1 / 54

Yoshinori Dobashi (Hokkaido University) Yusuke Shinzo (Hokkaido University)

Modeling of Clouds from a Single Photograph. Yoshinori Dobashi (Hokkaido University) Yusuke Shinzo (Hokkaido University) Tsuyoshi Yamamoto (Hokkaido University). Overview. Introduction Related Work Proposed Method Results Conclusion. Introduction. Realistic Display of Clouds

lauren
Download Presentation

Yoshinori Dobashi (Hokkaido University) Yusuke Shinzo (Hokkaido University)

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Modeling of Clouds from a Single Photograph Yoshinori Dobashi (Hokkaido University) Yusuke Shinzo (Hokkaido University) Tsuyoshi Yamamoto (Hokkaido University)

  2. Overview • Introduction • Related Work • Proposed Method • Results • Conclusion

  3. Introduction • Realistic Display of Clouds • Synthesizing images of outdoor scenes • Realistic density distribution • Long research history • Modeling of clouds • Procedural approach • Physically-based approach outdoor scene

  4. Introduction • Procedural approach • Use of simple heuristically-defined rules • Low computational cost • Difficult to adjust parameters • Physically-based approach • Simulating physical process of cloud formation • Realistic clouds • High computational cost

  5. Our Approach • Image-based approach • Use of a single photograph • Not to reconstruct the same clouds • Using the photo as a guide to synthesize similar clouds • Three types of clouds cirrus altocumulus cumulus

  6. Overview • Introduction • Related Work • Proposed Method • Results • Conclusion

  7. Related Work • Procedural modeling • Fractals [Vos83] • Textured ellipsoids [Gar85] • Metaballs + noise function [Ebe97] • Spectral synthesis [Sak93] • Real-time modeling/rendering system [SSEH03] Difficult to adjust parameters [Gar85] [Ebe97] [SSEH03]

  8. Related Work • Physically-based modeling • Numerical solution of atmospheric fluid dynamics [KH84, MYDN01, MYDN02] • Controlling cloud simulation [DKNY08] High computational cost [KH84] [MYDN01] [MYDN02] [DKNY08]

  9. Related Work • Image-based modeling • Use of infrared satellite images [DNYO98] Not applicable to photo taken from the ground No cloud types satellite image synthesized clouds

  10. Overview • Introduction • Related Work • Proposed Method • Results • Conclusion

  11. Overview of Our Method cirrus alto- cumulus cumulus

  12. Overview of Our Method cirrus alto- cumulus cumulus Calculation of cloud image (common process)

  13. intensity opacity cirrus intensity opacity altocumulus cumulus intensity opacity Calculation of Cloud Image • Overview (input photographs) (cloud images)

  14. 1. Removing cloud pixels 2. Interpolating sky color 3. calculation of intensity/opacity Calculation of Cloud Image • Processes input image sky image intensity opacity

  15. 3. calculation of intensity/opacity Calculation of Cloud Image • Processes 1. Removing cloud pixels 2. Interpolating sky color input image sky image intensity opacity

  16. 1. Removing cloud pixels Calculation of Cloud Image • Processes 2. Interpolating sky color input image sky image 3. calculation of intensity/opacity intensity opacity

  17. 1. Removing cloud pixels 2. Interpolating sky color Calculation of Cloud Image • Processes input image sky image 3. calculation of intensity/opacity = cloud image intensity opacity

  18. Removing Cloud Pixels • Use of chroma to identify cloud pixels • Clouds are generally white (or gray) • Remove pixel if chroma < threshold Input photograph removed cloud pixels

  19. Calculation of Sky Image • Interpolation of sky colors by solving Poisson equation: pc: cloud pixel l = R, G, B cloud pixels pc sky image

  20. Calculation of Cloud Image • Intensity of clouds (single scattering) sun sky viewpoint cloud

  21. Calculation of Cloud Image • Intensity of clouds (single scattering) sun Isun sky bIsun viewpoint cloud

  22. Calculation of Cloud Image • Intensity of clouds (single scattering) sun Isun sky bIsun viewpoint Isky cloud aIsky

  23. Calculation of Cloud Image • Intensity of clouds (single scattering) sun Isun sky bIsun viewpoint Isky cloud aIsky Icld (p: pixel, l = R, G, B)

  24. Calculation of Cloud Image • Computing cloud intensity (b) & opacity (a) by minimizing: store these in cloud image unknowns sky image input image

  25. Overview of Our Method cirrus Modeling of cirrus alto- cumulus cumulus

  26. Modeling of Cirrus • Cirrus clouds • Thin and no self-shadows • Two-dimensional texture input photograph

  27. input image cloud image cloud plane cirrus cloud texture Modeling of Cirrus • Use of cloud image as 2D texture • Removing effect of perspective transformation by specifying cloud plane

  28. Overview of Our Method cirrus alto- cumulus Modeling of altocumulus cumulus

  29. metaball position radius center density field function Modeling of Altocumulus • Altocumulus • Thin but with self-shadows are observed • Using Metaballs to define three-dimensional density distribution metaball input photo

  30. Generating Metaballs • Converting cloud image into binary image cloud image binary image

  31. Generating Metaballs • Distance transform of binary image • Distance from a white pixel to the nearest black pixel distance image binary image

  32. Generating Metaballs • Generating 2D metaballs at white pixels • Use distance for radius of metaball binary image

  33. Generating Metaballs • Generating 2D metaballs at white pixels • Use distance for radius of metaball distance image binary image

  34. Optimizing Metaball Density • Minimizing difference between cumulative density and cloud intensity center density field function [Wyvill90] cloud image cumulative density at pixel p

  35. Optimizing Metaball Density • Example cumulative density image cloud image

  36. Specifying Cloud Plane • Interactive specification • Orientation of cloud plane and viewing angle cloud plane

  37. cloud plane 2D metaballs viewpoint Projecting Metaballs • Projecting metaball center onto cloud plane • Scaling metaball radius in proportion to distance from viewpoint

  38. Computing Density Distribution • Calculating bounding box of all metaballs • Subdividing bounding box into grid • Computing density at each grid point density distribution bounding box

  39. Overview of Our Method cirrus alto- cumulus cumulus cumulus

  40. Modeling of Cumulus • Cumulus • Generating surface shape • Calculating densities inside surface shape cloud image surface shape 3D density distribution

  41. Computing Surface Shape • Converting cloud image into binary image cloud image binary image

  42. medial axis binary image distance image distance image Computing Surface Shape • Distance transform of binary image • Extracting medial axes • Pixels where distances are local maxima

  43. medial axis cloud image thickness image colorization by optimization [Levin 04] Computing Surface Shape • Use distance at medial axis as thickness of clouds • Propagate thickness by optimization

  44. front view side view Computing Surface Shape • Constructing surface shape • Assuming symmetric shape with respect to image plane

  45. Computing Density Distribution • Calculating bounding box of surface shape • Subdividing bounding box into grid • Calculating density at each grid point

  46. Overview • Introduction • Related Work • Proposed Method • Results • Conclusion

  47. Results • Computer • CPU: Intel Corei7 (3.33 GHz) • Main memory: 4GB • GPU: NVIDIA GeForce GTX 295 • Computation time • within 10 seconds

  48. input image cloud texture input image cloud texture Results • Cirrus

  49. Results • Altocumulus input input synthesized synthesized

  50. Results • Cumulus input input synthesized synthesized

More Related