Video understanding
Powered by Alibaba Cloud's deep learning technology, video understanding analyzes video content to generate engaging clips or frames for video covers. This helps boost click-through rates and enhance the user experience.
Service activation
Click Activate Now on the product page to activate the service. For detailed instructions, see the Beginner's Guide.
Features
The Alibaba Cloud Vision AI Platform provides the following video understanding features:
Category | Feature | Description |
video understanding | Analyzes an input video, extracts engaging content, and returns one or more video covers. | |
Segments an input video into shots and returns the timestamps for each shot change. | ||
Recognizes text in videos across various scenarios, including news, film and TV, entertainment, and sports. It can identify text in Simplified and Traditional Chinese, English, and sports scores. This feature supports multiple text types, such as standard subtitles, fixed captions, scrolling text, text in natural scenes, vertical text, and stylized fonts. | ||
Analyzes a video's shots and topics to divide it into smaller segments. It returns the boundary timestamps and a summary for each segment but does not return the actual video clips. | ||
Evaluates the quality of an input video in two modes: basic assessment and artifact assessment. It outputs both a summary report and a detailed report. |
Use cases
Common use cases for video understanding include:
Highlight video recommendations
Use the intelligent video cover feature to quickly select high-quality, representative covers for both long and short videos. This enhances the visual experience, helps users filter content faster, and increases user retention.
Compelling video covers
Generate compelling covers for your video content to highlight its most engaging moments, increasing click-through rates and user watch time.
For more product updates, follow the Alibaba Cloud Visual Intelligence API.