QuestionQ85

VerifiedImplement computer vision solutions

You have a Microsoft Foundry project that uses Azure Content Understanding in Foundry Tools to analyze marketing videos.

Video segmentation is enabled.

You need to configure an analyzer to output a generated JSON field that describes the color scheme of each video segment.

How should you configure the analyzer? Each value may be used once, more than once, or not at all.

Drag & Drop
Field value type:
Field method:
Explanation

In Azure Content Understanding's fieldSchema, each field defines a 'type' (value type) and a 'method' (how the value is populated). The value type 'string' is used for natural-language text output such as a description of a color scheme. The method 'generate' is used when the model must infer or interpret content rather than extract literal text or classify into a fixed enum — this is required for video analyzers since 'extract' is supported only for document analyzers. Microsoft's official video documentation explicitly cites 'Color scheme' as an example custom field used to convey mood and atmosphere in each video segment, and demonstrates comparable descriptive fields (e.g., brandLogo) configured with "type": "string" and "method": "generate". Since describing a segment's color scheme requires the generative model to interpret and synthesize a description from the visual frames rather than copy literal text or pick from a fixed category list, 'string' + 'generate' is the correct configuration.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!