Computer Vision Statistical Shape Analysis Shape Shape is the - - PowerPoint PPT Presentation

computer vision
SMART_READER_LITE
LIVE PREVIEW

Computer Vision Statistical Shape Analysis Shape Shape is the - - PowerPoint PPT Presentation

Computer Vision Statistical Shape Analysis Shape Shape is the geometric information that remains when translation, rotation, and scale are factored out. Landmark points. D'Arcy Thompson. David George Kendall Fred Bookstein.


slide-1
SLIDE 1

Computer Vision

Statistical Shape Analysis

slide-2
SLIDE 2

Shape

  • Shape is the geometric information that remains when translation,

rotation, and scale are factored out.

  • Landmark points.
  • D'Arcy Thompson.
  • David George Kendall
  • Fred Bookstein.
slide-3
SLIDE 3

Landmark points

An anatomical landmark is a point assigned by an ex pen that corresponds between organisms in some biologically meaningful way, e.g. the corner of an eye or the meeting of two sutures on a skull. Mathematical landmarks are points located on an object according to some mathematical or geometrical property of the figure. e.g. at a point of high curvature. Pseudo-landmarks are constructed points on an organism, located either around the outline or in between anatomical or mathematical landmarks. For example, equally spaced points on the outlines

  • f micro-fossils.
slide-4
SLIDE 4

Active Shape Models

slide-5
SLIDE 5

3D Active Shape Models

Hamsici, Martinez, ICCV 2009

slide-6
SLIDE 6

Shape analysis with rotation invariant kernels

Hamsici, Martinez, PAMI 2008

slide-7
SLIDE 7

Rotation Invariant Kernels algorithm

Hamsici, Martinez, PAMI 2008

slide-8
SLIDE 8

Object Catagorization from Shape

Hamsici, Martinez, PAMI 2008

slide-9
SLIDE 9

Gait Recognition