1 / 60

Medial Object Shape Representations for Image Analysis & Object Synthesis

Explore the use of shape representation in analyzing and synthesizing objects, with a focus on kidney-shaped object extraction and physical object deformation.

minore
Download Presentation

Medial Object Shape Representations for Image Analysis & Object Synthesis

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. Medial Object Shape Representationsfor Image Analysis & Object Synthesis Stephen M. Pizer Kenan Professor Medical Image Display & Analysis Group University of North Carolina, USA Credits: Many on MIDAG, especially Daniel Fritsch, Andrew Thall, George Stetten, Paul Yushkevich

  2. Medial Object Shape Representationsfor Image Analysis & Object Synthesis

  3. Whatshape representation is for • Analysis from images • Extract the kidney-shaped object • Register based on the pelvic bone shapes • Extract shape information w/o model • Synthesis • Design the object • Deform the object, with physical realism • Shape science • Shape and biology • Shape-based diagnosis

  4. Whatshape representation is for • Analysis from images • Extract the kidney-shaped object • Register based on the pelvic bone shapes

  5. Whatshape representation is for • Synthesis • Design the object • Deform the object, with physical realism

  6. Whatshape representation is for • Shape science • Shape and biology • Shape-based diagnosis Brain structures (Gerig)

  7. Shape Sciences • Geometry • The spatial layout: via primitives • Landmarks • Boundary places and orientations • Medial places, figural sizes and orientations • Space itself • Statistics • The average shape • Modes of variation from the average • Computer Graphics • Image Analysis

  8. Options for Primitives • Space: xi for grid elements • Landmarks: xi described by local geometry • Boundary: (xi ,normali) spaced along boundary • Figural: nets of diatoms sampling figures

  9. Primitives for shape representation: Landmarks • Sets of points of special geometry

  10. Primitives for shape representation: Boundaries • Boundary points with normals

  11. Object Representation by M-Reps

  12. Each M-figure Represented by Net of Medial Primitives

  13. Each M-figure Represented by Net of Medial Primitives

  14. Figural Models • Figures: successive medial involution • Main figure • Protrusions • Indentations • Separate figures • Hierarchy of figures • Relative position • Relative width • Relative orientation

  15. Primitives’ Desired Properties • Geometry • Intuitive: simple, global + local • Efficiently deformable • Easily extracted or created • Spatial tolerance inherent • Statistics • Unimodality: normally distributed • Via geometrical, tolerance-sensitive metric

  16. Figural Models with Boundary Deviations • Hypothesis • At a global level, a figural model is the most intuitive • At a local level, boundary deviations are most intuitive

  17. Union and Difference of M-figures

  18. Medial Primitives • x, (b,n) frame, r, q (object angle) • Imply boundary segments with tolerance • Similarity transform equivariant • Zoom invariance implies width-proportionality of • tolerance of implied boundary • boundary curvature distribution • spacing along net • interrogation aperture for image n

  19. 3D kidney model extracted from CT Four figure model of the kidneys Red represents indentation figures

  20. Need for Special End Primitives • Represent • non-blobby objects • angulated edges, corners, creases • still allow rounded edges , corners, creases • allow bent edges • But • Avoid infinitely fine medial sampling • Maintain tangency, symmetry principles

  21. End Primitives Corner primitive in cross-section Rounded end primitive in cross-section

  22. Displacements from Figurally Implied Boundary Boundary implied by figural model Boundary after displacements

  23. Coarse-to-fine representation • For each of three levels • Figural hierarchy • For each figure, net chain, successively smaller tolerance • For each net tile, boundary displacement chain

  24. Multiscale Medial Model • From larger scale medial net • Coarsely sampled • Smooother figurally implied boundary • Larger tolerance • Interpolate smaller scale medial net • Finer sampled • More detail in figurally implied boundary • Smaller tolerance • Represent medial displacements

  25. Multiscale Medial Model • From larger scale medial net, interpolate smaller scale medial net and represent medial displacements b.

  26. Multiscale Medial/Boundary Model • From medial net • Coarsely sampled, smoother implied boundary • Larger tolerance • Represent boundary displacements along implied normals • Finer sampled, more detail in boundary • Smaller tolerance

  27. Shape Rep’n in Image Analysis • Segmentation • Extract an object from image • Registration • Find geometric transformation that brings two images into alignment • 3D/3D • 3D/2D • Shape Measurement • Find how probable a shape is

  28. Shape Repres’n in Image Analysis • Segmentation • Find the most probable deformed mean model, given the image • Probability involves • Probability of the deformed model (prior) • Probability of the image, given the deformed model (likelihood)

  29. Probability of a deformed model • From training set • via principal components analysis, coarse-to-fine • -C * Geometric difference from typical shape

  30. Medialness: medial strength of a medial primitive in an image • Probability of image | deformed model • Sum of boundariness values • at implied boundary positions • in implied normal directions • with apertures proportional to tolerance • Boundariness value • Intensity profile distance from mean (at scale) • statistical, based on training set • Intensity differences • via Gaussian derivatives

  31. Figurally implied boundaries and rendering, via 4-figure model

  32. 3D DSL Model Deformation Initial Position of Model in Target Image

  33. 3D DSL Model Deformation Figural DeformationIteration 3

  34. 3D DSL Model Deformationwith interfigural penalties Initial position After optimization

  35. Shape Repres’n in Image Analysis • Registration • Find the most probable deformation, given the image

  36. Shape Rep’n in Image Analysis • Prior-free medial shape analysis • Cores: height ridges of medialness (Pizer, Fritsch, Morse, Furst) • Statistical analysis of medial diatoms (Stetten)

  37. Shape Rep’n in Image Analysis • Cores: height ridges of medialness

  38. M I P @ U N C

  39. Shape Rep’n in Image Analysis • Statistical analysis of medial diatoms

  40. sphere slab cylinder

  41. sphere slab cylinder

  42. sphere slab cylinder

  43. sphere slab cylinder

  44. sphere slab cylinder

  45. Shape Rep’n in CAD/CAM • Stock figural models • Deformation tools: large scale • Coarse-to-fine specification • Figural connection tools • Direct rendering, according to display needs

  46. Deformation in CAD/CAM

  47. Shape Rep’n in CAD/CAM • Design models for image analysis

  48. Medial Object Shape Representationsfor Image Analysis & Object Synthesis • Figural models, at successive levels of tolerance • Boundary displacements • Work in progress • Segmentation and registration tools • Statistical analysis of object populations • CAD tools, incl. direct rendering • Connection relative critical manifolds • …

More Related