High level vision

1
High level vision
2
High level vision

• Models of object recognition
• Top down influences
• Navigation/Movement

3
Last Time. . ..
We spent a lot of time focusing on lines how you
get them, why you would want them, and so on.

• We need to move from lines to objects.
• How do you recognize an object from an
  organization of lines?
• How does perception connect to memory?

4
Models of object recognition

• Template
• Feature
• New wave of feature models (3D features)

5
Template model
6
Template--problems
Problems Size
Orientation
Need too many templates
7
Feature model--pandemonium
8
Feature models
Good visual input does seem to be decomposed
into features
Good Physiological evidence about simple
features from Hubel Wiesel
Problems orientation missing features
natural objects
9
Natural objects What are the features of a dog?

• Nose
• Ear
• Front Leg
• Tail
• Back Leg

10
Principle from Gestalt Psych
A
C
Good continuation
B
D
11
Good continuation can be used to find the parts
of objects
12
new wave of feature models
These models use three-dimensional features.
13
Biedermans geon model
Geons
Simple objects
You usually only need to see the edges of a
geon Geons have properties that are invariant to
rotations
14
Experiment
Which can be better identifiedat a very
briefexposure?
15
(No Transcript)
16
Problems with Geons

• Do geons really represent all shapes?
• How are relationships among geons coded?

17
An alternative Local viewpoints

• Note that you can identify objects from many
  different orientations. Templates Feature
  models couldnt account for thisgeons can.
• BUT how good are you at doing this really?

18
(No Transcript)
19
(No Transcript)
20
(No Transcript)
21
(No Transcript)
22
(No Transcript)
23
(No Transcript)
24
(No Transcript)
25
(No Transcript)
26
(No Transcript)
27
Alternative

• Some researchers have suggested that object
  representations are NOT viewpoint independent.
  Rather, we store views of objects the way we see
  them.

HUH? Isnt that the template theory?
The difference is that you do some fixing
(size, rotation) of the image to fit the template
28
Tarrs local view experiments
O 45 90
O 45 90
29
Tarr Results
30
Problems with local view

• Chicken egg how do you know how to rotate the
  image before you can identify it?
• What is stored is clearly not literal
  picturesbut what is it?
• How is what you see and what is in memory matched?

31
Chicken egg problem. . .
Bottom up processing refers to beginning with
relatively raw, unprocessed sensory information,
and building towards more conceptual
representations.

• Top-down processing refers to conceptual
  knowledge influencing the processing or
  interpretation of lower-level perceptual processes

32
NOTE--weve been acting as though all processing
were bottom up.
33
Example ambiguous figures
34
Example
35
Example
36
It appears that in top down processing you use
conceptual information to generate hypotheses
about what the stimulus might be, then test these
hypotheses
37
More formal work
Watch for the object appearing
38
(No Transcript)
39
(No Transcript)
40
(No Transcript)
41
(No Transcript)
42
(No Transcript)
43
The Parsing Paradox
If perceptual organization is a matter of mapping
sensations onto structural schema, which happens
first interpreting the whole or interpreting the
parts? How can someone recognize a face until he
has first recognized the eyes, nose, mouth and
ears? Then again, how can you recognize the parts
until you know that they are part of a
face? --Stephen Palmer
44
(No Transcript)
45
Question do you process the top-down information
atthe same time as the bottom-up info?
Youll see a circle--try to identify the
objectthat appears in the circle.
46
(No Transcript)
47
(No Transcript)
48
(No Transcript)
49
(No Transcript)
50
(No Transcript)
51
(No Transcript)
52
(No Transcript)
53
The result people are better at identifying the
objectwhen the scene makes sense, compared to
when its jumbled
54
How is this possible?

• Word superiority effect
• It is easier to recognize a letter in a word than
  in isolation.

55
Identify the letter that will appear in the circle
56
TAKE
57
WOLP
58
Word superiority effect

• Faster and more reliable in identifying a letter
  when its part of a word than a non-word.

Isnt there the same chicken-egg problem? Dont
you need to know the letters to identify the
word? So then how is the word helping to identify
the letters (which you already know?)
59
Model of word identification
60
Navigation vs. Object identification
There is increasing evidence that spatial
information that helps us get around is
independent of the information that helps us
identify objects.
61
Mishkin Ungerlieder
62
Mishkin Ungerlieger
63
Mishkin Ungerliedger
Spatial
Object