches
• Effective generation of vast, plausible-looking terrains
• Multi-biome landscapes
• Dense, complex asset distribution
• Easy-to-use, iterative workflow
• Unreal Engine 4 integration
Introduction Previous Work Overview Details Results Conclusion 4
![Page 20: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/20.jpg)
Our PTG Pipeline
Introduction Previous Work Overview Details Results Conclusion 5
Rough Base-Terrain
Climate Simulation
Biome-based Terrain
Refinement
Asset Placement
Terrain + Asset
Distribution
User
Parameters
![Page 21: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/21.jpg)
Our PTG Pipeline
Introduction Previous Work Overview Details Results Conclusion 5
Rough Base-Terrain
Climate Simulation
Biome-based Terrain
Refinement
Asset Placement
Terrain + Asset
Distribution
User
Parameters
Noise
![Page 22: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/22.jpg)
Our PTG Pipeline
Introduction Previous Work Overview Details Results Conclusion 5
Rough Base-Terrain
Climate Simulation
Biome-based Terrain
Refinement
Asset Placement
Terrain + Asset
Distribution
User
Parameters
Temperature Wind Biomes
Noise
Precipitation
![Page 23: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/23.jpg)
Our PTG Pipeline
Introduction Previous Work Overview Details Results Conclusion 5
Rough Base-Terrain
Climate Simulation
Biome-based Terrain
Refinement
Asset Placement
Terrain + Asset
Distribution
User
Parameters
Temperature Wind Biomes
Noise DEMs
Precipitation
![Page 24: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/24.jpg)
Our PTG Pipeline
Introduction Previous Work Overview Details Results Conclusion 5
Rough Base-Terrain
Climate Simulation
Biome-based Terrain
Refinement
Asset Placement
Terrain + Asset
Distribution
User
Parameters
Temperature Wind Biomes
Noise Local-to-Global
DEMs
Precipitation
![Page 25: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/25.jpg)
Our PTG Pipeline
Introduction Previous Work Overview Details Results Conclusion 5
Rough Base-Terrain
Climate Simulation
Biome-based Terrain
Refinement
Asset Placement
Terrain + Asset
Distribution
User
Parameters
Temperature Wind Biomes
Noise Local-to-Global
DEMs
Precipitation
Synthetic
![Page 26: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/26.jpg)
Our PTG Pipeline
Introduction Previous Work Overview Details Results Conclusion 5
Rough Base-Terrain
Climate Simulation
Biome-based Terrain
Refinement
Asset Placement
Terrain + Asset
Distribution
User
Parameters
Temperature Wind Biomes
Noise Local-to-Global
DEMs
Precipitation
Synthetic Physics-based
![Page 27: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/27.jpg)
Our PTG Pipeline
Introduction Previous Work Overview Details Results Conclusion 5
Rough Base-Terrain
Climate Simulation
Biome-based Terrain
Refinement
Asset Placement
Terrain + Asset
Distribution
User
Parameters
Temperature Wind Biomes
Noise Local-to-Global
DEMs
Precipitation
Synthetic Physics-based Example-based
![Page 28: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/28.jpg)
Our PTG Pipeline
Introduction Previous Work Overview Details Results Conclusion 5
Rough Base-Terrain
Climate Simulation
Biome-based Terrain
Refinement
Asset Placement
Terrain + Asset
Distribution
User
Parameters
Temperature Wind Biomes
Noise Local-to-Global
DEMs
Precipitation
Synthetic Physics-based Example-based Mainly Synthetic
![Page 29: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/29.jpg)
Our PTG Pipeline
Introduction Previous Work Overview Details Results Conclusion 5
Rough Base-Terrain
Climate Simulation
Biome-based Terrain
Refinement
Asset Placement
Terrain + Asset
Distribution
User
Parameters
Temperature Wind Biomes
Noise Local-to-Global
DEMs
Precipitation
![Page 30: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/30.jpg)
Our PTG Pipeline
Introduction Previous Work Overview Details Results Conclusion 5
• Each step is customizable and repeatable
Rough Base-Terrain
Climate Simulation
Biome-based Terrain
Refinement
Asset Placement
Terrain + Asset
Distribution
User
Parameters
Temperature Wind Biomes
Noise Local-to-Global
DEMs
Precipitation
![Page 31: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/31.jpg)
Our PTG Pipeline
Introduction Previous Work Overview Details Results Conclusion 5
• Each step is customizable and repeatable
Rough Base-Terrain
Climate Simulation
Biome-based Terrain
Refinement
Asset Placement
Terrain + Asset
Distribution
User
Parameters
Temperature Wind Biomes
Noise Local-to-Global
DEMs
Precipitation
![Page 32: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/32.jpg)
Our PTG Pipeline
Introduction Previous Work Overview Details Results Conclusion 5
• Each step is customizable and repeatable
• Direct proxy visualization for quick workflow
Rough Base-Terrain
Climate Simulation
Biome-based Terrain
Refinement
Asset Placement
Terrain + Asset
Distribution
User
Parameters
Temperature Wind Biomes
Noise Local-to-Global
DEMs
Precipitation
![Page 33: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/33.jpg)
Our PTG Pipeline
Introduction Previous Work Overview Details Results Conclusion 5
• Each step is customizable and repeatable
• Direct proxy visualization for quick workflow
Rough Base-Terrain
Climate Simulation
Biome-based Terrain
Refinement
Asset Placement
Terrain + Asset
Distribution
User
Parameters
Temperature Wind Biomes
Noise Local-to-Global
DEMs
Precipitation
![Page 34: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/34.jpg)
Base Terrain Generation
Introduction Previous Work Overview Details Results Conclusion
• Multiple octaves of simplex noise
6
![Page 35: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/35.jpg)
Base Terrain Generation
Introduction Previous Work Overview Details Results Conclusion
• Multiple octaves of simplex noise
• Lots of parameters for flexibility, fast to compute, scalable
• Other noise combinations possible
6
![Page 36: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/36.jpg)
Base Terrain Generation
Introduction Previous Work Overview Details Results Conclusion
• Multiple octaves of simplex noise
• Lots of parameters for flexibility, fast to compute, scalable
• Other noise combinations possible
• Only serves as rough starting terrain, refined later
• No tedious fine-tuning needed
6
Base terrain, water bodies in blue
![Page 37: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/37.jpg)
Base Terrain Generation
Introduction Previous Work Overview Details Results Conclusion
• Multiple octaves of simplex noise
• Lots of parameters for flexibility, fast to compute, scalable
• Other noise combinations possible
• Only serves as rough starting terrain, refined later
• No tedious fine-tuning needed
• Easily extendable with sketch-basedediting techniques
6
Base terrain, water bodies in blue
![Page 38: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/38.jpg)
• Two adjustable interpolation modes:
• Bi-linear interpolation
Climate Simulation - Temperature
Introduction Previous Work Overview Details Results Conclusion 7
Temperature, blue = cold, red = hot
-15 32
25 35
![Page 39: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/39.jpg)
• Two adjustable interpolation modes:
• Bi-linear interpolation
• Sine-based interpolation
Climate Simulation - Temperature
Introduction Previous Work Overview Details Results Conclusion 7
Temperature, blue = cold, red = hot
-15 32
25 35
![Page 40: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/40.jpg)
• Two adjustable interpolation modes:
• Bi-linear interpolation
• Sine-based interpolation
• Adjustable altitude-based decline
Climate Simulation - Temperature
Introduction Previous Work Overview Details Results Conclusion 7
Temperature, blue = cold, red = hot
-15 32
25 35
![Page 41: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/41.jpg)
Climate Simulation - Wind
Introduction Previous Work Overview Details Results Conclusion
• Prevailing wind for moisture distribution
• Iterative, simplified semi-Lagrangian approach
8
![Page 42: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/42.jpg)
Climate Simulation - Wind
Introduction Previous Work Overview Details Results Conclusion
• Prevailing wind for moisture distribution
• Iterative, simplified semi-Lagrangian approach
• Only self advection and external forces
8
![Page 43: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/43.jpg)
Climate Simulation - Wind
Introduction Previous Work Overview Details Results Conclusion
• Prevailing wind for moisture distribution
• Iterative, simplified semi-Lagrangian approach
• Only self advection and external forces
• User specified, persistent external forces on corners
8
0
![Page 44: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/44.jpg)
1
Climate Simulation - Wind
Introduction Previous Work Overview Details Results Conclusion
• Prevailing wind for moisture distribution
• Iterative, simplified semi-Lagrangian approach
• Only self advection and external forces
• User specified, persistent external forces on corners
• Iteratively averaging wind vectors with adjacent ones in forward direction
8
![Page 45: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/45.jpg)
2
Climate Simulation - Wind
Introduction Previous Work Overview Details Results Conclusion
• Prevailing wind for moisture distribution
• Iterative, simplified semi-Lagrangian approach
• Only self advection and external forces
• User specified, persistent external forces on corners
• Iteratively averaging wind vectors with adjacent ones in forward direction
8
![Page 46: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/46.jpg)
2
Climate Simulation - Wind
Introduction Previous Work Overview Details Results Conclusion
• Prevailing wind for moisture distribution
• Iterative, simplified semi-Lagrangian approach
• Only self advection and external forces
• User specified, persistent external forces on corners
• Iteratively averaging wind vectors with adjacent ones in forward direction
• Enables creation of smooth, believable prevailing wind currents
8
Wind vector field
![Page 47: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/47.jpg)
Climate Simulation - Precipitation
Introduction Previous Work Overview Details Results Conclusion
• Iterative computation
9
![Page 48: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/48.jpg)
Climate Simulation - Precipitation
Introduction Previous Work Overview Details Results Conclusion
• Iterative computation
• Water bodies/world borders as moisture sources
9
![Page 49: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/49.jpg)
Climate Simulation - Precipitation
Introduction Previous Work Overview Details Results Conclusion
• Iterative computation
• Water bodies/world borders as moisture sources
• Temperature-dependent evaporation
9
![Page 50: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/50.jpg)
Climate Simulation - Precipitation
Introduction Previous Work Overview Details Results Conclusion
• Iterative computation
• Water bodies/world borders as moisture sources
• Temperature-dependent evaporation
• Wind distributes moisture
9
![Page 51: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/51.jpg)
Climate Simulation - Precipitation
Introduction Previous Work Overview Details Results Conclusion
• Iterative computation
• Water bodies/world borders as moisture sources
• Temperature-dependent evaporation
• Wind distributes moisture
• Dispersion and equalization
9
![Page 52: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/52.jpg)
Grey indicates receiving moisture amount
Climate Simulation - Precipitation
Introduction Previous Work Overview Details Results Conclusion
• Iterative computation
• Water bodies/world borders as moisture sources
• Temperature-dependent evaporation
• Wind distributes moisture
• Dispersion and equalization
9
![Page 53: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/53.jpg)
Diffusion based on current moisture
Climate Simulation - Precipitation
Introduction Previous Work Overview Details Results Conclusion
• Iterative computation
• Water bodies/world borders as moisture sources
• Temperature-dependent evaporation
• Wind distributes moisture
• Dispersion and equalization
9
![Page 54: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/54.jpg)
Diffusion based on current moisture
Climate Simulation - Precipitation
Introduction Previous Work Overview Details Results Conclusion
• Iterative computation
• Water bodies/world borders as moisture sources
• Temperature-dependent evaporation
• Wind distributes moisture
• Dispersion and equalization
9
![Page 55: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/55.jpg)
Climate Simulation - Precipitation
Introduction Previous Work Overview Details Results Conclusion
• Iterative computation
• Water bodies/world borders as moisture sources
• Temperature-dependent evaporation
• Wind distributes moisture
• Dispersion and equalization
9
Moisture, low = red, high = blue
![Page 56: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/56.jpg)
Climate Simulation - Precipitation
Introduction Previous Work Overview Details Results Conclusion
• Iterative computation
• Water bodies/world borders as moisture sources
• Temperature-dependent evaporation
• Wind distributes moisture
• Dispersion and equalization
• Temperature and moisture-dependent precipitation
9
Moisture, low = red, high = blue
Precipitation, low = red, high = blue
![Page 57: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/57.jpg)
Climate Simulation - Precipitation
Introduction Previous Work Overview Details Results Conclusion
• Iterative computation
• Water bodies/world borders as moisture sources
• Temperature-dependent evaporation
• Wind distributes moisture
• Dispersion and equalization
• Temperature and moisture-dependent precipitation
• Enables phenomena like rain shadows
9
Moisture, low = red, high = blue
Precipitation, low = red, high = blue
![Page 58: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/58.jpg)
• Biomes classified by temperature and precipitation
Climate Simulation - Biomes
Introduction Previous Work Overview Details Results Conclusion 10
![Page 59: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/59.jpg)
• Biomes classified by temperature and precipitation
• Discretized Whittaker diagram as lookup table
• Fully customizable or replaceable
Climate Simulation - Biomes
Introduction Previous Work Overview Details Results Conclusion 10
Temperature
Pre
cip
ita
tio
n
![Page 60: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/60.jpg)
• Biomes classified by temperature and precipitation
• Discretized Whittaker diagram as lookup table
• Fully customizable or replaceable
Climate Simulation - Biomes
Introduction Previous Work Overview Details Results Conclusion 10
Biome map
Temperature
Pre
cip
ita
tio
n
![Page 61: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/61.jpg)
Biome-Based Terrain Refinement
Introduction Previous Work Overview Details Results Conclusion 11
• Adds biome-specific structures
• DEMs as examples, inherently realistic
Biome map
DEMs
![Page 62: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/62.jpg)
Biome-Based Terrain Refinement
Introduction Previous Work Overview Details Results Conclusion 11
• Adds biome-specific structures
• DEMs as examples, inherently realistic
• DEMs, ℎ𝑑, blended with base terrain, ℎ𝑏
Biome map Base terrain Refined terrain
DEMs
![Page 63: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/63.jpg)
Biome-Based Terrain Refinement
Introduction Previous Work Overview Details Results Conclusion 11
• Adds biome-specific structures
• DEMs as examples, inherently realistic
• DEMs, ℎ𝑑, blended with base terrain, ℎ𝑏
Biome map Base terrain Refined terrain
DEMs
ℎ 𝑝 = 𝑤𝑏 ⋅ ℎ𝑏 𝑝 + 𝑤𝑑 ⋅ ℎ𝑑(𝑝)
![Page 64: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/64.jpg)
Biome-Based Terrain Refinement
Introduction Previous Work Overview Details Results Conclusion 11
• Adds biome-specific structures
• DEMs as examples, inherently realistic
• DEMs, ℎ𝑑, blended with base terrain, ℎ𝑏
Biome map Base terrain Refined terrain
DEMs
ℎ 𝑝 = 𝑤𝑏 ⋅ ℎ𝑏 𝑝 + 𝑤𝑑 ⋅ ℎ𝑑(𝑝)
![Page 65: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/65.jpg)
Biome-Based Terrain Refinement
Introduction Previous Work Overview Details Results Conclusion 11
• Adds biome-specific structures
• DEMs as examples, inherently realistic
• DEMs, ℎ𝑑, blended with base terrain, ℎ𝑏
Biome map Base terrain Refined terrain
DEMs
ℎ 𝑝 = 𝑤𝑏 ⋅ ℎ𝑏 𝑝 + 𝑤𝑑 ⋅ ℎ𝑑(𝑝)
![Page 66: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/66.jpg)
Biome-Based Terrain Refinement
Introduction Previous Work Overview Details Results Conclusion 11
• Adds biome-specific structures
• DEMs as examples, inherently realistic
• DEMs, ℎ𝑑, blended with base terrain, ℎ𝑏
• Natural biome transitions essential
Biome map Base terrain Refined terrain
DEMs
ℎ 𝑝 = 𝑤𝑏 ⋅ ℎ𝑏 𝑝 + 𝑤𝑑 ⋅ ℎ𝑑(𝑝)
![Page 67: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/67.jpg)
Biome-Based Terrain Refinement
Introduction Previous Work Overview Details Results Conclusion 11
• Adds biome-specific structures
• DEMs as examples, inherently realistic
• DEMs, ℎ𝑑, blended with base terrain, ℎ𝑏
• Natural biome transitions essential
• Further noise-based border distortion
Biome map Base terrain Refined terrain
DEMs
ℎ 𝑝 = 𝑤𝑏 ⋅ ℎ𝑏 𝑝 + 𝑤𝑑 ⋅ ℎ𝑑(𝑝)
![Page 68: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/68.jpg)
Biome-Based Terrain Refinement
Introduction Previous Work Overview Details Results Conclusion 11
• Adds biome-specific structures
• DEMs as examples, inherently realistic
• DEMs, ℎ𝑑, blended with base terrain, ℎ𝑏
• Natural biome transitions essential
• Further noise-based border distortion
• Weighted blending of adjacent DEMs, 𝑖, via 2D kernel
Biome map Base terrain Refined terrain
DEMs
ℎ 𝑝 = 𝑤𝑏 ⋅ ℎ𝑏 𝑝 + 𝑤𝑑 ⋅ ℎ𝑑(𝑝)
![Page 69: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/69.jpg)
Biome-Based Terrain Refinement
Introduction Previous Work Overview Details Results Conclusion 11
• Adds biome-specific structures
• DEMs as examples, inherently realistic
• DEMs, ℎ𝑑, blended with base terrain, ℎ𝑏
• Natural biome transitions essential
• Further noise-based border distortion
• Weighted blending of adjacent DEMs, 𝑖, via 2D kernel
ℎ 𝑝 = 𝑤𝑏 ⋅ ℎ𝑏 𝑝 + 𝑤𝑑 ⋅
𝑖=1
𝑛
𝑤𝑖 ⋅ ℎ𝑑𝑖(𝑝)
Biome map Base terrain Refined terrain
DEMs
![Page 70: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/70.jpg)
Procedural Asset Placement
Introduction Previous Work Overview Details Results Conclusion 12
• Iterative, rule-based local-to-global model
![Page 71: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/71.jpg)
Procedural Asset Placement
Introduction Previous Work Overview Details Results Conclusion 12
• Iterative, rule-based local-to-global model
• Assets with bilateral placement rules
![Page 72: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/72.jpg)
Procedural Asset Placement
Introduction Previous Work Overview Details Results Conclusion 12
• Iterative, rule-based local-to-global model
• Assets with bilateral placement rules
• Asset hierarchy, assigned to biomes
![Page 73: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/73.jpg)
Procedural Asset Placement
Introduction Previous Work Overview Details Results Conclusion 12
• Iterative, rule-based local-to-global model
• Assets with bilateral placement rules
• Asset hierarchy, assigned to biomes
• Constrained-based placement via dart throwingSample Rnd.
PositionCheck
ConstraintsValid Pos.
n tries
![Page 74: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/74.jpg)
Procedural Asset Placement
Introduction Previous Work Overview Details Results Conclusion 12
• Iterative, rule-based local-to-global model
• Assets with bilateral placement rules
• Asset hierarchy, assigned to biomes
• Constrained-based placement via dart throwing
• Enables emergent multi-object distributions
Sample Rnd.Position
CheckConstraints
Valid Pos.
n tries
![Page 75: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/75.jpg)
Procedural Asset Placement
Introduction Previous Work Overview Details Results Conclusion 12
• Iterative, rule-based local-to-global model
• Assets with bilateral placement rules
• Asset hierarchy, assigned to biomes
• Constrained-based placement via dart throwing
• Enables emergent multi-object distributions
• All seasons with one placement
• Switching asset variants
Sample Rnd.Position
CheckConstraints
Valid Pos.
n tries
![Page 76: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/76.jpg)
Result: Proxy View of Final Terrains
Introduction Previous Work Overview Details Results Conclusion 13
Represents 1600 𝑘𝑚2
![Page 77: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/77.jpg)
Result: Proxy View of Final Terrains
Introduction Previous Work Overview Details Results Conclusion 13
Represents 1600 𝑘𝑚2
Wind
![Page 78: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/78.jpg)
Result: Proxy View of Final Terrains
Introduction Previous Work Overview Details Results Conclusion 13
Represents 1600 𝑘𝑚2
![Page 79: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/79.jpg)
Result: Asset Placement
Introduction Previous Work Overview Details Results Conclusion 14
~ 200,000 instances on final terrain
![Page 80: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/80.jpg)
Result: Asset Placement
Introduction Previous Work Overview Details Results Conclusion 14
Shrubs exclusively in shadow of dense tree clusters Tight clusters of shrubs in open spaces between trees Dense, clumped shrubs around loosely grouped trees
![Page 81: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/81.jpg)
Result: Asset Placement - Seasons
Introduction Previous Work Overview Details Results Conclusion 15
![Page 82: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/82.jpg)
Result: Performance
Introduction Previous Work Overview Details Results Conclusion 16
Complexity:
𝑂 𝑁 ⋅ 𝑘 - N: # cells𝑘: # iterations
Pipeline Step Res. 1 Res. 2 Res. 3
Terrain Gen. 1024 2048 4096
Asset Placing 30 60 120
Rest 128 256 512
![Page 83: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/83.jpg)
Result: Terrain Refinement Performance
Introduction Previous Work Overview Details Results Conclusion 17
![Page 84: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/84.jpg)
Conclusion
Introduction Previous Work Overview Details Results Conclusion
• Successful combination of synthetic, physics-basedand example-based PTG
18
![Page 85: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/85.jpg)
Conclusion
Introduction Previous Work Overview Details Results Conclusion
• Successful combination of synthetic, physics-basedand example-based PTG
• Effective generation of vast, plausible-lookinglandscapes
18
![Page 86: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/86.jpg)
Conclusion
Introduction Previous Work Overview Details Results Conclusion
• Successful combination of synthetic, physics-basedand example-based PTG
• Effective generation of vast, plausible-lookinglandscapes
• Varied landscapes as combination of biomes
18
![Page 87: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/87.jpg)
Conclusion
Introduction Previous Work Overview Details Results Conclusion
• Successful combination of synthetic, physics-basedand example-based PTG
• Effective generation of vast, plausible-lookinglandscapes
• Varied landscapes as combination of biomes
• Procedural, complex rule-based asset placement
18
![Page 88: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/88.jpg)
Conclusion
Introduction Previous Work Overview Details Results Conclusion
• Successful combination of synthetic, physics-basedand example-based PTG
• Effective generation of vast, plausible-lookinglandscapes
• Varied landscapes as combination of biomes
• Procedural, complex rule-based asset placement
• Quick, easy-to-use iterative workflow
18
![Page 89: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/89.jpg)
Conclusion
Introduction Previous Work Overview Details Results Conclusion
• Successful combination of synthetic, physics-basedand example-based PTG
• Effective generation of vast, plausible-lookinglandscapes
• Varied landscapes as combination of biomes
• Procedural, complex rule-based asset placement
• Quick, easy-to-use iterative workflow
• Unreal Engine 4 integration
18
![Page 90: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/90.jpg)
Future Work
• Consider geological properties and soil types
Introduction Previous Work Overview Details Results Conclusion 19
![Page 91: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/91.jpg)
Future Work
• Consider geological properties and soil types
• Add rivers/water bodies and erosion
Introduction Previous Work Overview Details Results Conclusion 19
![Page 92: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/92.jpg)
Future Work
• Consider geological properties and soil types
• Add rivers/water bodies and erosion
• Generate and combine DEMs using neural networks
Introduction Previous Work Overview Details Results Conclusion 19
![Page 93: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/93.jpg)
Future Work
• Consider geological properties and soil types
• Add rivers/water bodies and erosion
• Generate and combine DEMs using neural networks
• Add sketch-based user control
Introduction Previous Work Overview Details Results Conclusion 19
![Page 94: Procedural Generation of Multi-Biome Landscapes...Motivation •Huge landscapes as combination of different biomes •Populating the terrain with objects Introduction Previous Work](https://reader035.fdocuments.us/reader035/viewer/2022071607/6144fa8734130627ed50b234/html5/thumbnails/94.jpg)
Future Work
• Consider geological properties and soil types
• Add rivers/water bodies and erosion
• Generate and combine DEMs using neural networks
• Add sketch-based user control
• Improve efficiency, e.g. multi-threading
Introduction Previous Work Overview Details Results Conclusion 19