ÃÛ¶¹ÊÓƵ

Attribute categories

An attribute category is a classification group that organizes related attributes that share a common characteristic. These categories help streamline the discovery, identification, and understanding of specific attributes by providing greater context and facilitating their application and use.

GenStudio for Performance Marketing uses ÃÛ¶¹ÊÓƵ’s AI and machine learning capabilities to study images, videos, and text and apply Asset attributes based on a probability of correctness. The attributes list of an asset is not exhaustive. Assets that contain a rich set of features may be limited to the three most dominant features identified by the AI.

Image features

Image features represent distinct and informative elements or patterns within an image that are used for analysis with Insights. The following table lists the image feature categories recognized by the GenStudio for Performance Marketing AI.

Category

Description

Example

Camera angle

The location and angle of the camera relative to the subject.

Subject distance

The distance between the camera and the subject of an image.

close up, mid shot, long shot

Camera setting

The configuration of the camera’s controls to produce the image.

Color and tone

Evaluates colors used in image elements. Identifies one to three colors from a set of 40 pre-determined colors in the following image layers:
Foreground colors—elements in the front layer of the image
Background colors—elements in the back layer of an image

Color temperature describes the general warmth or coolness of colors in the image.
Tone or temperature values: warm, cool, neutral

{width="200" modal="regular"}

Image style

The visual treatment of an image.

Lighting condition

The type of light in an image.

Objects

Identifies one or more items, entities, and elements that make up the image.

{width="200" modal="regular"}

Orientation

The position of the image relative to the aspect ratio.

landscape, portrait, square

People

When at least one person is present, one or more attributes may describe the person or persons in the image.

{width="200" modal="regular"}

Photography genres

Detects the subject and technique used to capture an image, such as abstract or landscape (not the same as landscape orientation).

Scenes

Detects the setting or environment depicted in an image.

Tags

Detects objects, elements, and other image characteristics that do not fall under a specific classification.

Video features

Image features represent distinct and informative elements, sounds, or patterns within a video for analysis with Insights. The following table lists the video feature categories recognized by the GenStudio for Performance Marketing AI.

Category

Description

Example

Audio Genre

When music is present, the video may receive one classification of music style, such as electronic or classical.

Audio Genre Category

When music is present, the video may receive one broad classification of music genre, such as acoustic or traditional.

Audio Mood

Describes the general atmosphere or tone of the audio, such as relaxing or energetic.

Audio Types

When audio is present, the video may receive a tag for one or more audio types, such as music or speech.

Objects

Identifies one or more items, entities, and elements that appear throughout the video.

{width="200" modal="regular"}

Orientation

The position of the video relative to the aspect ratio of the frame.

landscape, portrait, square

People

When at least one person is present, one or more attributes may describe the person or persons in the video.

Scenes

The setting or environment depicted in the video.

Styles

Detects visual treatments applied to elements in the video, such as matte or neon.

Tags

Detects objects, elements, and other video characteristics that do not fall under a specific classification.

Text features

Text features include counts for certain text elements, such as words, sentences, emojis, and classifications for semantics, emotion, and tone that are used for analysis with Insights. Text may also receive a readability score. Coming soon.