SlideShare a Scribd company logo
1 of 74
Download to read offline
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Apache MXNet/Gluon을 이용한
입술 읽기(Lip Reading) 모델 만들기
김형준
Research Engineer
SK Telecom
발표자료 바로 공개
발표자료는 발표 종료 후 해당
사이트에서 바로 보실 수 있습니다
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Who am I
김 형 준
- AI Research Engineer
- AI/Machine Learning Application Engineer
- Data Analysis Platform Engineer
- Data Scientist
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
SK Telecom AI Center
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Open Source & Community
(2018.07 ~ ) Apache MXNet Medium(Tech Blog) Writer & Code Contributor
(2018.06) A way to Benchmark Your Deep Learning Framework, Psygrams
(2018.01) Sentiment Analysis with grad-CAM and Target2Vec, Microsoft X 바벨피쉬
(2017.10) Machine Learning Platform Development, 데이터야놀자
(2015.12) Anomaly Detection using MRFs, Naver D2 개발자 커뮤니티
(2015.06) Sentiment Analysis, R User Conference in Korea 2015
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Contribution
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
MXNet/Gluon Model
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Agenda
1. 도움이 된 것들
2. 처음 Pull Request 하던 날
3. Lip Model 예제 작성
4. Review & Merged
5. gluonCV, gluonNLP
6. Discussion
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
MODEL: Deep Learning Frameworks
https://www.slideshare.net/awskorea/2-deep-learning-with-gluon
* Low level-control
- Hybridized Gluon(Symbolic)
* High level
- Gluon (Imperative)
* GluonCV, Gluon-NLP
* Semantic Versioning
- MXNet Community
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
MODEL: MXNet/Gluon Study (17.12 ~ 18.06)
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
MODEL: MXNet/Gluon Study (17.12 ~ 18.06)
- (NLP) Word2vec (skip-gram)
- (Vision) GAN
- (Vision) Pix2Pix
- (Vision) Grad-cam & Integrated-gradients
- (NLP) End to end Memory Network
- (NLP) LSTM with Custom Embedding
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
MODEL: MXNet/Gluon
- 때로는 다른 누군가가 만들어주더라도
- 때로는 실패하고
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
MODEL: MXNet/Gluon
- 성공했을 때의 경험은 잊을 수 없습니다
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
SOSCON 2018(https://www.slideshare.net/awskorea/apache-mxnet-soscon-2018/)
Contribution: Apache MXNet 공헌하는 방법
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Contribution: Apache MXNet 공헌하는 방법
- Source Code Contribution
: [MXNet Github](https://github.com/apache/incubator-mxnet)
- Blog
: [Medium Blog](https://medium.com/apache-mxnet)
- Community
: [MXNet Forum](https://discuss.mxnet.io/)
[Slack – mxnet, AWSKRUG ]
[Facebook – MXNet Korea User Group](https://www.facebook.com/groups/mxnetkr/)
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Medium Blog
(18.07) A way to Benchmark Your Deep Learning Framework On-premise
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Medium Blog
MXNet Makes Us Faster And Stronger!
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Source Code Contribution
[MXNet Github](https://github.com/apache/incubator-mxnet)
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Source Code Contribution
오타 수정도
번역도
가이드 문서 작성도
디자인 작업도
의견 제시도
모두 컨트리뷰션입니다
“More Common Than You Think: An In-Depth Study of Casual
Contributors”, 2016”
https://naver.github.io/OpenSourceGuide/book/BetterContribution/why-contribute-to-open-source.html
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Source Code Contribution
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Source Code Contribution
https://mxnet.apache.org/versions/master/community/contribute.html#setup-mxnet-for-development
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Source Code Contribution
[MXNet Github](https://github.com/apache/incubator-mxnet)
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Source Code Contribution
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Source Code Contribution
해당 이슈에 다음 행동들을 통해 기여해 주세요
- 의견 및 아이디어 제시
- 관심 어필
- Pull Request
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Bug Fix
- Issues
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Bug Fix
- PR(Pull Request)
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet
- https://arxiv.org/abs/1611.01599
- LipNet: End-to-End Sentence-level Lipreading (2016)
- please
입술 읽기(Lip reading)는 화자의 입의 움직임으로부터 텍스트를 디코딩하는 과제입니다.
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet
- https://arxiv.org/abs/1611.01599
- LipNet: End-to-End Sentence-level Lipreading (2016)
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
MXNet 공헌하기
- Example
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet
- https://github.com/apache/incubator-mxnet/tree/master/example/gluon/lipnet
Data
Collect
Data
Prepare
- Example
Model
Build
Model
Train
Model
Inference
utils./download_data.py utils/preprocess_data.py data_loader..py
models/network.py
trainer.py
main.py
Infer.py
- 정승환, 김형준(SK Telecom)
- 18.12.01 ~ 12.14 (약 2주간 개발)
- 18.12.14 ~ 02.15 (약 2달간 리뷰)
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet
Data
Collect
- Example
bbaf3s.mpg
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet
Data
Collect
- Example
- 16GB
- Multiple processes
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet
Data
Prepare
- Example
- Using Face
Landmark
Detection(http://dlib
.net/), It’s able to
extract the mouth
from a video.
- Using ffmpeg, It
changes 1 video to
75 frames
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet
Data
Prepare
- Example
- Preprocessing
: Video →
cropped images
- DataLoader
MXNet/Gluon
: Generator
(nd.array)
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet
Data
Prepare
전처리 시간이 너무 오래 걸림
• 54 hours  1 process
• 1 hour  54 processes
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet
Data
Prepare
Public S3 이용
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet
Data
Prepare
“직접 다운로드 & 전처리” 또는 “전처리된 데이터 이용”
- https://github.com/apache/incubator-mxnet/tree/master/example/gluon/lipnet
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet
- Model: (ST)CNN + GRU
Model
Build
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet
- Model: (ST)CNN + GRU
Model
Build
https://arxiv.org/pdf/1611.01599.pdf
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet
- Model: (ST)CNN + GRU
Model
Build
https://arxiv.org/pdf/1611.01599.pdf
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet
- Model: (ST)CNN + GRU
Model
Build
https://arxiv.org/pdf/1611.01599.pdf
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet
- 학습
Model
Train
trainer = Train(config)
trainer.build_model()
trainer.load_dataloader()
trainer.run()
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet
- 추론
Model
Inference
trainer = Train(config)
trainer.build_model()
trainer.load_dataloader()
trainer.infer_batch(trainer.valid_dataloader)
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet
- 추론 예제
Model
Inference
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet Pull Requests
- (1) Fork
- (2) New branch
- (3) Code/Doc
- (4) Test
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet Pull Requests
- (5) Fill in PR template
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet Pull Requests
- (6) Review
https://github.com/apache/incubator-mxnet/pull/13647
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet Pull Requests
- (6) Review
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet Pull Requests
- (7) Merged
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet Pull Requests
- 고려사항들 요약
- pylint
- huge data → s3
- Multi gpus
- Hybridize
- Test code
- Pretrained model
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet Pull Requests
- pylint
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet Pull Requests
- huge data → s3
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet Pull Requests
- Multi gpus
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet Pull Requests
- Hybridize
https://github.com/apache/incubator-mxnet/pull/13647/commits/f487255be48f89231ba917c570266af759cc1116
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet Pull Requests
- Test code
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet Pull Requests
- Pretrained model
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet Pull Requests
- 실수?!
- https://github.com/apache/incubator-
mxnet/tree/master/example/gluon/lipnet
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
LipNet Pull Requests
가장 힘들었던 점
- 미국과 한국의 시간 차
- 의사소통: 영어
- 체력 & 끈기: 과도한 요구? 이쯤이면 ‘approve’ 해 줍시다
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
GluonCV
- Model zoo
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
GluonCV
- Model zoo
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
GluonCV
- Model zoo
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
GluonNLP
- Model zoo
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Lip Reading
- Large-scale Visual Speech Recognition
: https://arxiv.org/pdf/1807.05162.pdf
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Lip Reading
- Large-scale Visual Speech Recognition
: https://arxiv.org/pdf/1807.05162.pdf
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Why Contribution?
- 시간 & 노력 vs 보람, 성취감, 자부심
- 실력 향상
- 최고의 실력자들로 부터 피드백
- 협업
- 테스트 코드
- 컨벤션
- “스프린트 / 핸즈업 / 밋업” 통한 사용자 확대 및 인적 네트워크
- 타인에게 Inspire
- 스터디
- 발표
- 스프린트 / 핸즈업 / 밋업
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Life is too short, Dive into Deep Learning!
http://ko.d2l.ai/
여러분의 피드백을 기다립니다!
#AWSSummit 해시태그로
소셜미디어에 여러분의
행사소감을 올려주세요.
AWS Summit Seoul 2019
모바일 앱과 QR코드를 통해
강연평가 및 설문조사에
참여하시고 재미있는 기념품을
받아가세요.
내년 Summit을 만들 여러분의
소중한 의견 부탁 드립니다.
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.

More Related Content

What's hot

5 g ran architcture
5 g ran architcture5 g ran architcture
5 g ran architctureHemraj Kumar
 
AWS Black Belt Techシリーズ Amazon Elastic Compute Cloud (Amazon EC2)
AWS Black Belt Techシリーズ Amazon Elastic Compute Cloud (Amazon EC2)AWS Black Belt Techシリーズ Amazon Elastic Compute Cloud (Amazon EC2)
AWS Black Belt Techシリーズ Amazon Elastic Compute Cloud (Amazon EC2)Amazon Web Services Japan
 
Application Load Balancer and the integration with AutoScaling and ECS - Pop-...
Application Load Balancer and the integration with AutoScaling and ECS - Pop-...Application Load Balancer and the integration with AutoScaling and ECS - Pop-...
Application Load Balancer and the integration with AutoScaling and ECS - Pop-...Amazon Web Services
 
Secure Socket Layer
Secure Socket LayerSecure Socket Layer
Secure Socket LayerPina Parmar
 
AWS Certificate Management and Private Certificate Authority Deep Dive (SEC41...
AWS Certificate Management and Private Certificate Authority Deep Dive (SEC41...AWS Certificate Management and Private Certificate Authority Deep Dive (SEC41...
AWS Certificate Management and Private Certificate Authority Deep Dive (SEC41...Amazon Web Services
 
Data Protection in Transit and at Rest
Data Protection in Transit and at RestData Protection in Transit and at Rest
Data Protection in Transit and at RestAmazon Web Services
 
AWS re:Invent 2016: Get the Most from AWS KMS: Architecting Applications for ...
AWS re:Invent 2016: Get the Most from AWS KMS: Architecting Applications for ...AWS re:Invent 2016: Get the Most from AWS KMS: Architecting Applications for ...
AWS re:Invent 2016: Get the Most from AWS KMS: Architecting Applications for ...Amazon Web Services
 
AWS기반 서버리스 데이터레이크 구축하기 - 김진웅 (SK C&C) :: AWS Community Day 2020
AWS기반 서버리스 데이터레이크 구축하기 - 김진웅 (SK C&C) :: AWS Community Day 2020 AWS기반 서버리스 데이터레이크 구축하기 - 김진웅 (SK C&C) :: AWS Community Day 2020
AWS기반 서버리스 데이터레이크 구축하기 - 김진웅 (SK C&C) :: AWS Community Day 2020 AWSKRUG - AWS한국사용자모임
 
Amazon Elasticsearch Service Deep Dive - AWS Online Tech Talks
Amazon Elasticsearch Service Deep Dive - AWS Online Tech TalksAmazon Elasticsearch Service Deep Dive - AWS Online Tech Talks
Amazon Elasticsearch Service Deep Dive - AWS Online Tech TalksAmazon Web Services
 
[AWS Innovate 온라인 컨퍼런스] Amazon Forecast를 통한 시계열 예측 활용하기 - 김종선, AWS 솔루션즈 아키텍트
[AWS Innovate 온라인 컨퍼런스] Amazon Forecast를 통한 시계열 예측 활용하기 - 김종선, AWS 솔루션즈 아키텍트[AWS Innovate 온라인 컨퍼런스] Amazon Forecast를 통한 시계열 예측 활용하기 - 김종선, AWS 솔루션즈 아키텍트
[AWS Innovate 온라인 컨퍼런스] Amazon Forecast를 통한 시계열 예측 활용하기 - 김종선, AWS 솔루션즈 아키텍트Amazon Web Services Korea
 
SID305 AWS Certificate Manager Private CA
SID305 AWS Certificate Manager Private CASID305 AWS Certificate Manager Private CA
SID305 AWS Certificate Manager Private CAAmazon Web Services
 
아키텍처 현대화 분야 신규 서비스 - 주성식, AWS 솔루션즈 아키텍트 :: AWS re:Invent re:Cap 2021
아키텍처 현대화 분야 신규 서비스 - 주성식, AWS 솔루션즈 아키텍트 :: AWS re:Invent re:Cap 2021아키텍처 현대화 분야 신규 서비스 - 주성식, AWS 솔루션즈 아키텍트 :: AWS re:Invent re:Cap 2021
아키텍처 현대화 분야 신규 서비스 - 주성식, AWS 솔루션즈 아키텍트 :: AWS re:Invent re:Cap 2021Amazon Web Services Korea
 
Dual transfer mode
Dual transfer modeDual transfer mode
Dual transfer modeMorg
 
AWS로 데이터 마이그레이션을 위한 방안과 옵션 - 박성훈 스토리지 스페셜리스트 테크니컬 어카운트 매니저, AWS :: AWS Summit...
AWS로 데이터 마이그레이션을 위한 방안과 옵션 - 박성훈 스토리지 스페셜리스트 테크니컬 어카운트 매니저, AWS :: AWS Summit...AWS로 데이터 마이그레이션을 위한 방안과 옵션 - 박성훈 스토리지 스페셜리스트 테크니컬 어카운트 매니저, AWS :: AWS Summit...
AWS로 데이터 마이그레이션을 위한 방안과 옵션 - 박성훈 스토리지 스페셜리스트 테크니컬 어카운트 매니저, AWS :: AWS Summit...Amazon Web Services Korea
 
AWS Summit Seoul 2023 | 바쁘다 바빠, 현대사회! Amazon Kendra로 원하는 자료를 적재적소에 찾아서 활용하기
AWS Summit Seoul 2023 | 바쁘다 바빠, 현대사회! Amazon Kendra로 원하는 자료를 적재적소에 찾아서 활용하기AWS Summit Seoul 2023 | 바쁘다 바빠, 현대사회! Amazon Kendra로 원하는 자료를 적재적소에 찾아서 활용하기
AWS Summit Seoul 2023 | 바쁘다 바빠, 현대사회! Amazon Kendra로 원하는 자료를 적재적소에 찾아서 활용하기Amazon Web Services Korea
 

What's hot (20)

5 g ran architcture
5 g ran architcture5 g ran architcture
5 g ran architcture
 
AWS Black Belt Techシリーズ Amazon Elastic Compute Cloud (Amazon EC2)
AWS Black Belt Techシリーズ Amazon Elastic Compute Cloud (Amazon EC2)AWS Black Belt Techシリーズ Amazon Elastic Compute Cloud (Amazon EC2)
AWS Black Belt Techシリーズ Amazon Elastic Compute Cloud (Amazon EC2)
 
Apache Spark
Apache SparkApache Spark
Apache Spark
 
Application Load Balancer and the integration with AutoScaling and ECS - Pop-...
Application Load Balancer and the integration with AutoScaling and ECS - Pop-...Application Load Balancer and the integration with AutoScaling and ECS - Pop-...
Application Load Balancer and the integration with AutoScaling and ECS - Pop-...
 
Secure Socket Layer
Secure Socket LayerSecure Socket Layer
Secure Socket Layer
 
20 palo alto site to site
20 palo alto site to site20 palo alto site to site
20 palo alto site to site
 
AWS Certificate Management and Private Certificate Authority Deep Dive (SEC41...
AWS Certificate Management and Private Certificate Authority Deep Dive (SEC41...AWS Certificate Management and Private Certificate Authority Deep Dive (SEC41...
AWS Certificate Management and Private Certificate Authority Deep Dive (SEC41...
 
Data Protection in Transit and at Rest
Data Protection in Transit and at RestData Protection in Transit and at Rest
Data Protection in Transit and at Rest
 
AWS re:Invent 2016: Get the Most from AWS KMS: Architecting Applications for ...
AWS re:Invent 2016: Get the Most from AWS KMS: Architecting Applications for ...AWS re:Invent 2016: Get the Most from AWS KMS: Architecting Applications for ...
AWS re:Invent 2016: Get the Most from AWS KMS: Architecting Applications for ...
 
AWS기반 서버리스 데이터레이크 구축하기 - 김진웅 (SK C&C) :: AWS Community Day 2020
AWS기반 서버리스 데이터레이크 구축하기 - 김진웅 (SK C&C) :: AWS Community Day 2020 AWS기반 서버리스 데이터레이크 구축하기 - 김진웅 (SK C&C) :: AWS Community Day 2020
AWS기반 서버리스 데이터레이크 구축하기 - 김진웅 (SK C&C) :: AWS Community Day 2020
 
Amazon Elasticsearch Service Deep Dive - AWS Online Tech Talks
Amazon Elasticsearch Service Deep Dive - AWS Online Tech TalksAmazon Elasticsearch Service Deep Dive - AWS Online Tech Talks
Amazon Elasticsearch Service Deep Dive - AWS Online Tech Talks
 
[AWS Innovate 온라인 컨퍼런스] Amazon Forecast를 통한 시계열 예측 활용하기 - 김종선, AWS 솔루션즈 아키텍트
[AWS Innovate 온라인 컨퍼런스] Amazon Forecast를 통한 시계열 예측 활용하기 - 김종선, AWS 솔루션즈 아키텍트[AWS Innovate 온라인 컨퍼런스] Amazon Forecast를 통한 시계열 예측 활용하기 - 김종선, AWS 솔루션즈 아키텍트
[AWS Innovate 온라인 컨퍼런스] Amazon Forecast를 통한 시계열 예측 활용하기 - 김종선, AWS 솔루션즈 아키텍트
 
SID305 AWS Certificate Manager Private CA
SID305 AWS Certificate Manager Private CASID305 AWS Certificate Manager Private CA
SID305 AWS Certificate Manager Private CA
 
아키텍처 현대화 분야 신규 서비스 - 주성식, AWS 솔루션즈 아키텍트 :: AWS re:Invent re:Cap 2021
아키텍처 현대화 분야 신규 서비스 - 주성식, AWS 솔루션즈 아키텍트 :: AWS re:Invent re:Cap 2021아키텍처 현대화 분야 신규 서비스 - 주성식, AWS 솔루션즈 아키텍트 :: AWS re:Invent re:Cap 2021
아키텍처 현대화 분야 신규 서비스 - 주성식, AWS 솔루션즈 아키텍트 :: AWS re:Invent re:Cap 2021
 
AWS Fargate on EKS 실전 사용하기
AWS Fargate on EKS 실전 사용하기AWS Fargate on EKS 실전 사용하기
AWS Fargate on EKS 실전 사용하기
 
Dual transfer mode
Dual transfer modeDual transfer mode
Dual transfer mode
 
5g-Air-Interface-pptx.pptx
5g-Air-Interface-pptx.pptx5g-Air-Interface-pptx.pptx
5g-Air-Interface-pptx.pptx
 
AWS로 데이터 마이그레이션을 위한 방안과 옵션 - 박성훈 스토리지 스페셜리스트 테크니컬 어카운트 매니저, AWS :: AWS Summit...
AWS로 데이터 마이그레이션을 위한 방안과 옵션 - 박성훈 스토리지 스페셜리스트 테크니컬 어카운트 매니저, AWS :: AWS Summit...AWS로 데이터 마이그레이션을 위한 방안과 옵션 - 박성훈 스토리지 스페셜리스트 테크니컬 어카운트 매니저, AWS :: AWS Summit...
AWS로 데이터 마이그레이션을 위한 방안과 옵션 - 박성훈 스토리지 스페셜리스트 테크니컬 어카운트 매니저, AWS :: AWS Summit...
 
AWS Summit Seoul 2023 | 바쁘다 바빠, 현대사회! Amazon Kendra로 원하는 자료를 적재적소에 찾아서 활용하기
AWS Summit Seoul 2023 | 바쁘다 바빠, 현대사회! Amazon Kendra로 원하는 자료를 적재적소에 찾아서 활용하기AWS Summit Seoul 2023 | 바쁘다 바빠, 현대사회! Amazon Kendra로 원하는 자료를 적재적소에 찾아서 활용하기
AWS Summit Seoul 2023 | 바쁘다 바빠, 현대사회! Amazon Kendra로 원하는 자료를 적재적소에 찾아서 활용하기
 
c1 & c2 values
c1 & c2 values c1 & c2 values
c1 & c2 values
 

Similar to Apache MXNet/Gluon을 이용한 입술 읽기(Lipreading) 모델 만들기 - 김형준, SK텔레콤 :: AWS Summit Seoul 2019

Technological Accelerants for Organizational Transformation - DVC303 - re:Inv...
Technological Accelerants for Organizational Transformation - DVC303 - re:Inv...Technological Accelerants for Organizational Transformation - DVC303 - re:Inv...
Technological Accelerants for Organizational Transformation - DVC303 - re:Inv...Amazon Web Services
 
DVC303-Technological Accelerants for Organizational Transformation
DVC303-Technological Accelerants for Organizational TransformationDVC303-Technological Accelerants for Organizational Transformation
DVC303-Technological Accelerants for Organizational TransformationAmazon Web Services
 
Demystifying Machine Learning On AWS - AWS Summit Sydney 2018
Demystifying Machine Learning On AWS - AWS Summit Sydney 2018Demystifying Machine Learning On AWS - AWS Summit Sydney 2018
Demystifying Machine Learning On AWS - AWS Summit Sydney 2018Amazon Web Services
 
re:Invent CON320 Tracing and Debugging for Containerized Services
re:Invent CON320 Tracing and Debugging for Containerized Servicesre:Invent CON320 Tracing and Debugging for Containerized Services
re:Invent CON320 Tracing and Debugging for Containerized ServicesCalvin French-Owen
 
AI APIs as a Catalyst for Machine Learning Initiatives
AI APIs as a Catalyst for Machine Learning InitiativesAI APIs as a Catalyst for Machine Learning Initiatives
AI APIs as a Catalyst for Machine Learning InitiativesNicholas Walsh
 
Supercharge Your Machine Learning Model with Amazon SageMaker
Supercharge Your Machine Learning Model with Amazon SageMakerSupercharge Your Machine Learning Model with Amazon SageMaker
Supercharge Your Machine Learning Model with Amazon SageMakerAmazon Web Services
 
Introduction to AI services for Developers - Builders Day Israel
Introduction to AI services for Developers - Builders Day IsraelIntroduction to AI services for Developers - Builders Day Israel
Introduction to AI services for Developers - Builders Day IsraelAmazon Web Services
 
Introduction to AI services for Developers - Builders Day Israel
Introduction to AI services for Developers - Builders Day IsraelIntroduction to AI services for Developers - Builders Day Israel
Introduction to AI services for Developers - Builders Day IsraelAmazon Web Services
 
Racing with Artificial Intelligence
Racing with Artificial IntelligenceRacing with Artificial Intelligence
Racing with Artificial IntelligenceDaniel Zivkovic
 
MXNet Paris Workshop - Intro To MXNet
MXNet Paris Workshop - Intro To MXNetMXNet Paris Workshop - Intro To MXNet
MXNet Paris Workshop - Intro To MXNetApache MXNet
 
AWS Toronto Summit 2019 - AIM302 - Build, train, and deploy ML models with Am...
AWS Toronto Summit 2019 - AIM302 - Build, train, and deploy ML models with Am...AWS Toronto Summit 2019 - AIM302 - Build, train, and deploy ML models with Am...
AWS Toronto Summit 2019 - AIM302 - Build, train, and deploy ML models with Am...Jonathan Dion
 
Frontend and Mobile with AWS Amplify | AWS Summit Tel Aviv 2019
Frontend and Mobile with AWS Amplify | AWS Summit Tel Aviv 2019Frontend and Mobile with AWS Amplify | AWS Summit Tel Aviv 2019
Frontend and Mobile with AWS Amplify | AWS Summit Tel Aviv 2019AWS Summits
 
Frontend and Mobile with AWS Amplify | AWS Summit Tel Aviv 2019
Frontend and Mobile with AWS Amplify | AWS Summit Tel Aviv 2019Frontend and Mobile with AWS Amplify | AWS Summit Tel Aviv 2019
Frontend and Mobile with AWS Amplify | AWS Summit Tel Aviv 2019Amazon Web Services
 
AWS Initiate Day Manchester 2019 – Moving to DevOps the Amazon Way
AWS Initiate Day Manchester 2019 – Moving to DevOps the Amazon WayAWS Initiate Day Manchester 2019 – Moving to DevOps the Amazon Way
AWS Initiate Day Manchester 2019 – Moving to DevOps the Amazon WayAmazon Web Services
 
Debugging Gluon and Apache MXNet (AIM423) - AWS re:Invent 2018
Debugging Gluon and Apache MXNet (AIM423) - AWS re:Invent 2018Debugging Gluon and Apache MXNet (AIM423) - AWS re:Invent 2018
Debugging Gluon and Apache MXNet (AIM423) - AWS re:Invent 2018Amazon Web Services
 
Reinforcement Learning with Sagemaker, DeepRacer and Robomaker
Reinforcement Learning with Sagemaker, DeepRacer and RobomakerReinforcement Learning with Sagemaker, DeepRacer and Robomaker
Reinforcement Learning with Sagemaker, DeepRacer and RobomakerAlex Barbosa Coqueiro
 
Building Volkswagen Group's Digital Ecosystem (AMT304) - AWS re:Invent 2018
Building Volkswagen Group's Digital Ecosystem (AMT304) - AWS re:Invent 2018Building Volkswagen Group's Digital Ecosystem (AMT304) - AWS re:Invent 2018
Building Volkswagen Group's Digital Ecosystem (AMT304) - AWS re:Invent 2018Amazon Web Services
 
Debugging and Performance tricks for MXNet Gluon
Debugging and Performance tricks for MXNet GluonDebugging and Performance tricks for MXNet Gluon
Debugging and Performance tricks for MXNet GluonApache MXNet
 

Similar to Apache MXNet/Gluon을 이용한 입술 읽기(Lipreading) 모델 만들기 - 김형준, SK텔레콤 :: AWS Summit Seoul 2019 (20)

Technological Accelerants for Organizational Transformation - DVC303 - re:Inv...
Technological Accelerants for Organizational Transformation - DVC303 - re:Inv...Technological Accelerants for Organizational Transformation - DVC303 - re:Inv...
Technological Accelerants for Organizational Transformation - DVC303 - re:Inv...
 
DVC303-Technological Accelerants for Organizational Transformation
DVC303-Technological Accelerants for Organizational TransformationDVC303-Technological Accelerants for Organizational Transformation
DVC303-Technological Accelerants for Organizational Transformation
 
Demystifying Machine Learning On AWS - AWS Summit Sydney 2018
Demystifying Machine Learning On AWS - AWS Summit Sydney 2018Demystifying Machine Learning On AWS - AWS Summit Sydney 2018
Demystifying Machine Learning On AWS - AWS Summit Sydney 2018
 
re:Invent CON320 Tracing and Debugging for Containerized Services
re:Invent CON320 Tracing and Debugging for Containerized Servicesre:Invent CON320 Tracing and Debugging for Containerized Services
re:Invent CON320 Tracing and Debugging for Containerized Services
 
AI APIs as a Catalyst for Machine Learning Initiatives
AI APIs as a Catalyst for Machine Learning InitiativesAI APIs as a Catalyst for Machine Learning Initiatives
AI APIs as a Catalyst for Machine Learning Initiatives
 
Supercharge Your Machine Learning Model with Amazon SageMaker
Supercharge Your Machine Learning Model with Amazon SageMakerSupercharge Your Machine Learning Model with Amazon SageMaker
Supercharge Your Machine Learning Model with Amazon SageMaker
 
Introduction to AI services for Developers - Builders Day Israel
Introduction to AI services for Developers - Builders Day IsraelIntroduction to AI services for Developers - Builders Day Israel
Introduction to AI services for Developers - Builders Day Israel
 
Introduction to AI services for Developers - Builders Day Israel
Introduction to AI services for Developers - Builders Day IsraelIntroduction to AI services for Developers - Builders Day Israel
Introduction to AI services for Developers - Builders Day Israel
 
Racing with Artificial Intelligence
Racing with Artificial IntelligenceRacing with Artificial Intelligence
Racing with Artificial Intelligence
 
MXNet Paris Workshop - Intro To MXNet
MXNet Paris Workshop - Intro To MXNetMXNet Paris Workshop - Intro To MXNet
MXNet Paris Workshop - Intro To MXNet
 
AWS Toronto Summit 2019 - AIM302 - Build, train, and deploy ML models with Am...
AWS Toronto Summit 2019 - AIM302 - Build, train, and deploy ML models with Am...AWS Toronto Summit 2019 - AIM302 - Build, train, and deploy ML models with Am...
AWS Toronto Summit 2019 - AIM302 - Build, train, and deploy ML models with Am...
 
Frontend and Mobile with AWS Amplify | AWS Summit Tel Aviv 2019
Frontend and Mobile with AWS Amplify | AWS Summit Tel Aviv 2019Frontend and Mobile with AWS Amplify | AWS Summit Tel Aviv 2019
Frontend and Mobile with AWS Amplify | AWS Summit Tel Aviv 2019
 
Frontend and Mobile with AWS Amplify | AWS Summit Tel Aviv 2019
Frontend and Mobile with AWS Amplify | AWS Summit Tel Aviv 2019Frontend and Mobile with AWS Amplify | AWS Summit Tel Aviv 2019
Frontend and Mobile with AWS Amplify | AWS Summit Tel Aviv 2019
 
AWS Initiate Day Manchester 2019 – Moving to DevOps the Amazon Way
AWS Initiate Day Manchester 2019 – Moving to DevOps the Amazon WayAWS Initiate Day Manchester 2019 – Moving to DevOps the Amazon Way
AWS Initiate Day Manchester 2019 – Moving to DevOps the Amazon Way
 
Moving to DevOps the Amazon Way
Moving to DevOps the Amazon WayMoving to DevOps the Amazon Way
Moving to DevOps the Amazon Way
 
Debugging Gluon and Apache MXNet (AIM423) - AWS re:Invent 2018
Debugging Gluon and Apache MXNet (AIM423) - AWS re:Invent 2018Debugging Gluon and Apache MXNet (AIM423) - AWS re:Invent 2018
Debugging Gluon and Apache MXNet (AIM423) - AWS re:Invent 2018
 
Reinforcement Learning with Sagemaker, DeepRacer and Robomaker
Reinforcement Learning with Sagemaker, DeepRacer and RobomakerReinforcement Learning with Sagemaker, DeepRacer and Robomaker
Reinforcement Learning with Sagemaker, DeepRacer and Robomaker
 
Moving to DevOps the Amazon Way
Moving to DevOps the Amazon WayMoving to DevOps the Amazon Way
Moving to DevOps the Amazon Way
 
Building Volkswagen Group's Digital Ecosystem (AMT304) - AWS re:Invent 2018
Building Volkswagen Group's Digital Ecosystem (AMT304) - AWS re:Invent 2018Building Volkswagen Group's Digital Ecosystem (AMT304) - AWS re:Invent 2018
Building Volkswagen Group's Digital Ecosystem (AMT304) - AWS re:Invent 2018
 
Debugging and Performance tricks for MXNet Gluon
Debugging and Performance tricks for MXNet GluonDebugging and Performance tricks for MXNet Gluon
Debugging and Performance tricks for MXNet Gluon
 

More from Amazon Web Services Korea

AWS Modern Infra with Storage Roadshow 2023 - Day 2
AWS Modern Infra with Storage Roadshow 2023 - Day 2AWS Modern Infra with Storage Roadshow 2023 - Day 2
AWS Modern Infra with Storage Roadshow 2023 - Day 2Amazon Web Services Korea
 
AWS Modern Infra with Storage Roadshow 2023 - Day 1
AWS Modern Infra with Storage Roadshow 2023 - Day 1AWS Modern Infra with Storage Roadshow 2023 - Day 1
AWS Modern Infra with Storage Roadshow 2023 - Day 1Amazon Web Services Korea
 
사례로 알아보는 Database Migration Service : 데이터베이스 및 데이터 이관, 통합, 분리, 분석의 도구 - 발표자: ...
사례로 알아보는 Database Migration Service : 데이터베이스 및 데이터 이관, 통합, 분리, 분석의 도구 - 발표자: ...사례로 알아보는 Database Migration Service : 데이터베이스 및 데이터 이관, 통합, 분리, 분석의 도구 - 발표자: ...
사례로 알아보는 Database Migration Service : 데이터베이스 및 데이터 이관, 통합, 분리, 분석의 도구 - 발표자: ...Amazon Web Services Korea
 
Amazon DocumentDB - Architecture 및 Best Practice (Level 200) - 발표자: 장동훈, Sr. ...
Amazon DocumentDB - Architecture 및 Best Practice (Level 200) - 발표자: 장동훈, Sr. ...Amazon DocumentDB - Architecture 및 Best Practice (Level 200) - 발표자: 장동훈, Sr. ...
Amazon DocumentDB - Architecture 및 Best Practice (Level 200) - 발표자: 장동훈, Sr. ...Amazon Web Services Korea
 
Amazon Elasticache - Fully managed, Redis & Memcached Compatible Service (Lev...
Amazon Elasticache - Fully managed, Redis & Memcached Compatible Service (Lev...Amazon Elasticache - Fully managed, Redis & Memcached Compatible Service (Lev...
Amazon Elasticache - Fully managed, Redis & Memcached Compatible Service (Lev...Amazon Web Services Korea
 
Internal Architecture of Amazon Aurora (Level 400) - 발표자: 정달영, APAC RDS Speci...
Internal Architecture of Amazon Aurora (Level 400) - 발표자: 정달영, APAC RDS Speci...Internal Architecture of Amazon Aurora (Level 400) - 발표자: 정달영, APAC RDS Speci...
Internal Architecture of Amazon Aurora (Level 400) - 발표자: 정달영, APAC RDS Speci...Amazon Web Services Korea
 
[Keynote] 슬기로운 AWS 데이터베이스 선택하기 - 발표자: 강민석, Korea Database SA Manager, WWSO, A...
[Keynote] 슬기로운 AWS 데이터베이스 선택하기 - 발표자: 강민석, Korea Database SA Manager, WWSO, A...[Keynote] 슬기로운 AWS 데이터베이스 선택하기 - 발표자: 강민석, Korea Database SA Manager, WWSO, A...
[Keynote] 슬기로운 AWS 데이터베이스 선택하기 - 발표자: 강민석, Korea Database SA Manager, WWSO, A...Amazon Web Services Korea
 
Demystify Streaming on AWS - 발표자: 이종혁, Sr Analytics Specialist, WWSO, AWS :::...
Demystify Streaming on AWS - 발표자: 이종혁, Sr Analytics Specialist, WWSO, AWS :::...Demystify Streaming on AWS - 발표자: 이종혁, Sr Analytics Specialist, WWSO, AWS :::...
Demystify Streaming on AWS - 발표자: 이종혁, Sr Analytics Specialist, WWSO, AWS :::...Amazon Web Services Korea
 
Amazon EMR - Enhancements on Cost/Performance, Serverless - 발표자: 김기영, Sr Anal...
Amazon EMR - Enhancements on Cost/Performance, Serverless - 발표자: 김기영, Sr Anal...Amazon EMR - Enhancements on Cost/Performance, Serverless - 발표자: 김기영, Sr Anal...
Amazon EMR - Enhancements on Cost/Performance, Serverless - 발표자: 김기영, Sr Anal...Amazon Web Services Korea
 
Amazon OpenSearch - Use Cases, Security/Observability, Serverless and Enhance...
Amazon OpenSearch - Use Cases, Security/Observability, Serverless and Enhance...Amazon OpenSearch - Use Cases, Security/Observability, Serverless and Enhance...
Amazon OpenSearch - Use Cases, Security/Observability, Serverless and Enhance...Amazon Web Services Korea
 
Enabling Agility with Data Governance - 발표자: 김성연, Analytics Specialist, WWSO,...
Enabling Agility with Data Governance - 발표자: 김성연, Analytics Specialist, WWSO,...Enabling Agility with Data Governance - 발표자: 김성연, Analytics Specialist, WWSO,...
Enabling Agility with Data Governance - 발표자: 김성연, Analytics Specialist, WWSO,...Amazon Web Services Korea
 
Amazon Redshift Deep Dive - Serverless, Streaming, ML, Auto Copy (New feature...
Amazon Redshift Deep Dive - Serverless, Streaming, ML, Auto Copy (New feature...Amazon Redshift Deep Dive - Serverless, Streaming, ML, Auto Copy (New feature...
Amazon Redshift Deep Dive - Serverless, Streaming, ML, Auto Copy (New feature...Amazon Web Services Korea
 
From Insights to Action, How to build and maintain a Data Driven Organization...
From Insights to Action, How to build and maintain a Data Driven Organization...From Insights to Action, How to build and maintain a Data Driven Organization...
From Insights to Action, How to build and maintain a Data Driven Organization...Amazon Web Services Korea
 
[Keynote] Accelerating Business Outcomes with AWS Data - 발표자: Saeed Gharadagh...
[Keynote] Accelerating Business Outcomes with AWS Data - 발표자: Saeed Gharadagh...[Keynote] Accelerating Business Outcomes with AWS Data - 발표자: Saeed Gharadagh...
[Keynote] Accelerating Business Outcomes with AWS Data - 발표자: Saeed Gharadagh...Amazon Web Services Korea
 
Amazon DynamoDB - Use Cases and Cost Optimization - 발표자: 이혁, DynamoDB Special...
Amazon DynamoDB - Use Cases and Cost Optimization - 발표자: 이혁, DynamoDB Special...Amazon DynamoDB - Use Cases and Cost Optimization - 발표자: 이혁, DynamoDB Special...
Amazon DynamoDB - Use Cases and Cost Optimization - 발표자: 이혁, DynamoDB Special...Amazon Web Services Korea
 
LG전자 - Amazon Aurora 및 RDS 블루/그린 배포를 이용한 데이터베이스 업그레이드 안정성 확보 - 발표자: 이은경 책임, L...
LG전자 - Amazon Aurora 및 RDS 블루/그린 배포를 이용한 데이터베이스 업그레이드 안정성 확보 - 발표자: 이은경 책임, L...LG전자 - Amazon Aurora 및 RDS 블루/그린 배포를 이용한 데이터베이스 업그레이드 안정성 확보 - 발표자: 이은경 책임, L...
LG전자 - Amazon Aurora 및 RDS 블루/그린 배포를 이용한 데이터베이스 업그레이드 안정성 확보 - 발표자: 이은경 책임, L...Amazon Web Services Korea
 
KB국민카드 - 클라우드 기반 분석 플랫폼 혁신 여정 - 발표자: 박창용 과장, 데이터전략본부, AI혁신부, KB카드│강병억, Soluti...
KB국민카드 - 클라우드 기반 분석 플랫폼 혁신 여정 - 발표자: 박창용 과장, 데이터전략본부, AI혁신부, KB카드│강병억, Soluti...KB국민카드 - 클라우드 기반 분석 플랫폼 혁신 여정 - 발표자: 박창용 과장, 데이터전략본부, AI혁신부, KB카드│강병억, Soluti...
KB국민카드 - 클라우드 기반 분석 플랫폼 혁신 여정 - 발표자: 박창용 과장, 데이터전략본부, AI혁신부, KB카드│강병억, Soluti...Amazon Web Services Korea
 
SK Telecom - 망관리 프로젝트 TANGO의 오픈소스 데이터베이스 전환 여정 - 발표자 : 박승전, Project Manager, ...
SK Telecom - 망관리 프로젝트 TANGO의 오픈소스 데이터베이스 전환 여정 - 발표자 : 박승전, Project Manager, ...SK Telecom - 망관리 프로젝트 TANGO의 오픈소스 데이터베이스 전환 여정 - 발표자 : 박승전, Project Manager, ...
SK Telecom - 망관리 프로젝트 TANGO의 오픈소스 데이터베이스 전환 여정 - 발표자 : 박승전, Project Manager, ...Amazon Web Services Korea
 
코리안리 - 데이터 분석 플랫폼 구축 여정, 그 시작과 과제 - 발표자: 김석기 그룹장, 데이터비즈니스센터, 메가존클라우드 ::: AWS ...
코리안리 - 데이터 분석 플랫폼 구축 여정, 그 시작과 과제 - 발표자: 김석기 그룹장, 데이터비즈니스센터, 메가존클라우드 ::: AWS ...코리안리 - 데이터 분석 플랫폼 구축 여정, 그 시작과 과제 - 발표자: 김석기 그룹장, 데이터비즈니스센터, 메가존클라우드 ::: AWS ...
코리안리 - 데이터 분석 플랫폼 구축 여정, 그 시작과 과제 - 발표자: 김석기 그룹장, 데이터비즈니스센터, 메가존클라우드 ::: AWS ...Amazon Web Services Korea
 
LG 이노텍 - Amazon Redshift Serverless를 활용한 데이터 분석 플랫폼 혁신 과정 - 발표자: 유재상 선임, LG이노...
LG 이노텍 - Amazon Redshift Serverless를 활용한 데이터 분석 플랫폼 혁신 과정 - 발표자: 유재상 선임, LG이노...LG 이노텍 - Amazon Redshift Serverless를 활용한 데이터 분석 플랫폼 혁신 과정 - 발표자: 유재상 선임, LG이노...
LG 이노텍 - Amazon Redshift Serverless를 활용한 데이터 분석 플랫폼 혁신 과정 - 발표자: 유재상 선임, LG이노...Amazon Web Services Korea
 

More from Amazon Web Services Korea (20)

AWS Modern Infra with Storage Roadshow 2023 - Day 2
AWS Modern Infra with Storage Roadshow 2023 - Day 2AWS Modern Infra with Storage Roadshow 2023 - Day 2
AWS Modern Infra with Storage Roadshow 2023 - Day 2
 
AWS Modern Infra with Storage Roadshow 2023 - Day 1
AWS Modern Infra with Storage Roadshow 2023 - Day 1AWS Modern Infra with Storage Roadshow 2023 - Day 1
AWS Modern Infra with Storage Roadshow 2023 - Day 1
 
사례로 알아보는 Database Migration Service : 데이터베이스 및 데이터 이관, 통합, 분리, 분석의 도구 - 발표자: ...
사례로 알아보는 Database Migration Service : 데이터베이스 및 데이터 이관, 통합, 분리, 분석의 도구 - 발표자: ...사례로 알아보는 Database Migration Service : 데이터베이스 및 데이터 이관, 통합, 분리, 분석의 도구 - 발표자: ...
사례로 알아보는 Database Migration Service : 데이터베이스 및 데이터 이관, 통합, 분리, 분석의 도구 - 발표자: ...
 
Amazon DocumentDB - Architecture 및 Best Practice (Level 200) - 발표자: 장동훈, Sr. ...
Amazon DocumentDB - Architecture 및 Best Practice (Level 200) - 발표자: 장동훈, Sr. ...Amazon DocumentDB - Architecture 및 Best Practice (Level 200) - 발표자: 장동훈, Sr. ...
Amazon DocumentDB - Architecture 및 Best Practice (Level 200) - 발표자: 장동훈, Sr. ...
 
Amazon Elasticache - Fully managed, Redis & Memcached Compatible Service (Lev...
Amazon Elasticache - Fully managed, Redis & Memcached Compatible Service (Lev...Amazon Elasticache - Fully managed, Redis & Memcached Compatible Service (Lev...
Amazon Elasticache - Fully managed, Redis & Memcached Compatible Service (Lev...
 
Internal Architecture of Amazon Aurora (Level 400) - 발표자: 정달영, APAC RDS Speci...
Internal Architecture of Amazon Aurora (Level 400) - 발표자: 정달영, APAC RDS Speci...Internal Architecture of Amazon Aurora (Level 400) - 발표자: 정달영, APAC RDS Speci...
Internal Architecture of Amazon Aurora (Level 400) - 발표자: 정달영, APAC RDS Speci...
 
[Keynote] 슬기로운 AWS 데이터베이스 선택하기 - 발표자: 강민석, Korea Database SA Manager, WWSO, A...
[Keynote] 슬기로운 AWS 데이터베이스 선택하기 - 발표자: 강민석, Korea Database SA Manager, WWSO, A...[Keynote] 슬기로운 AWS 데이터베이스 선택하기 - 발표자: 강민석, Korea Database SA Manager, WWSO, A...
[Keynote] 슬기로운 AWS 데이터베이스 선택하기 - 발표자: 강민석, Korea Database SA Manager, WWSO, A...
 
Demystify Streaming on AWS - 발표자: 이종혁, Sr Analytics Specialist, WWSO, AWS :::...
Demystify Streaming on AWS - 발표자: 이종혁, Sr Analytics Specialist, WWSO, AWS :::...Demystify Streaming on AWS - 발표자: 이종혁, Sr Analytics Specialist, WWSO, AWS :::...
Demystify Streaming on AWS - 발표자: 이종혁, Sr Analytics Specialist, WWSO, AWS :::...
 
Amazon EMR - Enhancements on Cost/Performance, Serverless - 발표자: 김기영, Sr Anal...
Amazon EMR - Enhancements on Cost/Performance, Serverless - 발표자: 김기영, Sr Anal...Amazon EMR - Enhancements on Cost/Performance, Serverless - 발표자: 김기영, Sr Anal...
Amazon EMR - Enhancements on Cost/Performance, Serverless - 발표자: 김기영, Sr Anal...
 
Amazon OpenSearch - Use Cases, Security/Observability, Serverless and Enhance...
Amazon OpenSearch - Use Cases, Security/Observability, Serverless and Enhance...Amazon OpenSearch - Use Cases, Security/Observability, Serverless and Enhance...
Amazon OpenSearch - Use Cases, Security/Observability, Serverless and Enhance...
 
Enabling Agility with Data Governance - 발표자: 김성연, Analytics Specialist, WWSO,...
Enabling Agility with Data Governance - 발표자: 김성연, Analytics Specialist, WWSO,...Enabling Agility with Data Governance - 발표자: 김성연, Analytics Specialist, WWSO,...
Enabling Agility with Data Governance - 발표자: 김성연, Analytics Specialist, WWSO,...
 
Amazon Redshift Deep Dive - Serverless, Streaming, ML, Auto Copy (New feature...
Amazon Redshift Deep Dive - Serverless, Streaming, ML, Auto Copy (New feature...Amazon Redshift Deep Dive - Serverless, Streaming, ML, Auto Copy (New feature...
Amazon Redshift Deep Dive - Serverless, Streaming, ML, Auto Copy (New feature...
 
From Insights to Action, How to build and maintain a Data Driven Organization...
From Insights to Action, How to build and maintain a Data Driven Organization...From Insights to Action, How to build and maintain a Data Driven Organization...
From Insights to Action, How to build and maintain a Data Driven Organization...
 
[Keynote] Accelerating Business Outcomes with AWS Data - 발표자: Saeed Gharadagh...
[Keynote] Accelerating Business Outcomes with AWS Data - 발표자: Saeed Gharadagh...[Keynote] Accelerating Business Outcomes with AWS Data - 발표자: Saeed Gharadagh...
[Keynote] Accelerating Business Outcomes with AWS Data - 발표자: Saeed Gharadagh...
 
Amazon DynamoDB - Use Cases and Cost Optimization - 발표자: 이혁, DynamoDB Special...
Amazon DynamoDB - Use Cases and Cost Optimization - 발표자: 이혁, DynamoDB Special...Amazon DynamoDB - Use Cases and Cost Optimization - 발표자: 이혁, DynamoDB Special...
Amazon DynamoDB - Use Cases and Cost Optimization - 발표자: 이혁, DynamoDB Special...
 
LG전자 - Amazon Aurora 및 RDS 블루/그린 배포를 이용한 데이터베이스 업그레이드 안정성 확보 - 발표자: 이은경 책임, L...
LG전자 - Amazon Aurora 및 RDS 블루/그린 배포를 이용한 데이터베이스 업그레이드 안정성 확보 - 발표자: 이은경 책임, L...LG전자 - Amazon Aurora 및 RDS 블루/그린 배포를 이용한 데이터베이스 업그레이드 안정성 확보 - 발표자: 이은경 책임, L...
LG전자 - Amazon Aurora 및 RDS 블루/그린 배포를 이용한 데이터베이스 업그레이드 안정성 확보 - 발표자: 이은경 책임, L...
 
KB국민카드 - 클라우드 기반 분석 플랫폼 혁신 여정 - 발표자: 박창용 과장, 데이터전략본부, AI혁신부, KB카드│강병억, Soluti...
KB국민카드 - 클라우드 기반 분석 플랫폼 혁신 여정 - 발표자: 박창용 과장, 데이터전략본부, AI혁신부, KB카드│강병억, Soluti...KB국민카드 - 클라우드 기반 분석 플랫폼 혁신 여정 - 발표자: 박창용 과장, 데이터전략본부, AI혁신부, KB카드│강병억, Soluti...
KB국민카드 - 클라우드 기반 분석 플랫폼 혁신 여정 - 발표자: 박창용 과장, 데이터전략본부, AI혁신부, KB카드│강병억, Soluti...
 
SK Telecom - 망관리 프로젝트 TANGO의 오픈소스 데이터베이스 전환 여정 - 발표자 : 박승전, Project Manager, ...
SK Telecom - 망관리 프로젝트 TANGO의 오픈소스 데이터베이스 전환 여정 - 발표자 : 박승전, Project Manager, ...SK Telecom - 망관리 프로젝트 TANGO의 오픈소스 데이터베이스 전환 여정 - 발표자 : 박승전, Project Manager, ...
SK Telecom - 망관리 프로젝트 TANGO의 오픈소스 데이터베이스 전환 여정 - 발표자 : 박승전, Project Manager, ...
 
코리안리 - 데이터 분석 플랫폼 구축 여정, 그 시작과 과제 - 발표자: 김석기 그룹장, 데이터비즈니스센터, 메가존클라우드 ::: AWS ...
코리안리 - 데이터 분석 플랫폼 구축 여정, 그 시작과 과제 - 발표자: 김석기 그룹장, 데이터비즈니스센터, 메가존클라우드 ::: AWS ...코리안리 - 데이터 분석 플랫폼 구축 여정, 그 시작과 과제 - 발표자: 김석기 그룹장, 데이터비즈니스센터, 메가존클라우드 ::: AWS ...
코리안리 - 데이터 분석 플랫폼 구축 여정, 그 시작과 과제 - 발표자: 김석기 그룹장, 데이터비즈니스센터, 메가존클라우드 ::: AWS ...
 
LG 이노텍 - Amazon Redshift Serverless를 활용한 데이터 분석 플랫폼 혁신 과정 - 발표자: 유재상 선임, LG이노...
LG 이노텍 - Amazon Redshift Serverless를 활용한 데이터 분석 플랫폼 혁신 과정 - 발표자: 유재상 선임, LG이노...LG 이노텍 - Amazon Redshift Serverless를 활용한 데이터 분석 플랫폼 혁신 과정 - 발표자: 유재상 선임, LG이노...
LG 이노텍 - Amazon Redshift Serverless를 활용한 데이터 분석 플랫폼 혁신 과정 - 발표자: 유재상 선임, LG이노...
 

Recently uploaded

Benefits Of Flutter Compared To Other Frameworks
Benefits Of Flutter Compared To Other FrameworksBenefits Of Flutter Compared To Other Frameworks
Benefits Of Flutter Compared To Other FrameworksSoftradix Technologies
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationRadu Cotescu
 
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Patryk Bandurski
 
Breaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountBreaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountPuma Security, LLC
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationSafe Software
 
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 3652toLead Limited
 
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j
 
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...shyamraj55
 
Pigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping ElbowsPigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping ElbowsPigging Solutions
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking MenDelhi Call girls
 
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking MenDelhi Call girls
 
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024BookNet Canada
 
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure serviceWhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure servicePooja Nehwal
 
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxFactors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxKatpro Technologies
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationMichael W. Hawkins
 
Understanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitectureUnderstanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitecturePixlogix Infotech
 
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | DelhiFULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhisoniya singh
 
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationBeyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationSafe Software
 
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmaticsKotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmaticscarlostorres15106
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking MenDelhi Call girls
 

Recently uploaded (20)

Benefits Of Flutter Compared To Other Frameworks
Benefits Of Flutter Compared To Other FrameworksBenefits Of Flutter Compared To Other Frameworks
Benefits Of Flutter Compared To Other Frameworks
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organization
 
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
 
Breaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path MountBreaking the Kubernetes Kill Chain: Host Path Mount
Breaking the Kubernetes Kill Chain: Host Path Mount
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
 
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
 
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
 
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
 
Pigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping ElbowsPigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping Elbows
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men
 
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
 
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
 
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure serviceWhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
 
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxFactors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day Presentation
 
Understanding the Laravel MVC Architecture
Understanding the Laravel MVC ArchitectureUnderstanding the Laravel MVC Architecture
Understanding the Laravel MVC Architecture
 
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | DelhiFULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
FULL ENJOY 🔝 8264348440 🔝 Call Girls in Diplomatic Enclave | Delhi
 
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationBeyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
 
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmaticsKotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men
 

Apache MXNet/Gluon을 이용한 입술 읽기(Lipreading) 모델 만들기 - 김형준, SK텔레콤 :: AWS Summit Seoul 2019

  • 1. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Apache MXNet/Gluon을 이용한 입술 읽기(Lip Reading) 모델 만들기 김형준 Research Engineer SK Telecom
  • 2. 발표자료 바로 공개 발표자료는 발표 종료 후 해당 사이트에서 바로 보실 수 있습니다 © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
  • 3. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Who am I 김 형 준 - AI Research Engineer - AI/Machine Learning Application Engineer - Data Analysis Platform Engineer - Data Scientist
  • 4. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. SK Telecom AI Center
  • 5. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Open Source & Community (2018.07 ~ ) Apache MXNet Medium(Tech Blog) Writer & Code Contributor (2018.06) A way to Benchmark Your Deep Learning Framework, Psygrams (2018.01) Sentiment Analysis with grad-CAM and Target2Vec, Microsoft X 바벨피쉬 (2017.10) Machine Learning Platform Development, 데이터야놀자 (2015.12) Anomaly Detection using MRFs, Naver D2 개발자 커뮤니티 (2015.06) Sentiment Analysis, R User Conference in Korea 2015
  • 6. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
  • 7. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Contribution
  • 8. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. MXNet/Gluon Model
  • 9. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Agenda 1. 도움이 된 것들 2. 처음 Pull Request 하던 날 3. Lip Model 예제 작성 4. Review & Merged 5. gluonCV, gluonNLP 6. Discussion
  • 10. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
  • 11. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. MODEL: Deep Learning Frameworks https://www.slideshare.net/awskorea/2-deep-learning-with-gluon * Low level-control - Hybridized Gluon(Symbolic) * High level - Gluon (Imperative) * GluonCV, Gluon-NLP * Semantic Versioning - MXNet Community
  • 12. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. MODEL: MXNet/Gluon Study (17.12 ~ 18.06)
  • 13. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. MODEL: MXNet/Gluon Study (17.12 ~ 18.06) - (NLP) Word2vec (skip-gram) - (Vision) GAN - (Vision) Pix2Pix - (Vision) Grad-cam & Integrated-gradients - (NLP) End to end Memory Network - (NLP) LSTM with Custom Embedding
  • 14. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. MODEL: MXNet/Gluon - 때로는 다른 누군가가 만들어주더라도 - 때로는 실패하고
  • 15. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. MODEL: MXNet/Gluon - 성공했을 때의 경험은 잊을 수 없습니다
  • 16. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. SOSCON 2018(https://www.slideshare.net/awskorea/apache-mxnet-soscon-2018/) Contribution: Apache MXNet 공헌하는 방법
  • 17. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Contribution: Apache MXNet 공헌하는 방법 - Source Code Contribution : [MXNet Github](https://github.com/apache/incubator-mxnet) - Blog : [Medium Blog](https://medium.com/apache-mxnet) - Community : [MXNet Forum](https://discuss.mxnet.io/) [Slack – mxnet, AWSKRUG ] [Facebook – MXNet Korea User Group](https://www.facebook.com/groups/mxnetkr/)
  • 18. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Medium Blog (18.07) A way to Benchmark Your Deep Learning Framework On-premise
  • 19. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Medium Blog MXNet Makes Us Faster And Stronger!
  • 20. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Source Code Contribution [MXNet Github](https://github.com/apache/incubator-mxnet)
  • 21. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Source Code Contribution 오타 수정도 번역도 가이드 문서 작성도 디자인 작업도 의견 제시도 모두 컨트리뷰션입니다 “More Common Than You Think: An In-Depth Study of Casual Contributors”, 2016” https://naver.github.io/OpenSourceGuide/book/BetterContribution/why-contribute-to-open-source.html
  • 22. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Source Code Contribution
  • 23. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Source Code Contribution https://mxnet.apache.org/versions/master/community/contribute.html#setup-mxnet-for-development
  • 24. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Source Code Contribution [MXNet Github](https://github.com/apache/incubator-mxnet)
  • 25. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Source Code Contribution
  • 26. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Source Code Contribution 해당 이슈에 다음 행동들을 통해 기여해 주세요 - 의견 및 아이디어 제시 - 관심 어필 - Pull Request
  • 27. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
  • 28. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Bug Fix - Issues
  • 29. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Bug Fix - PR(Pull Request)
  • 30. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
  • 31. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet - https://arxiv.org/abs/1611.01599 - LipNet: End-to-End Sentence-level Lipreading (2016) - please 입술 읽기(Lip reading)는 화자의 입의 움직임으로부터 텍스트를 디코딩하는 과제입니다.
  • 32. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet - https://arxiv.org/abs/1611.01599 - LipNet: End-to-End Sentence-level Lipreading (2016)
  • 33. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. MXNet 공헌하기 - Example
  • 34. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet - https://github.com/apache/incubator-mxnet/tree/master/example/gluon/lipnet Data Collect Data Prepare - Example Model Build Model Train Model Inference utils./download_data.py utils/preprocess_data.py data_loader..py models/network.py trainer.py main.py Infer.py - 정승환, 김형준(SK Telecom) - 18.12.01 ~ 12.14 (약 2주간 개발) - 18.12.14 ~ 02.15 (약 2달간 리뷰)
  • 35. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet Data Collect - Example bbaf3s.mpg
  • 36. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet Data Collect - Example - 16GB - Multiple processes
  • 37. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet Data Prepare - Example - Using Face Landmark Detection(http://dlib .net/), It’s able to extract the mouth from a video. - Using ffmpeg, It changes 1 video to 75 frames
  • 38. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet Data Prepare - Example - Preprocessing : Video → cropped images - DataLoader MXNet/Gluon : Generator (nd.array)
  • 39. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet Data Prepare 전처리 시간이 너무 오래 걸림 • 54 hours  1 process • 1 hour  54 processes
  • 40. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet Data Prepare Public S3 이용
  • 41. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet Data Prepare “직접 다운로드 & 전처리” 또는 “전처리된 데이터 이용” - https://github.com/apache/incubator-mxnet/tree/master/example/gluon/lipnet
  • 42. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet - Model: (ST)CNN + GRU Model Build
  • 43. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet - Model: (ST)CNN + GRU Model Build https://arxiv.org/pdf/1611.01599.pdf
  • 44. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet - Model: (ST)CNN + GRU Model Build https://arxiv.org/pdf/1611.01599.pdf
  • 45. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet - Model: (ST)CNN + GRU Model Build https://arxiv.org/pdf/1611.01599.pdf
  • 46. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet - 학습 Model Train trainer = Train(config) trainer.build_model() trainer.load_dataloader() trainer.run()
  • 47. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet - 추론 Model Inference trainer = Train(config) trainer.build_model() trainer.load_dataloader() trainer.infer_batch(trainer.valid_dataloader)
  • 48. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet - 추론 예제 Model Inference
  • 49. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
  • 50. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet Pull Requests - (1) Fork - (2) New branch - (3) Code/Doc - (4) Test
  • 51. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet Pull Requests - (5) Fill in PR template
  • 52. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet Pull Requests - (6) Review https://github.com/apache/incubator-mxnet/pull/13647
  • 53. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet Pull Requests - (6) Review
  • 54. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet Pull Requests - (7) Merged
  • 55. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet Pull Requests - 고려사항들 요약 - pylint - huge data → s3 - Multi gpus - Hybridize - Test code - Pretrained model
  • 56. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet Pull Requests - pylint
  • 57. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet Pull Requests - huge data → s3
  • 58. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet Pull Requests - Multi gpus
  • 59. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet Pull Requests - Hybridize https://github.com/apache/incubator-mxnet/pull/13647/commits/f487255be48f89231ba917c570266af759cc1116
  • 60. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet Pull Requests - Test code
  • 61. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet Pull Requests - Pretrained model
  • 62. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet Pull Requests - 실수?! - https://github.com/apache/incubator- mxnet/tree/master/example/gluon/lipnet
  • 63. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. LipNet Pull Requests 가장 힘들었던 점 - 미국과 한국의 시간 차 - 의사소통: 영어 - 체력 & 끈기: 과도한 요구? 이쯤이면 ‘approve’ 해 줍시다
  • 64. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
  • 65. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. GluonCV - Model zoo
  • 66. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. GluonCV - Model zoo
  • 67. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. GluonCV - Model zoo
  • 68. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. GluonNLP - Model zoo
  • 69. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
  • 70. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Lip Reading - Large-scale Visual Speech Recognition : https://arxiv.org/pdf/1807.05162.pdf
  • 71. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Lip Reading - Large-scale Visual Speech Recognition : https://arxiv.org/pdf/1807.05162.pdf
  • 72. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Why Contribution? - 시간 & 노력 vs 보람, 성취감, 자부심 - 실력 향상 - 최고의 실력자들로 부터 피드백 - 협업 - 테스트 코드 - 컨벤션 - “스프린트 / 핸즈업 / 밋업” 통한 사용자 확대 및 인적 네트워크 - 타인에게 Inspire - 스터디 - 발표 - 스프린트 / 핸즈업 / 밋업
  • 73. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Life is too short, Dive into Deep Learning! http://ko.d2l.ai/
  • 74. 여러분의 피드백을 기다립니다! #AWSSummit 해시태그로 소셜미디어에 여러분의 행사소감을 올려주세요. AWS Summit Seoul 2019 모바일 앱과 QR코드를 통해 강연평가 및 설문조사에 참여하시고 재미있는 기념품을 받아가세요. 내년 Summit을 만들 여러분의 소중한 의견 부탁 드립니다. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.