Feature Integration Theory - PowerPoint PPT Presentation

1 / 11
About This Presentation
Title:

Feature Integration Theory

Description:

Preattentive Stage: perception of primitives - visual ... Tilt. Line ends. Movement. Color. Brightness. Direction of illumination. Feature Integration Theory ... – PowerPoint PPT presentation

Number of Views:537
Avg rating:3.0/5.0
Slides: 12
Provided by: Mik7191
Category:

less

Transcript and Presenter's Notes

Title: Feature Integration Theory


1
Feature Integration Theory
  • Preattentive Stage perception of primitives -
    orientation, contours, curvature, color and
    movement

2
Feature Integration Theory
  • Preattentive Stage perception of primitives -
    texture differences produce pop-out boundaries

3
Feature Integration Theory
  • Preattentive Stage perception of primitives -
    visual search method - The Os pop-out in these
    pictures - The Os and Vs do not share primitive
    features

4
Feature Integration Theory
  • Preattentive Stage perception of primitives -
    visual search method - The Rs do not pop-out,
    search takes longer when there are more
    distractors - P and R share primitive features

5
Feature Integration Theory
  • Preattentive Stage perception of primitives -
    these are perceived before the object is
    recognized
  • Curvature
  • Tilt
  • Line ends
  • Movement
  • Color
  • Brightness
  • Direction of illumination

6
Feature Integration Theory
  • Focused Attention Stage primitive features are
    integrated, the object is recognized and
    associated with information in memory-attention
    is important in doing this

7
Object Recognition by Components
  • Object recognition by volumetric primitives,
    called geons (geometric ions)

8
Object Recognition by Components
  • View invariance geons are perceived from many
    different perspectives and angles
  • Discriminability geons are well-discriminated
    from each other, at many angles
  • Geons are resistant to visual noise

9
Object Recognition by Components
  • Geons are resistant to visual noise

10
Object Recognition by Components
  • Componential Recovery Perception of an object
    based on a subset of its geons.

11
Object Recognition by Components
  • Geons and 3D Modeling
Write a Comment
User Comments (0)
About PowerShow.com