Class Notes (836,562)
United States (324,575)
Psychology (367)
PSC 131 (20)
Geng Joy (20)
Lecture 7

PSC 131 Lecture 7: Middle Vision

5 Pages
Unlock Document

PSC 131
Geng Joy

I. Lecture 7 – Middle Vision (April 26, 2017) A. Class Summary 1. Object recognition is hard 2. Gestalt Principles: “the sum is more than the sum of the parts” 3. Inference: Heuristic Mechanisms (shortcuts the brain takes that work most of the time) a. Perceptual committees b. Bayesian inference c. Explanation for illusory contours, impossible figures, etc. B. Object Recognition is Hard 1. Occlusion – only able to see part of an object at a given time, but can automatically infer that what you see is a part of a whole 2. Unusual Viewpoints – even though you look at an object from a different perspective, you still know what it is 3. Novel examples – all are examples of chairs, fundamentally for sitting on; the commonality makes us able to infer new things 4. Many different examples a. They may not share visual features, all are different from each other b. Your brain picks out the common feature(s) 5. How to identify an object? a. Computer algorithms can find an object by the contours: edge detection b. Edge extraction picks out visible edges, but it’s not very good quality for you to identify objects i. You don’t get all of the important contours, you even get unimportant contours 6. We don’t merely “see” sensory elements but rather there is a process of object construction a. We impose the structure onto what is essentially noise (clouds, constellations) b. This process has a bias for constructing substance when it is abstract 7. Finding edges a. Object boundaries are inferred b. If we actually trace it, there is incomplete edge extraction c. We perceive it as continuous and encompassing contours 8. Illusory Contours a. A = triangle over circles; B = bar/cylinder; C = sphere; D = snake in water b. None of those things exist; our brain makes them to perceive it c. It occurs because it’s what we infer – prior experiences are very important 9. A lot of object recognition is inferential – the brain does a lot of “good guessing” (or predicting) a. When things violate those predictions, it’s surprising and can be used or ignored b. We get lots of illusions from guessing – illusory contours i. When things exploit what you’re used to c. Gestalt laws are “rules of evidence” d. Perceptual “committees” (population coding) 10. Perceptual Organization a. Structuralism – perception is created by “adding up” sensations (Wundt, 1879) i. Ex: edge models b. Gestalt approach – the whole is more than the sum of the parts (Wertheimer, 1912) i. Perception is not built up from sensations, but is a result of perceptual organization ii. Gestalt heuristics C. Gestalt Principles of Perceptual Organization 1. Pragnanz – every stimulus is seen as simply as possible a. We expect things to be occluded all the time b. We always infer the circles to be five overlapping circles, not a collection of objects that are meeting each other 2. Contour Completion – whatever path seems to be a continuation of whatever is existing, it’s more likely that it will be completed smoothly 3. Good Continuation – connected points resulting in straight or smooth curves belong together a. Lines are seen as following the smoothest path b. Illusory contours: the only way the line can be continuous is if there is an object in front of it c. Occlusion: go with the most parsimonious (simplest) explanation i. Circles: could be full or half circle ii. MASK – you assume there’s a white bar over the letters 4. Texture Segmentation – different textures show different objects (and/or parts of objects) a. Summary (Average) Orientation: picking what is mostly there for a simple solution i. To the left: assumed vertical red lines, horizontal blue lines ii. We get an impression of the objects iii. People can extract the average out of a group of objects (vertical/horizontal) 5. Similarity – things that look the same are grouped together a. We’ve learned over time that similar objects tend to group together    6. Similarity and Proximity a. Can shift between two options (4 lines or 2 squares, you don’t see them both) b. But, you can only see one at a time 7. Proximity – things that are near to each other are grouped together as a unit 8. Common Region –
More Less

Related notes for PSC 131

Log In


Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.