1 / 25

Microtiles : Extracting Building Blocks from Correspondences

Microtiles : Extracting Building Blocks from Correspondences. Javor Kalojanov , Martin Bokeloh , Michael Wand, Leonidas Guibas , Hans-Peter Seidel, Philipp Slusallek. Overview. Partial Symmetries. Inverse Procedural M odeling. Part I. Structuring Partial Symmetries. Goal.

foy
Download Presentation

Microtiles : Extracting Building Blocks from Correspondences

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. Microtiles: Extracting Building Blocks from Correspondences JavorKalojanov, Martin Bokeloh, Michael Wand, Leonidas Guibas, Hans-Peter Seidel, Philipp Slusallek

  2. Overview Partial Symmetries Inverse Procedural Modeling

  3. Part I Structuring Partial Symmetries

  4. Goal • Exchangeable building blocks

  5. Pairwise matches

  6. Transformation Groups • Global symmetries • Closed groups of s • Partial symmetries

  7. Related Work • Mitra et al. • Transformation space clustering • Pauly et al. • Regular patterns in transformation space • Lipman et al. • Point-wise symmetry aware distance

  8. Problem • Input • A geometric shape • Point-wise partial correspondences • Partial symmetries • Output • Characteristic building blocks

  9. Assumptions • Point-wise equivalence relation • Reflexive: • Symmetric: • Transitive: • Connectivity is preserved • Homeomorphisms

  10. Correspondence Graph Properties • Set of disconnected cliques (orbits) • Infinitely many cliques • Point-wise correspondence sets • Set of edge labels

  11. Microtiles – Instances and Classes • Microtile (definition) • Connected set of points • Same set of mappings • Microtile classes • Equivalent microtiles

  12. Properties • Existence and uniqueness • Disjoint • No partial correspondences between tiles • Encode all symmetries • Algebraic model: permutation groups

  13. Part II Understanding Inverse Procedural Modeling

  14. Overview • Compute rules, describing a class of similar models • Bokeloh et al. [2010] Output Input

  15. -Similarity • Local neighborhoods match exemplar radius radius Output radius Input

  16. -Symmetry • Correspondence set • -Symmetry w.r.t. rigid transformations • if the r-neighborhoods match

  17. Continuous Symmetries • -Slippable points • Infinitely many equivalent points • Gelfand and Guibas [2004]

  18. Microtiles 1-slippable 2-slippable

  19. The Space of r-Similar Shapes • Theorem: Given a shape S, all -similar shapes to S can be constructed out of the -microtiles of S • Unique construction

  20. Proof Outline • Show that tile boundaries remain invariant • - similarity implies 𝑟-similarity at tile boundaries

  21. Towards a Shape Grammar • Constraint for construction of shapes -similar to S • Microtile adjacency has to be present in S • Necessary for -similarity • Sufficiency not yet shown

  22. Results

  23. Conclusion • Structure for partial symmetries • Cannonical • Encodes all symmetries • Tiles do not match partially • Maximal permutation groups • Connection to inverse procedural modeling • Describes all r-similar shapes • Pairwise assembly • Unique construction

  24. Limitations & Future Work • Extraction • Robust and scalable • Ill-posed problems e.g. noisy scans • Applications • Shape understanding • Instatiation patterns • Generating new shapes • Constructive rules (grammar)

  25. Thank You!

More Related