Object recognition II Flashcards by Mel Nelte

Flashcards in Object recognition II Deck (8)

Loading flashcards...

Biedermann model

Proposed that we recognize objects by components, goons (geometric shapes)
critics are: how is it possible to distinguish between objects of similar shape
what are the brains mechanisms and where are the goons located?

Biederman on viewpoint dependent recognition

object recognition is affected by viewpoint, research shows that objects seen in less familiar viewpoints are less recognized

Ullmans model

explains how images/objects are segmented and how they are categorized based on visual representations (hierarchically organized)

Basic features of the model

highly informative pictorial fragments that are hierarchal organized
a feature is informative when it reduces uncertainty of the objects category.

informative features: need distinction and frequency

feature hierarchies

help to generalize objects
atomic fragments: lines, edges, corners
bottom up: incomplete information about the object
top-down segmentation: groups together different fragments and then classify it as a certain object.

Summary of Ullmans model

A hierarchy of abstract features continuously extracted from examples. they are usually very class-specific, non generic, and learned based on usefulness.
pros: can extract known facts of object
cons: takes a huge amount of extracting fragments

IT neurons

respond to pictorial objects preferentially but are not highly selective. Size and color can change.

types of features

low: generic and correlate directly with picture
mid: allows generalization across categories
high: very specific to a class: e.g. fruits

Object recognition II Flashcards Preview

PS3012 Perception > Object recognition II > Flashcards

Biedermann model

Proposed that we recognize objects by components, goons (geometric shapes)
critics are: how is it possible to distinguish between objects of similar shape
what are the brains mechanisms and where are the goons located?

Biederman on viewpoint dependent recognition

object recognition is affected by viewpoint, research shows that objects seen in less familiar viewpoints are less recognized

Ullmans model

explains how images/objects are segmented and how they are categorized based on visual representations (hierarchically organized)

Basic features of the model

highly informative pictorial fragments that are hierarchal organized
a feature is informative when it reduces uncertainty of the objects category.

informative features: need distinction and frequency

feature hierarchies

help to generalize objects
atomic fragments: lines, edges, corners
bottom up: incomplete information about the object
top-down segmentation: groups together different fragments and then classify it as a certain object.

Summary of Ullmans model

A hierarchy of abstract features continuously extracted from examples. they are usually very class-specific, non generic, and learned based on usefulness.
pros: can extract known facts of object
cons: takes a huge amount of extracting fragments

IT neurons

respond to pictorial objects preferentially but are not highly selective. Size and color can change.

types of features

low: generic and correlate directly with picture
mid: allows generalization across categories
high: very specific to a class: e.g. fruits

Decks in PS3012 Perception Class (6):

Key Links

Subjects

Company

Find Us

Brainscape's Knowledge GenomeTM

Object recognition II Flashcards Preview

PS3012 Perception > Object recognition II > Flashcards

Biedermann model

Proposed that we recognize objects by components, goons (geometric shapes)critics are: how is it possible to distinguish between objects of similar shapewhat are the brains mechanisms and where are the goons located?

Biederman on viewpoint dependent recognition

object recognition is affected by viewpoint, research shows that objects seen in less familiar viewpoints are less recognized

Ullmans model

explains how images/objects are segmented and how they are categorized based on visual representations (hierarchically organized)

Basic features of the model

highly informative pictorial fragments that are hierarchal organizeda feature is informative when it reduces uncertainty of the objects category.informative features: need distinction and frequency

feature hierarchies

help to generalize objectsatomic fragments: lines, edges, cornersbottom up: incomplete information about the objecttop-down segmentation: groups together different fragments and then classify it as a certain object.

Summary of Ullmans model

A hierarchy of abstract features continuously extracted from examples. they are usually very class-specific, non generic, and learned based on usefulness. pros: can extract known facts of objectcons: takes a huge amount of extracting fragments

IT neurons

respond to pictorial objects preferentially but are not highly selective. Size and color can change.

types of features

low: generic and correlate directly with picturemid: allows generalization across categorieshigh: very specific to a class: e.g. fruits

Key Links

Subjects

Company

Find Us

Brainscape's Knowledge Genome^TM

Proposed that we recognize objects by components, goons (geometric shapes)
critics are: how is it possible to distinguish between objects of similar shape
what are the brains mechanisms and where are the goons located?

highly informative pictorial fragments that are hierarchal organized
a feature is informative when it reduces uncertainty of the objects category.

informative features: need distinction and frequency

help to generalize objects
atomic fragments: lines, edges, corners
bottom up: incomplete information about the object
top-down segmentation: groups together different fragments and then classify it as a certain object.

A hierarchy of abstract features continuously extracted from examples. they are usually very class-specific, non generic, and learned based on usefulness.
pros: can extract known facts of object
cons: takes a huge amount of extracting fragments

low: generic and correlate directly with picture
mid: allows generalization across categories
high: very specific to a class: e.g. fruits