Object recognition and perception

Question 1

What is perception

Answer 1

The process of discovering from our senses what is present in the world and where it is

Question 2

what is object recognition

Answer 2

the process of identifying the objects that surround us.

Question 3

In object recognition, what is the input of the visual system like

Answer 3

amosaic, with each photoreceptor seeing a small part of the image.

Question 4

are retinal images organised

Answer 4

no

Question 5

what is necessary for object recognition and describe what happens in the early stages of perception

Answer 5

Perceptual organisation is necessary. Early perception starts as unstructured elements (like edges and colours) that must be grouped into a coherent figure to recognise objects.

Question 6

what are cones essential for and what are rods essential for

Answer 6

rods- luminance
cones- colour such as the grey,red,green and blue channels

Question 7

There are 4 different receptors in our retina what do they provide

Answer 7

they provide different types of visual information , they need to be processed so the object can be recognised /found

Question 8

how do we seperate the object from it’s background using edges and boundries

Answer 8

e detect edges and boundaries that define where one object ends and another begins. The primary visual cortex (V1) is sensitive to lines and contrasts, helping identify the figure (object) from the background.

Question 9

hue is important for object recognition why?

Answer 9

The retina contains different colour channels that provide information about hue. When an object has a different hue than the background, the contrast in colour—known as chromatic contrast—helps the visual system distinguish the object from its surroundings.

Question 10

What other cue is important for object recognition

Answer 10

vision contains binocular and monocular cues ( use knowledge from previous lecture) for depth perception. As we have two eyes, binocular disparity occurs, making the object appear either closer or further away. This creates a depth map

Question 11

in perceptual organisation , what is figure-background segregation

Answer 11

It is the process of identifying one object as a figure, whereas the rest forms the ground

Question 12

what type of processing is involved in figure-background segregation in perceptual organization

Answer 12

from the bottom up ( we have cues from our environment)

from the top-down- cognition driven

Question 13

What does Gestalt psychology look at

Answer 13

look at the mind and behaviour as a whole.

Organism perceives patterns or configurations, not the elementary components

They created the laws of perceptual organisation

Question 14

what is perceptual segregation

Answer 14

process by which the visual system determines which elements in a scene belong together and form separate objects.

Question 15

what is figure-ground segregation in the laws of perceptual organisation

Answer 15

one object is identified as a figure, whereas the rest forms the ground

Question 16

What is the aims of perceptual organisation in the laws of perceptual organisation and what type of process is it

Answer 16

aims : perceptual segregation and figure ground segregation

bottom up prcessing- stimulus driven- independant from cognitive factors

Question 17

what is the law of similarity in the law of perceptual organisation

Answer 17

similar elements are grouped together perceptually, for example same colour, same shape and same size.

Question 18

give some examples of the law of similarity

Answer 18

Mil matches to the smaller circle and mal matches to the larger circle as they feel similar to use personally even though there is no distinction as to why. This is similar across all cultures and for those who are blind

Question 19

What is the law of pragnanz/ the law of good figure

Answer 19

the elements that result in the simplest form tend to be grouped together when they come from a complex object

e.g An example of this can be experienced with the Olympic logo. When you look at the logo, you see overlapping circles rather than an assortment of curved, connected lines.

Question 20

What is the law of continuity

Answer 20

elements in a line or curve seem more related to one another than those positioned randomly.

Question 21

What is the law of closure

Answer 21

Our brains tend to fill in missing parts of a shape or object so that we see it as complete, even if some parts are not actually there.

Example:
If you see a circle with small gaps in its outline, you will still see a whole circle because your mind fills in the gaps.

Question 22

What is the law of proximity

Answer 22

elements close together are grouped together

e.g. the lines are close together so visually they belong together

Question 23

what is the law of common region

Answer 23

elements that are within the same boundary are grouped together meaning objects located in the same closed region, we perceive them as belonging to the same group

Question 24

what are the legacy of the law of perceptual organisation

Answer 24

First one to understand the issue and to introduce the figure-ground segregation . What is the role to separate the figure and the background

The laws stood the test of time- no alternative explanation

Question 25

How can we criticise the law of perceptual organisation

Answer 25

Claim that it is a bottom up process, but miss and do not talk about top-down procceses ( dots being upside down takes the brain longer to process that the dots are in the same image) The laws are a description not an explanation - doesnt attempt to explain how the brain seperates the image from the backgroun

Question 26

What are natural scene statistics

Answer 26

refers to physical regulariies in our environemtn that provide information for perceptual processing and organisation

Question 27

what does light from above state as a natural scene statistic

Answer 27

When we see shaded images, we automatically interpret the shading as if the light source is above. This affects how we perceive surfaces: If the top of an object is lighter and the bottom is darker, we see it as convex (bulging out). If the top is darker and the bottom is lighter, we see it as concave (caving in).

Question 28

What is another natural scene statistic

Answer 28

Edges are usually vertical or horizontal. Gravity is always exerting effects. There are only two stable configuration either flat on the floor or standing in the centre of balance

Question 29

what is the hollow mask illusion

Answer 29

occurs when viewers percieve a concave mask as convex. This occurs as the brain rely on past experiences and knowledge of how faces typically appear. The brain interprets the visual cues which leads to the ppt believing the face portrudes than receeds.

Question 30

Why is the hollow-face illusion a natural statistical illusion?

Answer 30

Because the brain uses learned statistical expectations from real-world experience — it assumes faces are usually convex, not concave. When viewing a concave mask, the brain relies on this prior knowledge and misinterprets the shape, creating the illusion.

Question 31

how do we recognise an object in short

Answer 31

Cues from the environment interacts with our visual system

Question 32

Who purposed research into object recognition

Answer 32

Marr in Marr's theory

Question 33

what type of approach is Marr's theory of object recognition

Answer 33

Structuralist approach

Question 34

What did Marr look at

Answer 34

Looked at object recognition as a computational problem. The idea was to write a computer software that could recognise an image like a human would. Understand the visual system for object recognition

Question 35

What is the primal sketch for object recognition

Answer 35

visual system creates a two-dimensional description of the image based on changes in light intensity. This involves detecting edges, curves, and boundaries. The purpose of the primal sketch is to extract basic structural information from the scene, which is essential for further processing and understanding of what is being viewed.

Question 36

what is the 2.5 D sketch in Marr's theory

Answer 36

Objects begin to acquire volume; image gains horizontal, vertical, and depth dimensions. Depth and orientation of surfaces are calculated using cues such as texture, motion, and binocular disparity. Helps interpret how surfaces are positioned and related in space.

Question 37

What is the final sketch Marr purposed for Object recognition

Answer 37

produces a three-dimensional, view-independent description of the object . The brain compares the 3D image to stored representations in memory to categorise the object. Objects are decomposed into elementary units called generalised cones.

Question 38

what are generalised cones

Answer 38

Objects can be broken down into elementary units called generalized cones

Question 39

what can we do with the generalized cones for object recognition

Answer 39

The brain constructs a 3D model of the object from these cones. Object recognition happens by matching this constructed 3D model against a catalogue of stored 3D models in memory. If a match is found, the object is recognised.

Question 40

Who extended marrs research and what did he look at

Answer 40

Biderman's theory looked at object recognition as a computational problem . The model is a program for computer vision

Question 41

according to Bierderman's theory , what are non-acidental features

Answer 41

visual properties, such as parallel lines, symmetry, and curvature, that remain constant across different viewpoints. Because these features are unlikely to arise by chance from a particular angle, they provide reliable cues for object recognition.

Question 42

according to biederman theory of object recognition , what does the visual system do with non-accidental features

Answer 42

The visual system uses non-accidental features to identify objects regardless of the viewing angle, making them crucial for robust and efficient recognition. ( view-invariant)

Question 43

How are non-accidental features used in Biederman’s Recognition-by-Components theory?

Answer 43

They are used to identify which 3D parts (called geons) are present in an image and to determine how those geons are spatially related to one another.

Question 44

Why can we recognize so many objects with only a small number of geons?

Answer 44

Because a small set of geons can be combined in many different spatial arrangements, creating a huge number of possible object representations. | They are used to identify complex objects

Question 45

Stage 1 – Edge Extraction- biderman

Answer 45

The visual system detects edges and outlines of objects to provide the basic structure for identifying shapes.

Question 46

Detection of Non-Accidental Properties- biderman

Answer 46

The brain identifies stable features like parallel lines, symmetry, and curvature that remain consistent across viewpoints.

Question 47

stage 3 – Parsing at Regions of Concavity- biderman

Answer 47

We break objects into parts where the surface curves inward. These curved-in areas help us see the separate parts of an object, making it easier to recognise what it is.

Question 48

: Stage 4 – Determination of Components

Answer 48

Each part of the 3D object is identified as a geon.each part of the object is matched to one of these geons.

Question 49

Stage 5 – Matching Components to Object Representations

Answer 49

The identified geons and their spatial relationships are compared with stored mental representations of known objects.

Question 50

Stage 6 – Object Identification

Answer 50

when the geons match with the stored memory of the object, we can recognise the object

Question 51

How can degraded objects be recognised

Answer 51

we can percieve their non-accidental features, allowing the visual system to identify the object’s parts and match them to stored representations, even with incomplete information.

Question 52

What do structuralist theories of perception (like Biederman’s) claim about object recognition?

Answer 52

objects are recognized using viewpoint-invariant representations, meaning recognition doesn’t depend on the viewing angle or orientation

Question 53

How does the Thatcher illusion challenge structuralist theories?

Answer 53

The illusion shows that when a distorted face is upside down, it looks normal, but upright, the distortions are obvious — proving that recognition depends on orientation and viewpoint which goes against marrs and bidermans viewpoints ( viewpoint-dependant)

Question 54

how can we test independance/dependance

Answer 54

using rotations , where ppts have to say whether the 2 stimuli match or not. The angular disparity between the objects is manipulated

Question 55

what is found when we increase angular disparity

Answer 55

We find-wen we increase angular disparity, the accuracy decreases and reaction time increases. This shows that object recognition is not entirely viewpoint independant , as our accuracy decreases when objects are rotated

Question 56

what is modularity

Answer 56

areas of the brain are seperated into different units , called modules that are responsible for specific functions. If damage in a specific module occurs, it cam impair a specific function e.g damage to V1 - blindness

Question 57

describe the case of phineous gage

Answer 57

gage- rod went through his skull. Personality changed a lot Certain brain damage in areas affect behaviour

Question 58

what is agnosia

Answer 58

where a person cannot recognise objects, even though their vision and memory are normal

Question 59

Lissauer’s research (1890)

Answer 59

After head injury, patient G.L. had problems recognizing objects mistook pictures for boxes, jacket for trousers, couldn’t recognise cutlery. Vision was intact, as was his memory of objects

Question 60

what is apperceptive agnosia

Answer 60

Impaired ability to consciously perceive and discriminate stimuli even though their vision is fine localised in the posterior right hemisphere

Question 61

what is assossiative agnosia

Answer 61

patient has an preserved ability to percieve a stimulus but can't interpret what has been seen bilateral occipitotemporal junction

Question 62

during th drawing test how did those with appreicative and assossiative agnosia perform- Lissauer’s work

Answer 62

apperceptive amnesia, they couldn’t percieve the objects, and therefore could not copy the object Associative agnosia- ppt cannot recognise objects , but they can copy and pass the drawing test. It is not a perceptual problem , but retrieving the knowledge from memory about the objects

Question 63

Who purposed the hierarchal model

Answer 63

Riddoch and Humphreys (2001) proposed that object recognition involves several distinct, hierarchical stages. Recognition is not a single process but a sequence of steps, beginning with basic visual perception and progressing to meaningful interpretation and naming. Damage to different stages results in different types of agnosia. Apperceptive agnosia occurs when early perceptual processes are impaired and is commonly linked to damage in the posterior right hemisphere. Associative agnosia occurs when later interpretative stages are impaired and is associated with damage to the bilateral occipitotemporal junction.

Question 64

what condition arises from impaired ability to percieve faces . and which area of the brain is assossiated with face perception

Answer 64

– prosopagnosia Face perception - fusiform gyrus in the inferior parietal lobe

Question 65

what is bruce and youngs 1986 model of face recongition

Answer 65

is a multi-stage process involving several different cognitive modules, each responsible for a particular aspect of face processing. The model suggests that recognising a face is not a single, unified process, but rather a series of steps that involve both perceptual and memory components.

Question 66

what is the first stage to bruce and youngs object recongtion

Answer 66

Structural Encoding: focusing on both the features and the configuration of the face apperceptive varient

Question 67

what is the second stage

Answer 67

Expression Analysis: Decoding facial expressions to understand emotional states and intentions of others

Question 68

what is the third stage

Answer 68

Facial Speech Analysis: Using facial movements to aid in speech perception (e.g., lip-reading). associative varient

Question 69

what is the fourth stage

Answer 69

Directed Visual Processing: Focusing attention on particular features when required (e.g., searching for a familiar person).

Question 70

what is the fifth stage

Answer 70

Face Recognition Units (FRUs): Storage of representations of familiar faces. They contain the structures of familiar faces we have seen before

Question 70

what is the sixth stage

Answer 70

Person Identity Nodes (PINs): Accessing information about the person associated with a recognised face (e.g., name, occupation)

Question 71

what is the 7th stage

Answer 71

Name Generation: Retrieving the name associated with a familiar face. Significance

Question 72

provide a summary of visual imagery

Answer 72

refers to the ability to "see with the mind’s eye". It Occurs when a visual representation is present in short-term memory but the stimulus is not viewed​ ​ Can be compromised following brain damage

Question 73

The case study of Mr I. (“The Case of the Colorblind Painter”)

Answer 73

he lost the ability to perceive colour in both actual vision and in his visual imagery. Whether he looked with eyes open, tried to imagine colours, remembered colourful scenes, or dreamed, he experienced only shades of grey. “Colour perceiving and colour imagery seem to rely on the same mechanisms.”

Question 74

why does visual imagery and perception work similarly

Answer 74

Both visual imagery and visual perception activate similar areas of the brain. This means that the brain systems involved in processing actual visual input are also engaged when you generate mental images.

Question 75

describe shepards and meltzer (2017) research for measuring visual imagery-mental rotation

Answer 75

shown images of 3D object two shapes are the same or different

Question 76

what was found in rotation research

Answer 76

the time taken to answer increases with the degree of rotation required, suggesting participants use visual imagery to manipulate objects in their mind’s eye.