Object Perception Terms

Question 1

Recognition

Answer 1

perceiving something as previously known

eg. have you thought this before yes/no

Question 2

Identification

Answer 2

Naming or classifying an object

eg. what is it?

Question 3

Data-driven processing (bottom-up)

Answer 3

low-level stimulus information combined into larger wholes to create representation

Question 4

Concept-drive processing (top-down)

Answer 4

higher-level cognitive processes affect interpretations of the stimulus input

Eg. memories, beliefs, expectations

Question 5

Template Matching

Answer 5

compare input to a model or template stored in memory, stimulus categorized by exact match.

PRO: successfully used by machines

CON: can’t handle novel stimuli, can’t handle variations within a stimulus, too many templates required, cannot handle context

Eg. cookie cutter templates

Question 6

Prototype approach

Answer 6

individual instances not store, represented as a prototype (abstraction of typical or best example of an object). Categorization based on “distance” between perceived item and prototype.
PRO: more flexible than template,
CON: can’t handle context

Eg. great dane / chihuahua far away from prototypical dog

Question 7

Feature analysis

Answer 7

PRO: can identify a wide range of stimuli, just specify component features, feature-detectors physiologically relate to cells in the visual system.

CON: doesn’t define features, cannot handle organizational principles, cannot handle context effects, cannot be applied to 3D objects

Pandemonium (Selfridge 1959) Eg. Stage 1: “Image Demon” gets sensory input eg. R
Stage 2: “Feature Demons” analyze input in terms of feature; each activated by its specific feature
Stage 3: “Cognitive Demons” determine which patterns of features are present, corresponding to known objects
Stage 4: “Decision Demon” identifies the pattern by listening for the Cognitive Demon shouting the loudest

Question 8

Marr/Nishihara (1978) Structuralist Description Approach

Answer 8

problem: defining an object with an absolute frame of reference, solution: define objects characteristics with respect to object itself (object-centered), determine objects primary axis using generalized cones, create shape descriptions of object at different levels of detail, each level of hierarchy contains info about axes of cones, arrangement of axes of component cones, internal reference to a 3-D description of component models, comprises 3-D model description. Object identification finds match between 3-D model description and a stored catalog of 3-D models of known objects. Have specificity index, adjunct index, parent index.

PRO: doesn’t rely on a catalog of features, is economical, handles variation and novel stimuli, allows for top-down processing, accounts for organizational principle.

CONS: physiological evidence is questionable, identifies objects by gross features not details

Question 9

Generalized Cones

Answer 9

can be used to determine object’s primary axis. Have an axis of orientation, a certain location or centre of mass, overall size

Eg. pyramids, spheres, cylinders, oblongs, “arms” and “legs”

Question 10

Viewpoint Invariance

Answer 10

ability to identify an object from different points of view
Eg. can determine a chair is a chair even when viewed from the side, front and back

Question 11

geons

Answer 11

geometric icons, 36 basic volumetric shapes that can be modified (length, width etc.) and yet remain identifiable. Different geons have different non-accidental (not an artefact of viewing position but rather reflect a property of the world). Non-accidental properties would be curvature, collinearity, symmetry, parallel, co-termination

Eg. cylinder, brick, cone

Question 12

Principle of componential recovery

Answer 12

in an object’s geons can be determined then the object can be recognized or identified even if object is partially obscured. PRO: has well-defined components, can handle novel stimuli and variation, is economical CON: geons not always reliability determined, may be too broad (objects also differ in their details), is viewpoint-invariant; however objects are most easily identified from a canonical viewpoint

eg. Priming studies: 1) prime: present object eg. teacup 2a) viewpoint-invariant contour change: present object from same category made of different geons eg. cylindrical mug 2b) metric change: present object from same category made of some geons, but stretched eg. latte bowl. Responses were faster for 2b than 2a

Question 13

3-D Model description

Answer 13

object-centered, invariant over changes in position of the viewer.