PowerPoint Presentation

Size: px

Start display at page:

Download "PowerPoint Presentation"

嫱车
5 years ago
Views:

1 使用 Amazon Rekognition 实时分析图像王元恺,AWS 解决方案架构师

2 沟通交流你好

3 沟通交流你好

4 Amazon Rekognition Image 采用深度学习技术的完全托管式图像识别服务搜索, 鉴别并组织百万的图片对象和场景检测面部分析面部比较面部识别名人识别图像审核文字识别

对象和场景检测 DetectLabels { }, { }, { },... "Confidence": 94.

5 对象和场景检测 DetectLabels { }, { }, { },... "Confidence": , "Name": "adventure" "Confidence": , "Name": "boat" "Confidence": , "Name": "rafting"

6 对象和场景检测用户场景

面部分析 DetectFaces [ { "BoundingBox": { "Height": 0.3449999988079071, "Left": 0.09666666388511658, "Top": 0.

1335220336914, "Type": "HAPPY" }, {"Confidence": 3.3275485038757324, "Type": "CALM"}, {"Confidence": 0.

7 面部分析 DetectFaces [ { "BoundingBox": { "Height": , "Left": , "Top": , "Width": }, "Confidence": 100, "Emotions": [ {"Confidence": , "Type": "HAPPY" }, {"Confidence": , "Type": "CALM"}, {"Confidence": , "Type": "SAD"} ], "Eyeglasses": {"Confidence": , "Value": false}, "EyesOpen": {Confidence": , "Value": true}, "Gender": {"Confidence": 100, "Value": "Female }

面部分析 Demographic Data Age Range 29-45 Gender: Male 96.

Bounding Box Image Quality Brightness 23.6% Sharpness 99.9% Emotions Happy 83.8% Surprised 0.

8 面部分析 Demographic Data Age Range Gender: Male 96.5% Facial Landmarks EyeLeft,EyeRight,Nose RightPupil,LeftPupil MouthRight,LeftEyeBrowUp Bounding Box Image Quality Brightness 23.6% Sharpness 99.9% Emotions Happy 83.8% Surprised 0.65% General Attributes Smile:True 23.6% EyesOpen:True 99.8% Beard:True 99.5% Mustache:True 99.9% Facial Pose Pitch Roll Yaw 4.383

9 面部分析用户场景收集分析人脸图片应用统计和情感属性 AMAZON REKOGNITION DetectFaces 获取广告图片日志记录 Look Your Best All Day AMAZON S3 AMAZON DYNAMODB 更新统计属性 AMAZON REDSHIFT

面孔比较 CompareFaces { } "FaceMatches": [ {"Face": {"BoundingBox": { "Height": 0.2683333456516266, "Left": 0.5099999904632568, "Top": 0.

2383333295583725, "Left": 0.6233333349227905, "Top": 0.3016666769981384, "Width": 0.15888889133930206}, "Confidence": 99.

10 面孔比较 CompareFaces { } "FaceMatches": [ {"Face": {"BoundingBox": { "Height": , "Left": , "Top": , "Width": }, "Confidence": }, "Similarity": 96 }, {"Face": {"BoundingBox": { "Height": , "Left": , "Top": , "Width": }, "Confidence": }, "Similarity": 0 } ], "SourceImageFace": {"BoundingBox": { "Height": , "Left": , "Top": , "Width": }, "Confidence": }

11 面孔比较用户场景

12 transformed stored 面部识别 Face Face ID & vector<float> f7a3a278-2a a549-a12ab1a8cae8 & v1 IndexFace 02e b39-ba57-9afb0fd8782d & v2 Collection { f7a3a278-2a a549-a12ab1a8cae8, 02e b39-ba57-9afb0fd8782d, 4c55926e-69b3-5c80-8c9b-78ea01d30690 } 4c55926e-69b3-5c80-8c9b-78ea01d30690 & v3

13 面部识别用户场景相机实时信息应用图片索引 AMAZON S3 AWS LAMBDA AMAZON REKOGNITION IndexFaces 个人信息数据表人脸收集

名人识别 RecognizeCelebrities { "CelebrityFaces": [ { "Face": { "BoundingBox": { "Height":0.

4511111080646515 },, "Quality": { "Brightness":56.59690475463867, "Sharpness": 99.

14 名人识别 RecognizeCelebrities { "CelebrityFaces": [ { "Face": { "BoundingBox": { "Height": , "Left": , "Top": , "Width": },, "Quality": { "Brightness": , "Sharpness": } }, "Id": "1SK7cR8M", "MatchConfidence": 100, "Name": "Jeff Bezos", "Urls": [ " ] } ]}

图像审核 DetectModerationLabels { "ModerationLabels": [ { Confidence : 83.

15 图像审核 DetectModerationLabels { "ModerationLabels": [ { Confidence : , "Name": "Suggestive", "ParentName": "" }, { Confidence : , Name : Female Swimwear Or Underwear, "ParentName": "Suggestive } ] }

文字识别 DetectText { "TextDetections": [ { "Confidence": 97.4773178100586, "DetectedText": "IT'S", "Geometry": { "BoundingBox": { "Height": 0.10191240906715393, "Left": 0.6658040285110474, "Top": 0.

16 文字识别 DetectText { "TextDetections": [ { "Confidence": , "DetectedText": "IT'S", "Geometry": { "BoundingBox": { "Height": , "Left": , "Top": , "Width": }, "Polygon": [ { "X": , "Y": }, ] }, "Id": 0, "Type": "LINE" }, { "Confidence": , "DetectedText": "MONDAY", "Geometry": { } }

Amazon Rekognition API 总览 Non-storage API Operations Storage-based API Operations DetectModerationLabels IndexFaces SearchFacesByImage CompareFaces {

17888888716697693}, " DetectLabels Detect-Text CreateCollection ListCollections { 17888888716697693}, " SearchFaces DeleteCollection ListFaces DetectFaces

17 Amazon Rekognition API 总览 Non-storage API Operations Storage-based API Operations DetectModerationLabels IndexFaces SearchFacesByImage CompareFaces { "FaceMatches": [ {"Face": {"BoundingB "Height": , "Left": , "Top": , "Width": }, " DetectLabels Detect-Text CreateCollection ListCollections { "FaceMatches": [ {"Face": {"BoundingB "Height": , "Left": , "Top": , "Width": }, " SearchFaces DeleteCollection ListFaces DetectFaces RecognizeCelebrities GetCelebrityInfo DeleteFaces

18 Softmax Amazon Rekognition 深度学习 Labrador Probability Dog Beach Outdoors Layer 1 Layer 2 Layer n FC Feature map

19 Amazon Rekognition Image 客户案例 Extract the rich text captured in images at scale Processed with Amazon Rekognition for millions of Pins stored in Amazon S3 Match customers with art they will love thanks to recommendation tools built on AWS

20 DEMO 整体架构前端删除图片实时分析下载原图 Serverless 获取索引图片热度获取缩略图

21 DEMO 解决自定义标签 Amazon SageMaker Machine Learning AMI RiseML on Kubernetes 数据集打标签模型训练 Custom Algorithm using existing tags

22 DEMO 解决单图多面部识别 Index Collection Search Delete

Rekognition IndexFaces SearchFaces DeleteFaces Rekognition

23 DEMO 实时分析图像 Amazon S3 Photos Put Event AWS Lambda URL & Height Amazon DynamoDB Presign URL Rekognition DetectFaces Rekognition IndexFaces SearchFaces DeleteFaces Rekognition DetectLabels Amazon S3 Thumbnail Initial Heat Amazon ElasticSearch Metadata

24 Amazon Rekognition Video 采用深度学习技术的完全托管式视频识别分析服务对象和场景检测轨迹跟踪面部识别实时流分析不安全视频检测名人识别

25 轨迹跟踪

26 Amazon Rekognition Video 客户案例 Identify and track actors across millions of frames of content Allowing viewers to identify and buy the products they see on screen.

27 Amazon Rekognition Video 客户案例 Automating Footage Tagging with Amazon Rekognition Indexed 97,000 people Saves ~9,000 hours a year in labor

28 工作代码示例 Search for rekognition : Smart image cropping Twitter bot Find missing persons using social media Automatic content moderation Serverless hotdog detection And more

29 AWS 服务集成存储解耦 AWS Elemental MediaConvert 视频处理 Amazon API Gateway Amazon S3 AWS Elemental MediaLive 应用 AWS Lambda Amazon EFS AWS Batch Amazon SNS Amazon EC2 Amazon Kinesis Amazon DynamoDB Amazon ECS Amazon SQS 索引 Amazon ElasticSearch 计算

30 Amazon Rekognition 价值完全托管 AWS 集成经过验证的可扩展性安全低价

31 感谢参加 AWS INNOVATE 2018 在线技术大会我们希望您在这里找到感兴趣的内容! 也请帮助我们完成投票打分和反馈问卷欲获取关于 AWS 的更多信息和技术内容, 可以通过以下方式找到我们 : 微信公众号 :AWSChina 新浪微博 : 领英 : 知乎 : 视频中心 : 更多线上活动 :

PowerPoint Presentation

PowerPoint Presentation Amazon Rekognition 简介亚马逊图像视频分析服务简介与动手实践王元恺 AWS 解决方案架构师图片通用的, 普遍的 & 重要的 There are 3,700,000,000 internet users in 2017 1,200,000,000 photos will be taken in 2017 (9% YoY Growth) Source: InfoTrends Worldwide