Chapter 4 Flashcards

(35 cards)

1
Q

What is object recognition?

A

The ability to categorize and identify objects regardless of differences in viewing conditions.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What is the inverse projection problem?

A

Different objects can produce the same retinal image.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is viewpoint invariance?

A

The ability to recognize an object from different viewpoints.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is middle vision?

A

The stage between basic feature extraction and object recognition; involves edge detection and region grouping.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What are illusory contours?

A

Perceived edges without actual luminance or colour changes; the brain infers missing information.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What does Gestalt psychology state?

A

The whole is greater than the sum of its parts.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What are Gestalt grouping rules?

A

Rules describing when visual elements appear to group together.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Define good continuation.

A

Elements that form a continuous line or contour are grouped together.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Define similarity in Gestalt terms.

A

Elements similar in colour or shape tend to group together.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Define proximity.

A

Objects close together are grouped as part of the same object.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Define parallelism.

A

Parallel elements tend to be grouped together.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Define symmetry.

A

Symmetrical elements are perceived as belonging together.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Define common region.

A

Elements within a shared region are grouped, even if not close together.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Define common fate.

A

Elements moving together are grouped as a single object.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Define synchrony.

A

Elements that change or move at the same time are grouped together.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Define meaningfulness/familiarity.

A

Elements forming familiar or meaningful patterns are grouped together (e.g., cloud shapes).

17
Q

What are perceptual committees?

A

Neural systems that work together like a committee to interpret visual input by consensus.

18
Q

What are committee rules?

A

They obey physical laws and avoid interpreting accidental viewpoints.

19
Q

What is an ambiguous figure?

A

A stimulus that can be interpreted in more than one way.

20
Q

What is an accidental viewpoint?

A

A rare viewpoint that creates misleading regularities; committees assume such views are uncommon.

21
Q

What is the Bayesian approach to perception?

A

A probabilistic model estimating the likelihood of a hypothesis given observed data (P(H|O)).

22
Q

What is figure-ground assignment?

A

Determining which parts of an image are object (figure) and which are background (ground).

23
Q

Which cues help in figure-ground separation?

A

Symmetry, size, parallelism, meaningfulness, extremal edges, relative motion, and surroundedness.

24
Q

What are extremal edges?

A

Edges shaded to appear as if receding into depth, typically seen as figure.

25
What is the goal of middle vision?
To group appropriate elements, split others, use prior knowledge, avoid accidents, and reduce ambiguity.
26
What is the naïve template theory?
Object recognition through exact matching of visual input to stored templates.
27
What is the problem with template theory?
Would require a separate template for every size, angle, and orientation of an object.
28
What is structural description theory?
Objects are recognized by their parts and the relationships between them.
29
Who proposed recognition by components?
Irving Biederman (1987).
30
What are geons?
Basic geometric shapes (e.g., cylinders, cones) that serve as building blocks of objects.
31
Why is recognition by components useful?
It allows viewpoint invariance by combining geons into various configurations.
32
Is object recognition perfectly viewpoint invariant?
No, recognition time increases as an object’s orientation deviates from learned views.
33
What is an entry-level category?
The label that first comes to mind when seeing an object (e.g., 'dog').
34
What is a subordinate-level category?
A more specific classification of an object (e.g., 'poodle').
35
What is a superordinate-level category?
A more general category of an object (e.g., 'animal').