Object Recognition – part 2
,Topics of This Lecture
● Indexing with Local Features
n Inverted file index
n Visual Words
n Visual Vocabulary construction
n Use for content based image/video retrieval
● Object Categorization
n More exact task definition
n Basic level categories
n Category representations
● Bag-of-Words Representations
n Representation
n Learning
n Use for image classification
Slide 2
,Object Categorization
● Find this particular object • Recognize ANY car
• Recognize ANY cow
Slide 3
, Object Categorization
● Task Description
n “Given a small number of training images of a category,
recognize a-priori unknown instances of that category and assign
the correct category label.”
● Which categories are feasible visually?
n Extensively studied in Cognitive Psychology,
e.g. [Brown’58]
“Fido” German dog animal living
shepherd being
Slide 4
,Topics of This Lecture
● Indexing with Local Features
n Inverted file index
n Visual Words
n Visual Vocabulary construction
n Use for content based image/video retrieval
● Object Categorization
n More exact task definition
n Basic level categories
n Category representations
● Bag-of-Words Representations
n Representation
n Learning
n Use for image classification
Slide 2
,Object Categorization
● Find this particular object • Recognize ANY car
• Recognize ANY cow
Slide 3
, Object Categorization
● Task Description
n “Given a small number of training images of a category,
recognize a-priori unknown instances of that category and assign
the correct category label.”
● Which categories are feasible visually?
n Extensively studied in Cognitive Psychology,
e.g. [Brown’58]
“Fido” German dog animal living
shepherd being
Slide 4
