Análisis emocional - Poomi Docs

Sube una foto de tu mascota para analizar su expresión facial, lenguaje corporal y estado emocional.

POST/api/v1/emotion-analysis

Imagen5 Creditos por llamada

Campo	Tipo	Requerido	Descripción
`image`	`string`	Sí	Datos de imagen codificados en Base64 (JPG/PNG/WEBP, máx 3MB)
`animalType`	`string`	No	Tipo de animal para análisis (ej. dog, cat, rabbit, hamster)
`language`	`string`	No	Código de idioma de respuesta (en, zh, es)

API Playground

Tipo de animal

Idioma

Ejemplo de solicitud

bash

curl -X POST https://www.poomi.pet/api/v1/emotion-analysis \
  -H "Authorization: Bearer pk_your_api_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "image": "base64_encoded_data",
    "animalType": "dog",
    "language": "es"
  }'

Campos de respuesta

Campo	Tipo	Requerido	Descripción
`emotion`	`string`	Sí	Most likely current emotional state
`confidence`	`number`	Sí	Confidence score from 0 to 1
`summary`	`string`	Sí	A concise 2 to 4 sentence summary covering the visible cues, the emotional interpretation, and any important limitation

Schema de respuesta

json-schema

{
  "type": "object",
  "properties": {
    "emotion": {
      "type": "string",
      "description": "Most likely current emotional state"
    },
    "confidence": {
      "type": "number",
      "description": "Confidence score from 0 to 1",
      "minimum": 0,
      "maximum": 1
    },
    "summary": {
      "type": "string",
      "description": "A concise 2 to 4 sentence summary covering the visible cues, the emotional interpretation, and any important limitation"
    }
  },
  "required": [
    "emotion",
    "confidence",
    "summary"
  ],
  "additionalProperties": false
}

Ejemplo de respuesta

json200 OK

{
  "success": true,
  "request_id": "req_xxxxxxxx",
  "data": {
    "emotion": "relaxed",
    "confidence": 0.84,
    "summary": "The pet appears relaxed in this image. The visible posture looks loose and balanced rather than tense or defensive, and there are no strong signs of fear or distress in the visible frame. A clearer full-body view would help confirm tail and lower-body cues."
  }
}