ブラウザから画像をアップロードすると、CORS policy のエラーが発生しアップロードできない

前提・実現したいこと

まずやってみる機械学習～AWS SAGEMAKER/REKOGNITIONとVUEで作る画像判定WEBアプリケーション（後編）
を参考にしながら、SAGEMAKERのテストを行いたいのですが、
「2-3. 動作確認」の「3. 実行結果を確認します」にてCORS policy のエラーが発生しアップロードできない状態です。

流れとしては、以下のイメージです。
アーキテクチャ

発生している問題・エラーメッセージ

Chromeの開発者ツールでは、下記が出力されています。

Access to XMLHttpRequest at 'https://runtime.sagemaker.us-east-1.amazonaws.com/endpoints/sagemaker-imageclassification-notebook-ep--2019-06-24-06-45-35/invocations/api/classification' from origin 'http:/xxxxxxxxxx.s3-website-us-east-1.amazonaws.com' has been blocked by CORS policy: Response to preflight request doesn't pass access control check: No 'Access-Control-Allow-Origin' header is present on the requested resource.

該当のソースコード

index.html

<!DOCTYPE html>
<html>
<head>
  <meta charset="utf-8">
  <title>ITA</title>
  <link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/siimple@3.1.0/dist/siimple.min.css">
</head>
<body>
  <div id="app">
    <div>
      <div class="siimple-navbar siimple-navbar--extra-large siimple-navbar--primary">
          <div class="siimple-navbar-title">Image Classification Example</div>
          <div class="siimple-navbar-subtitle">ITA</div>      </div>
      
      <div class="siimple-content siimple-content--extra-large">
          <div class="siimple-grid">
              <div class="siimple-grid-row">
                  <div class="siimple-grid-col siimple-grid-col--6 siimple-grid-col--sm-6">
                      <div class="siimple-h2">Select Image file</div>
                        <input type="file" id="file" v-on:change="onFileChange">

                      <div class="siimple-content siimple-content--Medium">
                          <div v-if="image">
                              <img :src="image" /><br>
                              <div class="siimple-btn siimple-btn--primary siimple-btn" v-on:click="uploadImage()">Upload</div>
                          </div>
                      </div>
                  </div>

                  <div class="siimple-grid-col siimple-grid-col--6 siimple-grid-col--sm-6">
                      <div class="siimple-h2">Result - SageMaker</div>
                        <div class="siimple-list">
                            <div class="siimple-list-item" v-for="result in results_classification">{{ result }}</div>
                        </div>
                      <div class="siimple-h2">Result - Rekognition</div>
                        <div class="siimple-list">
                            <div class="siimple-list-item" v-for="result in results_rekognition">{{ result }}</div>
                        </div>
                  </div>
              </div>
          </div>
      </div>
    </div>
  </div>

  <script src="https://unpkg.com/vue@2.5.17/dist/vue.min.js"></script>
  <script src="https://unpkg.com/axios@0.18.0/dist/axios.min.js"></script>
  <script src="./main.js"></script>
</body>
</html>

main.js

let app = new Vue({
  el: '#app',
  data: {
    image: '',
    results_classification: [],
    results_rekognition: [],
  },
  methods: {
    onFileChange: function(e) {
      let files = e.target.files || e.dataTransfer.files;
      if (!files.length)
        return;

      this.showImage(files[0]);
      this.results_classification = [];
      this.results_rekognition = [];
    },
    showImage: function(file) {
      let freader = new FileReader();

      freader.onload = (e) => {
        this.image = e.target.result;
      };
      freader.readAsDataURL(file);
    },
    uploadImage: function(e) {
      let config = {
        headers: {
          'content-type': 'application/octet-stream',
        }
      };

      axios
        .post(
          "https://<<YOUR ENDPOINT URL>>/api/classification",
          this.image,
          config
        )
        .then(response => {
            this.results_classification = response.data.split(",");
        })
        .catch(error => {
            console.log(error);
        });

      axios
        .post(
          "https://<<YOUR ENDPOINT URL>>/api/rekognition",
          this.image,
          config
        )
        .then(response => {
            this.results_rekognition = response.data.split(",");
        })
        .catch(error => {
            console.log(error);
        });
    }
  }
})

APP.PY

import boto3
import base64
import json
import numpy as np
import logging
import traceback,sys
from chalice import Chalice
 
app = Chalice(app_name='image-classification')
app.debug = True
 
logger = logging.getLogger()
logger.setLevel(logging.DEBUG)
 
@app.route('/classification', methods=['POST'], content_types=['application/octet-stream'], cors=True)
def classification():
 
    try:
        body_data = app.current_request.raw_body
        body_data = body_data.split(b'base64,')
 
        image = base64.b64decode(body_data[1])
 
        sagemaker_client = boto3.client(service_name='sagemaker-runtime', region_name='us-east-1')
 
        logger.info('Invoke Endpoint')
        res = sagemaker_client.invoke_endpoint(
                        EndpointName='<<YOUR SAGEMAKER ENDPOINT NAME>>',
                        ContentType='application/x-image',
                        Body = image
                    )
 
        result = res['Body'].read()
        result = json.loads(result)
 
        # the result will output the probabilities for all classes
        object_categories = ['ak47', 'american-flag', 'greyhound', 'tennis-shoes', 'toad', 'clutter']
 
        out = ''
        index = np.argsort(result)
        for i in index[::-1]:
            out += '{} / [probability] {:.2%},'.format(object_categories[i], result[i])
            if result[i] < 0.1:
                break
 
        return out[:-1]
 
    except Exception as e:
        tb = sys.exc_info()[2]
        return 'error:{0}'.format(e.with_traceback(tb))
 
@app.route('/rekognition', methods=['POST'], content_types=['application/octet-stream'], cors=True)
def rekognition():
 
    try:
        body_data = app.current_request.raw_body
        body_data = body_data.split(b'base64,')
 
        image = base64.b64decode(body_data[1])
 
        rekognition_client = boto3.client(service_name='rekognition', region_name='us-east-1')
 
        logger.info('Invoke Rekognition')
        res = rekognition_client.detect_labels(
                        Image = { 'Bytes': image },
                        MaxLabels=5,
                        MinConfidence=10
                    )
 
        translate_client = boto3.client(service_name='translate', region_name='us-east-1')
        out = ''
        for label in res['Labels'] :
            trans = translate_client.translate_text(Text=label['Name'], 
                        SourceLanguageCode='en', TargetLanguageCode='ja')
 
            out += '[en] {} / [ja] {} / [Confidence] {:.2f}%,'.format(
                        label['Name'], trans.get('TranslatedText'), label['Confidence']
                    )
 
        return out[:-1]
 
    except Exception as e:
        tb = sys.exc_info()[2]
        return 'error:{0}'.format(e.with_traceback(tb))

補足情報（FW/ツールのバージョンなど）

自分なりに調べて、Access-Control-Allow-Origin ヘッダを返すようにすればいいのは分かったのですが、実際に返すための方法がわかりませんでした。

2019/07/01追記
S3のバケットプロパティ→アクセス権限→CORS設定にてCORS構成エディタがあったため、ここに下記を設定しています。

<?xml version="1.0" encoding="UTF-8"?>
<CORSConfiguration xmlns="http://s3.amazonaws.com/doc/2006-03-01/">
<CORSRule>
    <AllowedOrigin>*</AllowedOrigin>
    <AllowedMethod>GET</AllowedMethod>
    <AllowedMethod>POST</AllowedMethod>
    <AllowedMethod>PUT</AllowedMethod>
    <MaxAgeSeconds>3000</MaxAgeSeconds>
    <AllowedHeader>*</AllowedHeader>
    <AllowedHeader>Origin</AllowedHeader>
</CORSRule>
</CORSConfiguration>