3 Theories of Object Recognition Online Quiz Questions

In the interactive-activation model of letter perception, can candidate letters feed activation back to compatible visual features?

In Biederman's recognition-by-components theory, what are the simple volumetric components, such as cylinders, cones, and blocks, called? Enter the term.

A viewer sees several short line segments close together and perceives them as one shape. Which perceptual-organizing tendency best explains this grouping?

A recognition system compares each incoming image with one rigid stored pattern. Why might this system struggle to recognize the same object when its size, lighting, or viewpoint changes?

A recognition process begins with detected edges and features and builds toward candidate object identities. What is this processing direction called? Enter the established term.

A person identifies a familiar object more quickly from a commonly encountered angle than from an unusual angle. Which account most directly predicts this pattern?

A person recognizes a new bird by comparing it with several particular birds encountered before, including unusual examples. Which category-representation account best fits this strategy?

A partly obscured object on a dining table is perceived as a cup, although the visible shape does not actually match a cup. What does this example show about top-down processing?

Which representation best characterizes the structural-description approach proposed by Marr and Nishihara?

A bicycle is recognized both nearby and from far away, despite changes in its retinal image. What property makes an object representation useful in this situation?

A learner sees an unfamiliar bird and identifies it as belonging to a familiar category because it resembles that category’s average or most representative example. Which account best explains this judgment?

Select all statements consistent with feature-based theories of recognition.

Select all statements supported by parts-based structural accounts of object recognition.

True or false: In a parts-based account of object recognition, two objects made from the same parts must be identified as the same object, regardless of how those parts are arranged.

True or false: Object recognition research can combine different kinds of representation because no single representation needs to explain every recognition situation.

What is the term for the process that groups visual features into coherent shapes and parts?

In a view-based model, what is the process called when the system estimates a new appearance by using familiar stored views as reference points?

A visual system detects several diagnostic features in an unfamiliar object, and the object also appears to contain recognizable parts in a particular arrangement. Explain how a feature-based account and a parts-based structural account would each describe the information used to identify the object.