Image recognition

更新时间:
复制 MD 格式

Powered by Alibaba Cloud deep learning technology, image recognition accurately identifies visual content in images, including thousands of object labels and dozens of common scenes. This service includes image tagging, scene classification, color recognition, style recognition, and element recognition. Image recognition technology is widely used in industries such as digital marketing, new retail, and advertising design.

Service activation

Click Activate Now on the product page to activate the service. For details, see the Quick Start.

Capabilities

Image recognition on the Alibaba Cloud Vision AI Open Platform includes the following capabilities:

Category

Capability

Description

image tagging

general image tagging

Identifies the main subjects in an image and applies tags. It supports thousands of labels for common object categories.

ad material analysis

Tags content in ad materials, such as people (celebrities, non-celebrities, and computer-generated characters) and scenes. It supports thousands of content labels for broad coverage.

Basic recognition

color recognition

Analyzes an image's colors and returns their values (in RGB and HEX formats) and the percentage for each color.

element recognition

Recognizes elements in an image, marks their positions with bounding boxes, and classifies them into basic types (people/objects, decorations, and text).

style recognition

Analyzes the style of an input image and suggests relevant tags. For example:

  • Visual style: Chinese style

  • Semantic style: elegant

garbage classification recognition

Classifies garbage in an image and identifies each item.

ID photo quality review

Detects and identifies quality issues in an ID photo.

scene recognition

scene recognition

Identifies the scene in an image. It supports dozens of common scenes, such as sky and grassland.

Use cases

Common use cases for image recognition include:

  • Smart photo album editing and management

    You can classify photos in an album using smart tags. For example, you can categorize landscape photos into subcategories such as sky, beach, or sunset, and group event photos into categories such as dining, sports, or performances.

  • Video scene analysis

    This feature uses deep learning algorithms trained on large-scale image data to identify content in videos. The resulting automated recognition is fast and precise, helping you improve search efficiency and accuracy, increase view counts, and reduce labor costs. It also enables services like automated content retrieval, personalized recommendations, content review, and distribution.

For more product updates, follow the Alibaba Cloud Visual Intelligence API.