200枚のプロフィール写真から奇跡の1枚をAmazon Rekognitionに選んでもらいました

import json

import boto3

from decimal import Decimal

from urllib.parse import unquote_plus

from aws_xray_sdk.core import xray_recorder

from aws_xray_sdk.core import patch_all
patch_all()

def add_face_details(payload, face_details):

    print(payload)

    attributes = [

        'Smile',

        'Eyeglasses',

        'Sunglasses',

        'Gender',

        'Beard',

        'Mustache',

        'EyesOpen',

        'MouthOpen'

    ]


    for attribute in attributes:

        if attribute in face_details:

            payload[attribute] = Decimal(face_details[attribute]['Confidence'])

    return payload



def add_face_emotions(payload, emotions):

    for emotion in emotions:

        payload[emotion['Type']] = Decimal(emotion['Confidence'])

    return payload



def lambda_handler(event, context):

    rekognition = boto3.client('rekognition')

    table = boto3.resource('dynamodb').Table('ProfilePhotos')

    for record in event['Records']:

        bucket = record['s3']['bucket']['name']

        key = unquote_plus(record['s3']['object']['key'])

        face_response = rekognition.detect_faces(

            Image={

                'S3Object': {

                    'Bucket': bucket,

                    'Name': key

                }

            },

            Attributes=[

                'ALL',

            ]

        )

        payload = {

            'Prefix': key.split('/')[0],

            'ObjectKey': key

        }

        payload = add_face_details(

            payload,

            face_response['FaceDetails'][0]
        )

        if 'Emotions' in face_response['FaceDetails'][0]:

            payload = add_face_emotions(

                payload,

                face_response['FaceDetails'][0]['Emotions']

            )


        table.put_item(

            Item=payload

        )