object recognition Flashcards

1
Q

what is perception

A
  • Perception refers to our ability to extract meaning from sensory input

It includes:

  1. audition
  2. taste
  3. touch
  4. olfaction (smell)

but research is dominated by vision.

  • Vision alone accounts for over 50% of all neurons in our cortex
  • How many senses do we have? many types of senses[ eg sense of hunger sense of when you need the toilet] but limited amount of receptors
  • Perception is a constructive process
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

the visual system

A

LVF processed by right hemisphere

RVF processed by left hemisphere

  • corpus callosum acts as pathway between left and right hemisphere → study using ppl with epilepsy
    Two processing streams: ventral, dorsal
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

dorsal

processsing stream.

A

where things are → in the posterior partietal cortex

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

ventral

processing stream

A

what you’re seeing → in the inferior temporal cortex

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

a misleading impression of simplicity regarding vision

A
  • Studies like this (Tootell etal, 1982) support the notion that a near perfect representation of the external world is “projected”onto our primary visual cortex.
  • The stimuli were displayed for 25-30 minutes and the monkeys were not conscious during the experiment…
  • monkeys injected with radioactive isotope and became concentrated
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

a 3 stage model

A
  1. local features = edge detection / contrast etc [outlines]
  2. shape representation = gestalt principles / feature integration
  3. object representation = stored representation / knowledge
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

gestalt principle

A
  • The whole visual percept is more than the sum of its parts.
  • Our perceptual system constantly tries to impose organisation on its input
  • Components of an image are grouped together on the basis of certain visual properties
  • Laws of perceptual organisation e.g. “good continuation” and “closure” give rise to “illusory contours” → we use our previous experiences to create assumptions
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

shape perception

A

Primarily “bottom up” processes produce a “primal sketch” (Marr,1982)

This sketch contains “primitives”– edges / orientations / positions/ lengths / colours etc

“Top down” processes (such as Gestalt laws) are used to group collections of primitives together into “lines, curves, larger blobs, groups and small patches” - symbolic primitives.

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

object recognition- 3 models

A
  1. template matching [prototype theory]
  2. . feature analysis
  3. recognition by components
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

template matching

A
  • A template is an internal representation
  • A memory against which the visual input is matched.
  • Computer based object recognition programmes use templates.
  • Intuitively plausible – object recognition must involve some kind of contact with a “comparable internal form”

But what rules determine whether a match is made?

How many different templates are needed?

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

feature analysis

A

Assumes lower level features are analysed first

  • The perceptual system searches for simple but characteristic features of an object
  • Supported by neurological evidence (e.g.Orientation selective cells in visual cortex)* Most research focuses on letter / word recognition (used to read postcodes!)
  • Letter A made up of / \ and –
  • But what about V and X? Both made of /and \ … Spatial relationship important
  • What about complex natural objects?
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

recognition by components

A

Recognition By Components = (Feature analysis in 3d)

  • any specific view of an objectcan be represented as anarrangement of simple 3D shapes – geons (Biederman,1987)
  • Geons are “viewpoint invariant”– easily “recoverable” from a 2D retinal image
  • “Invariant properties” include cotermination & parallelism
  • Object recognition is impaired when geons are made “non-recoverable” by removing termination points
    but… Geons reappear with ‘splats’ - the ‘splats’ means your brain makes the assumption that its a 3D object
  • not easily explained by ‘recognition by components
  • colour can air recognition too
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q
A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly