Visual Cognition II Object Perception
description
Transcript of Visual Cognition II Object Perception
![Page 1: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/1.jpg)
Visual Cognition II
Object Perception
![Page 2: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/2.jpg)
Theories of Object Recognition
• Template matching models• Feature matching Models• Recognition-by-components• Configural models
![Page 3: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/3.jpg)
Template matching
Detect patterns by matching visual input with a set of templates stored in memory – see if any template matches.
TEST INSTANCE
“J” TEMPLATE “T” TEMPLATE
![Page 4: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/4.jpg)
Problem:
what if the object differs slightly from the template? E.g., it is rotated or scaled differently?
Solution:
use a set of transformations to best align the object with a template (using translation, rotation, scaling)
TEST INSTANCE
“J” TEMPLATE “T” TEMPLATE
rotation
![Page 5: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/5.jpg)
Template-matching works well in constrained environments
![Page 6: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/6.jpg)
Figure 2-15 (p. 58)Examples of the letter M.
Problem: template matching is not powerful enough for general object recognition
![Page 7: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/7.jpg)
Feature Theories
• Detect objects by the presence of features• Each object is broken down into features• E.g.
A = + +
![Page 8: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/8.jpg)
Problem
• Many objects consist of the same collection of features
• Need to also know how the features relate to each other structural theories
• One theory is recognition by components
Different objects, similar sets of features
![Page 9: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/9.jpg)
Recognition by Components
• Biederman (1987): Complex objects are made up of arrangements of basic, component parts: geons.
• “Alphabet” of 24 geons
• Recognition involves recognizing object elements (geons) and their configuration
![Page 10: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/10.jpg)
![Page 11: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/11.jpg)
Why these geons?
• Choice of shape vocabulary seems a bit arbitrary• However, choice of geons was based on non-accidental
properties. The same geon can be recognized across a variety of different perspectives:
except for a few “accidental” views:
![Page 12: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/12.jpg)
Viewpoint Invariance
![Page 13: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/13.jpg)
• Viewpoint invariance is possible except for a few accidental viewpoints, where geons cannot be uniquely identified
![Page 14: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/14.jpg)
Prediction
• Recognition is easier when geons can be recovered
• Disrupting vertices disrupts geon processing more than just deleting parts of lines
ObjectDeleting
line segments
Deleting vertices
![Page 15: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/15.jpg)
Evidence from priming experiments
![Page 16: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/16.jpg)
Problem
• Theory does not say how color, texture and small details are processed. These are often important to tell apart similar objects. E.g.:
![Page 17: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/17.jpg)
Configural models of recognition
• Individual instances are not stored; what is stored is an “exemplar” or representative element of a category
• Recognition based on “distance” between perceived item and prototype
prototype
match
“Face space”
no match
![Page 18: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/18.jpg)
Configural Models
![Page 19: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/19.jpg)
Configural effects in face processing
Do these faces have anything in common?
How about these ones? By disrupting holistic processing, it becomes easier to process the individual parts
![Page 20: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/20.jpg)
Face superiority effect
Farah (1994)
![Page 21: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/21.jpg)
Results
![Page 22: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/22.jpg)
Face superiority effect
• Parts of faces are not processed independently. The context of other face parts (e.g. mouth) influences recognition of a particular part (e.g. nose)
• Face superiority effect disappears when face is inverted
![Page 23: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/23.jpg)
Context and Object Recognition
![Page 24: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/24.jpg)
What do you see?
![Page 25: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/25.jpg)
Top-down vs. Bottom up
Visual Input
Low Level Vision
High Level Vision
Bottom-up processing
Stimulus driven
Knowledge
![Page 26: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/26.jpg)
Top-down vs. Bottom up
Visual Input
Low Level Vision
High Level Vision
Top-down processing
Knowledge drivenContext Effects
Knowledge
![Page 27: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/27.jpg)
Top-down effects: knowledge influences perception
![Page 28: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/28.jpg)
Slide from Rob Goldstone
![Page 29: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/29.jpg)
Problem for many object recognition theories. How to model role of context? Context can often help in
identification of an object
Later identification of objects is more accurate when object is embedded in coherent context
![Page 30: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/30.jpg)
Context can alter the interpretation of an object
![Page 31: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/31.jpg)
Context Effects in Letter Perception
The word superiority effect: discriminating between letters is easier in the context of a word than as letters alone or in the context of a nonword string.
DEMO:http://psiexp.ss.uci.edu/research/teachingP140C/demos/demo_wordsuperiorityeffect.ppt (Reicher, 1969)
![Page 32: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/32.jpg)
• Word superiority effect suggests that information at the word level might affect interpretation at the letter level
• Interactive activation model: neural network model for how different information processing levels interact
• Levels interact– bottom up: how letters combine to form words– top-down: how words affect detectability of letters
![Page 33: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/33.jpg)
The Interactive Activation Model
• Three levels: feature, letter, and word level
• Nodes represent features, letters and words; each has an activation level
• Connections between nodes are excitatory or inhibitory
• Activation flows from feature to letter to word level and back to letter level
(McClelland & Rumelhart, 1981)
![Page 34: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/34.jpg)
The Interactive Activation Model
• PDP: parallel distributed processing
• Bottom-up:– feature to word level
• Top-down: – word back to letter level
• Model predicts Word superiority effect because of top-down processing
(McClelland & Rumelhart, 1981)
![Page 35: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/35.jpg)
Predictions of the IA model – stimulus is “WORK”
• At word level, evidence for “WORK” accumulates over time• Small initial increase for “WORD”
WORK
WORD
WEAR
![Page 36: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/36.jpg)
Predictions of the IA model – stimulus is “WORK”
• At letter level, evidence for “K” accumulates over time – boost from word level
• “D” is never activated because of inhibitory influence from feature level
K
R
D
![Page 37: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/37.jpg)
For a demo of the IA model, see:
http://www.itee.uq.edu.au/~cogs2010/cmc/chapters/LetterPerception/
![Page 38: Visual Cognition II Object Perception](https://reader035.fdocuments.us/reader035/viewer/2022062322/56814fd8550346895dbd9df2/html5/thumbnails/38.jpg)
Take-home message
• What you see is not what is out there in the outside world (ie., not like “taking a picture”), but instead a result of visual computation -- only those computations that are critical for survival, shaped by the evolution.