External B-roll Search

This capability is available through partnership. This page describes the capability only and contains no API parameters. Contact business@ai-mcn.tv to discuss integration.

External B-roll Search adds public online video as a third source alongside your local footage library and material matrix. It is useful for finding historical footage, news imagery, and on-location scenes that are missing from your own assets. Describe the target picture in natural language, and the system prepares selectable candidates for your script beats.

What it does

• Searches public video across YouTube, Vimeo, TikTok, Bilibili, and other supported platforms;
• Generates low-resolution previews and returns one or more candidate clips for each narration beat;
• Keeps editable in-video time ranges for comparing and refining candidates;
• Collects selected candidates into a B-roll set and lays them onto the client timeline;
• Processes the original-quality version only after B-roll confirmation, reducing unnecessary full-video downloads;
• Optionally uses the existing visual purification capability to process watermarks or subtitles in the final footage.

Workflow

1. Describe the target picture in natural language, adding the era, people, location, action, and shot type when useful.
2. The cloud service searches public videos and creates scene analysis with usable time ranges for each candidate.
3. The client shows one or more candidate clips. You can replace, trim, and rearrange them in the B-roll set.
4. After refining the timeline, click Confirm B-roll. The service then processes the original-quality versions of the selected segments.

Search, low-resolution previews, candidate scene analysis, and initial arrangement run in the cloud. The client presents candidates and handles selection, manual refinement, and confirmation. Results are editable in-video time windows, so a target section from a longer video can be placed directly on the timeline.

Typical use cases

• Supplementing historical people, events, and archival footage;
• Quickly screening news, documentary, and public-scene footage;
• Filling visual gaps in narration videos when local assets do not have the required shot;
• Comparing multiple sources and shots while arranging B-roll.

How to get started

This capability is currently offered through partnership and can be combined with local footage, the material matrix, and existing video project workflows. Contact business@ai-mcn.tv to discuss access and integration.