51黑料不打烊

Category

Description

Example

Attention distribution

The level of viewer attention spread across an image, indicating how much focus different areas of the image may receive. A higher distribution means that no single area dominates the viewer鈥檚 focus, while a lower distribution means that one or two focal points capture the viewer鈥檚 attention.

high, medium, low

Example of low distribution on the left and high distribution on the right:

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

Camera angle

The perspective from which the camera captures the subject, which affects the viewer鈥檚 perception and interpretation of the image. If the Image style is photograph, then this trait is identified.

Low angle, High angle, Eye level, Overhead view, Dutch angle, Bird's eye view

Example of eye level:

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

Camera setting

The specific adjustments and configurations of the camera鈥檚 controls that influence the final appearance and quality of the image. If the Image style is photograph, then this trait is identified.

Fast shutter speed, Long exposure, Bokeh blur, Motion blur, Tilt-shift blur, Flash, Wide-angle, Black and white, Surreal, Double-exposure, Macro, Normal mode

Example of a macro setting:

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

Color and tone

The colors and tonal qualities within an image. Identifies up to three colors from a predefined set of 40 colors in different image layers:

Foreground colors鈥攃olors in the front layer of the image
Background colors鈥攃olors in the back layer of the image

Color values: Red, Dark Red, Green, Bright Green, Dark Green, Light Green, Mud Green, Blue, Dark Blue, Light Blue, Royal Blue, Black, White, Off White, Gray, Dark Gray, Silver, Cream, Magenta, Cyan, Yellow, Mustard, Khaki, Brown, Dark Brown, Violet, Pink, Dark Pink, Maroon, Tan, Purple, Lavender, Turquoise, Plum, Gold, Emerald, Orange, Beige, Lilac, Olive

Color temperature

Describes the general warmth or coolness of colors in the image.

Tone or temperature values: warm, cool, neutral
{width="200" modal="regular"}

Content density

The concentration of visual elements and details within an image, indicating how much information is packed into the visual space.

Unlike attention distribution, which measures how viewer focus is spread across different areas of an image, content density focuses on the quantity of visual information present. A higher content density means that more elements are present.

high, medium, low

Example of low density on the left and high density on the right:

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

Image style

The visual treatment of an image, such as a photograph or sketch. Once the AI determines the image style, other traits may be identified. For example, if the image is a photograph, then camera settings, camera proximity, and lighting conditions may apply.

Photograph, Sketch, Painting, Digital cartoon, Infographics, Graphic design, Collage, Software screenshot

Example of a digital cartoon image style {width="200" modal="regular"}

Lighting condition

Describes the quality and characteristics of light in an image, affecting its mood, tone, and visibility.

Golden hour, Blue hour, Midday, Overcast, Night, High-key, Low-key, Daylighting, Incandescent, Fluorescent, Colorful, Studio

Example of daylighting condition:

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

Objects

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

Values are too numerous, but some examples include: backpack, book, hawk, glasses, fish, pencil, mountain bike, soap

Example of toucan and bird objects:

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

Orientation

The alignment of the image in relation to its width and height. Detects whether it is wider than it is tall (landscape), taller than it is wide (portrait), or equal in width and height (square).

landscape, portrait, square

Example of a square orientation:

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

People

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

person, woman, man, girl, boy, social group, kid, crowd, people

Example of people woman and person categories:

{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).

Architecture, Astro, Landscape, Pet, Interior, Wildlife, Night, Cityscape, Seascape, Underwater, Storm, Adventure sports, Fashion, Portrait, Sports, Food, Street, Event, Lifestyle, Commercial, Group, Abstract, Minimalist, Composite

See

Scenes

Identifies the setting or environment within an image, providing context about where the image was captured or the type of location depicted.

Values are too numerous, but some examples include: lake, underwater, highway, hill, log cabin, island, beach, lounge

Example snow, sky, winter, and mountain scenes reflected on a helmet:

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

Subject distance

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

close up, mid shot, long shot

Example of a closeup shot:

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

Styles

Detects visual treatments applied to image elements, such as those used in Lightroom or Photoshop.

design, illustration, logo, square, cartoon, art, circle, circular

Example of circular style:

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

Tags

Detects other image characteristics that do not fall under a specific classification. Tags provide additional context and metadata about the image. For example, the AI may detect helmet and motorobike objects in an image, and include riding as a tag.

Values are too numerous, but some examples include: construction, gothic, healing, military, selfie, football, typing, dancer, dancing

Example of dancer and dancing tags:

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