biofrq.attributes

Face attribute prediction without the liveness pass. Returns the same shape as biofrq.analyze.attributes, scoped to the kinds you asked for.

In practice, most integrations call biofrq.analyze with a single kinds value; biofrq.attributes is the lighter-weight building block when you don't need liveness in the same call.

Params

Field	Type	Required	Default	Notes
`images[0]`	`ImageItem`	yes	—	One image. See Envelope → Image inputs.
`kinds`	`string[]`	no	all	Which attribute heads to run.

Supported kinds

Kind	Output	Notes
`ageGender`	`{ "age": int, "gender": "male" \| "female", "confidence": float }`	Coarse age band, not exact age.
`emotion`	`{ "label": string, "confidence": float }`	Six-class softmax.

New kinds may be added; clients should ignore unknown keys.

Request

json

{
  "action": "biofrq.attributes",
  "params": {
    "images": [
      { "kind": "image", "bytes": "<base64>", "contentType": "image/jpeg" }
    ],
    "kinds": ["emotion"]
  }
}

Response

The per-action payload lives at data.results[0].value:

json

{
  "emotion":   { "label": "neutral", "confidence": 0.71 },
  "ageGender": { "age": 31, "gender": "female", "confidence": 0.82 }
}

Only the kinds you asked for are populated. Omitted kinds are omitted from the response.

Common errors

INVALID_REQUEST — no face in the image, or unknown entry in kinds.

Open in playgroundbiofrq.attributes