custom-timelapse-analysis

To extract scenes from a long video, you provide a video file or public URL along with target keywords. The system selects matching segments and compiles them into a condensed highlight reel for fast retrieval of relevant moments.

Keyword-based video summarization targets scene-based summarization by extracting segments that match user-specified subjects like people or events. It works by assembling these relevant clips into an emotionally rich summary reel.

You can retrieve historical reports by running the list command with your open-id via the cloud API. This retrieves the latest report list and outputs it as a Markdown table with clickable links.

An open-id is required to run time-album condensation tasks. You must validate and retrieve your open-id before invoking the analysis APIs to extract keyword-based highlight reels from your videos.

Generating a time-album highlight summary requires a provided input video file or a public URL. You also need to specify your extraction target description to select the most relevant segments for the condensed compilation.

Automated clip selection targets scene-based summarization for contexts like home videos and travel logs. It requires strict cloud-based history listing and open-id validation before invoking the analysis APIs, limiting offline processing.