PerLecture 5 - Objects and Scene Perception Flashcards

Vocabulary-based flashcards covering concepts of object and scene perception, Gestalt laws, neurological pathways, and facial recognition based on lecture notes.

Perceptual Organisation

A concept consisting of scene segmentation and perceptual grouping that affects how we interpret things in our visual field.

Scene Segmentation

The process of partitioning a scene into multiple, meaningful elements, such as objects or object components.

Perceptual Grouping

The process of determining which elements in a scene belong to the same object.

Inverse Projection

The requirement in 3D interpretation to correct for 3D objects being projected onto the 2D retina, which can create ambiguity.

Viewpoint Invariance

A part of object recognition involving the ability to recognize an object regardless of the viewpoint, even when it creates a different retinotopic image.

Thorpe et al. (1996)

An EEG study that found the brain takes $160\,\text{ms}$ to recognize an animal in a scene presented to a participant.

Pareidolia

The phenomenon where facial recognition is so strong that people see meaningful shapes, including faces, in coincidental patterns with no meaning.

Structuralism Approach

An analytical approach to perception that identifies elementary sensations as the building blocks of the perceptual experience.

Introspective Method

A method used in structuralism that involves ignoring all aspects of an object except for primary or elementary features such as light and color levels.

Gestalt Approach

A holistic approach suggesting the relationship between features affects perception and that the entire conscious experience must be considered because the whole is different than the sum of the parts.

Necker Cube

A perceptual illusion demonstrating how the interpretation of overall configuration is the foundation of perception and suggests that Gestalt is unconscious.

Proximity

The Gestalt law stating that elements that are closer together tend to be grouped together.

Similarity

The Gestalt law stating that elements that are most similar tend to be grouped together.

Good Continuation

The tendency to perceive line segments as smooth continuations of the same line, and segments of abrupt changes as different lines.

Closure

A grouping cue where closed contours tend to determine perceptual groups.

Common Fate

A grouping cue where elements that move simultaneously in the same direction and speed tend to be grouped together.

Common Regions

A grouping cue where elements located within a common region of space are perceived as grouped together.

Uniform Connectedness

A grouping cue where connected elements appear as single objects with multiple parts.

Figure-Ground Segregation

The process of identifying what an object is in the foreground and background of an image using specific cues to segregate them.

Figure

The part of an ambiguous image that is more recognizable, thing-like, memorable, and possesses its own borders defining its shape in the foreground.

Ground

The undefined element of an image that extends behind the figure and has its borders imposed by the figure.

Bottom-Up Effects

Scene perception where property cues of the stimulus determine grouping and figure-ground segregation.

Top-Down Effects

Scene perception where familiarity, context, and knowledge determine the meaning of objects and how the scene is organized.

Scene Statistics

The concept that global image structure, such as a yellow and blue concave line representing a beach, can provide a gist of what a scene is.

Anterior Occipital Cortex

The brain region where semantic or meaning-based encoding of object recognition occurs.

Superior Temporal Sulcus

The area involved in facial processing that assesses where an individual is looking and mouth movement.

Fusiform Gyrus

The brain region containing the Fusiform Face Area (FFA) which is responsible for identifying faces.

Meng et al. (2012)

A study finding that the right FFA only reacted to real faces, while the left FFA reacted to anything resembling a face with increasing strength based on similarity.

Artificial Neural Networks

Systems using input, hidden, and output layers along with deep learning and algorithms to assess the likelihood of reality in scene perception.