video-understand

AI video analysis extracts scene descriptions and detects motion by processing temporal sequences and frames. This skill automates information extraction, identifying events and tracking objects without manual review.

Yes, you can analyze MP4, AVI, and MOV video files for content understanding. The skill processes these common formats to extract frame information and generate scene descriptions.

To generate a summary of key discussion points from a recorded meeting video, the skill analyzes the video's temporal sequence and audio frames to extract action items and decisions made.

Yes, the z-ai-web-dev-sdk is required as a dependency. It provides the underlying AI models needed to process video content, detect motion, and understand temporal sequences.

The best way to track objects and identify events is through AI-driven scene description, which analyzes temporal sequences across frames to understand complex motion and extract information automatically.