Submit Search
Upload
GluonCV
•
Download as PPTX, PDF
•
0 likes
•
126 views
S
Soji Adeshina
Follow
Amazon AI Conclave Workshops
Read less
Read more
Technology
Report
Share
Report
Share
1 of 25
Download now
Recommended
Challenges Encountered by Scaling Up Recommendation Services at Gravity R&D
Challenges Encountered by Scaling Up Recommendation Services at Gravity R&D
Domonkos Tikk
"A Shallow Dive into Training Deep Neural Networks," a Presentation from Deep...
"A Shallow Dive into Training Deep Neural Networks," a Presentation from Deep...
Edge AI and Vision Alliance
Lego blocks and pieces stacked on top of one another process 6 stages style ...
Lego blocks and pieces stacked on top of one another process 6 stages style ...
SlideTeam.net
Scalable Deep Learning on AWS with Apache MXNet
Scalable Deep Learning on AWS with Apache MXNet
Julien SIMON
Autoscaling near-persistent EBS
Autoscaling near-persistent EBS
Emil Philips
OpenStack in the Enterprise - NJ VMUG June 9, 2015 - Melissa Palmer
OpenStack in the Enterprise - NJ VMUG June 9, 2015 - Melissa Palmer
vmiss33
Introduction to GluonCV
Introduction to GluonCV
Apache MXNet
Emotion recognition in images: from idea to a model in production - Nordic DS...
Emotion recognition in images: from idea to a model in production - Nordic DS...
Hagay Lupesko
Recommended
Challenges Encountered by Scaling Up Recommendation Services at Gravity R&D
Challenges Encountered by Scaling Up Recommendation Services at Gravity R&D
Domonkos Tikk
"A Shallow Dive into Training Deep Neural Networks," a Presentation from Deep...
"A Shallow Dive into Training Deep Neural Networks," a Presentation from Deep...
Edge AI and Vision Alliance
Lego blocks and pieces stacked on top of one another process 6 stages style ...
Lego blocks and pieces stacked on top of one another process 6 stages style ...
SlideTeam.net
Scalable Deep Learning on AWS with Apache MXNet
Scalable Deep Learning on AWS with Apache MXNet
Julien SIMON
Autoscaling near-persistent EBS
Autoscaling near-persistent EBS
Emil Philips
OpenStack in the Enterprise - NJ VMUG June 9, 2015 - Melissa Palmer
OpenStack in the Enterprise - NJ VMUG June 9, 2015 - Melissa Palmer
vmiss33
Introduction to GluonCV
Introduction to GluonCV
Apache MXNet
Emotion recognition in images: from idea to a model in production - Nordic DS...
Emotion recognition in images: from idea to a model in production - Nordic DS...
Hagay Lupesko
MCL310_Building Deep Learning Applications with Apache MXNet and Gluon
MCL310_Building Deep Learning Applications with Apache MXNet and Gluon
Amazon Web Services
What is deep learning (and why you should care) - Talk at SJSU Oct 2018
What is deep learning (and why you should care) - Talk at SJSU Oct 2018
Hagay Lupesko
Maschinelles Lernen auf AWS für Entwickler, Data Scientists und Experten
Maschinelles Lernen auf AWS für Entwickler, Data Scientists und Experten
AWS Germany
MCL303-Deep Learning with Apache MXNet and Gluon
MCL303-Deep Learning with Apache MXNet and Gluon
Amazon Web Services
Machine Learning Models with Apache MXNet and AWS Fargate
Machine Learning Models with Apache MXNet and AWS Fargate
Amazon Web Services
LFS301-SAGE Bionetworks, Digital Mammography DREAM Challenge and How AWS Enab...
LFS301-SAGE Bionetworks, Digital Mammography DREAM Challenge and How AWS Enab...
Amazon Web Services
Deep Learning Using Caffe2 on AWS - MCL313 - re:Invent 2017
Deep Learning Using Caffe2 on AWS - MCL313 - re:Invent 2017
Amazon Web Services
From Notebook to production with Amazon SageMaker
From Notebook to production with Amazon SageMaker
Amazon Web Services
Training Chatbots and Conversational Artificial Intelligence Agents with Amaz...
Training Chatbots and Conversational Artificial Intelligence Agents with Amaz...
Amazon Web Services
DevOps on AWS
DevOps on AWS
Amazon Web Services
DevOps on AWS
DevOps on AWS
Amazon Web Services
Model Serving for Deep Learning with MXNet Model Server
Model Serving for Deep Learning with MXNet Model Server
Amazon Web Services
Practical Artificial Intelligence: Deep Learning Beyond Cats and Cars
Practical Artificial Intelligence: Deep Learning Beyond Cats and Cars
Alexey Rybakov
Emotion Recognition in Images
Emotion Recognition in Images
Apache MXNet
"Deep Learning Beyond Cats and Cars: Developing a Real-life DNN-based Embedde...
"Deep Learning Beyond Cats and Cars: Developing a Real-life DNN-based Embedde...
Edge AI and Vision Alliance
Issues in AI product development and practices in audio applications
Issues in AI product development and practices in audio applications
Taesu Kim
Build Deep Learning Applications Using Apache MXNet - Featuring Chick-fil-A (...
Build Deep Learning Applications Using Apache MXNet - Featuring Chick-fil-A (...
Amazon Web Services
Deep Learning Workshop
Deep Learning Workshop
Amazon Web Services
Amazon SageMaker (December 2018)
Amazon SageMaker (December 2018)
Julien SIMON
Julien Simon, Principal Technical Evangelist at Amazon - Machine Learning: Fr...
Julien Simon, Principal Technical Evangelist at Amazon - Machine Learning: Fr...
Codiax
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
Delhi Call girls
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
carlostorres15106
More Related Content
Similar to GluonCV
MCL310_Building Deep Learning Applications with Apache MXNet and Gluon
MCL310_Building Deep Learning Applications with Apache MXNet and Gluon
Amazon Web Services
What is deep learning (and why you should care) - Talk at SJSU Oct 2018
What is deep learning (and why you should care) - Talk at SJSU Oct 2018
Hagay Lupesko
Maschinelles Lernen auf AWS für Entwickler, Data Scientists und Experten
Maschinelles Lernen auf AWS für Entwickler, Data Scientists und Experten
AWS Germany
MCL303-Deep Learning with Apache MXNet and Gluon
MCL303-Deep Learning with Apache MXNet and Gluon
Amazon Web Services
Machine Learning Models with Apache MXNet and AWS Fargate
Machine Learning Models with Apache MXNet and AWS Fargate
Amazon Web Services
LFS301-SAGE Bionetworks, Digital Mammography DREAM Challenge and How AWS Enab...
LFS301-SAGE Bionetworks, Digital Mammography DREAM Challenge and How AWS Enab...
Amazon Web Services
Deep Learning Using Caffe2 on AWS - MCL313 - re:Invent 2017
Deep Learning Using Caffe2 on AWS - MCL313 - re:Invent 2017
Amazon Web Services
From Notebook to production with Amazon SageMaker
From Notebook to production with Amazon SageMaker
Amazon Web Services
Training Chatbots and Conversational Artificial Intelligence Agents with Amaz...
Training Chatbots and Conversational Artificial Intelligence Agents with Amaz...
Amazon Web Services
DevOps on AWS
DevOps on AWS
Amazon Web Services
DevOps on AWS
DevOps on AWS
Amazon Web Services
Model Serving for Deep Learning with MXNet Model Server
Model Serving for Deep Learning with MXNet Model Server
Amazon Web Services
Practical Artificial Intelligence: Deep Learning Beyond Cats and Cars
Practical Artificial Intelligence: Deep Learning Beyond Cats and Cars
Alexey Rybakov
Emotion Recognition in Images
Emotion Recognition in Images
Apache MXNet
"Deep Learning Beyond Cats and Cars: Developing a Real-life DNN-based Embedde...
"Deep Learning Beyond Cats and Cars: Developing a Real-life DNN-based Embedde...
Edge AI and Vision Alliance
Issues in AI product development and practices in audio applications
Issues in AI product development and practices in audio applications
Taesu Kim
Build Deep Learning Applications Using Apache MXNet - Featuring Chick-fil-A (...
Build Deep Learning Applications Using Apache MXNet - Featuring Chick-fil-A (...
Amazon Web Services
Deep Learning Workshop
Deep Learning Workshop
Amazon Web Services
Amazon SageMaker (December 2018)
Amazon SageMaker (December 2018)
Julien SIMON
Julien Simon, Principal Technical Evangelist at Amazon - Machine Learning: Fr...
Julien Simon, Principal Technical Evangelist at Amazon - Machine Learning: Fr...
Codiax
Similar to GluonCV
(20)
MCL310_Building Deep Learning Applications with Apache MXNet and Gluon
MCL310_Building Deep Learning Applications with Apache MXNet and Gluon
What is deep learning (and why you should care) - Talk at SJSU Oct 2018
What is deep learning (and why you should care) - Talk at SJSU Oct 2018
Maschinelles Lernen auf AWS für Entwickler, Data Scientists und Experten
Maschinelles Lernen auf AWS für Entwickler, Data Scientists und Experten
MCL303-Deep Learning with Apache MXNet and Gluon
MCL303-Deep Learning with Apache MXNet and Gluon
Machine Learning Models with Apache MXNet and AWS Fargate
Machine Learning Models with Apache MXNet and AWS Fargate
LFS301-SAGE Bionetworks, Digital Mammography DREAM Challenge and How AWS Enab...
LFS301-SAGE Bionetworks, Digital Mammography DREAM Challenge and How AWS Enab...
Deep Learning Using Caffe2 on AWS - MCL313 - re:Invent 2017
Deep Learning Using Caffe2 on AWS - MCL313 - re:Invent 2017
From Notebook to production with Amazon SageMaker
From Notebook to production with Amazon SageMaker
Training Chatbots and Conversational Artificial Intelligence Agents with Amaz...
Training Chatbots and Conversational Artificial Intelligence Agents with Amaz...
DevOps on AWS
DevOps on AWS
DevOps on AWS
DevOps on AWS
Model Serving for Deep Learning with MXNet Model Server
Model Serving for Deep Learning with MXNet Model Server
Practical Artificial Intelligence: Deep Learning Beyond Cats and Cars
Practical Artificial Intelligence: Deep Learning Beyond Cats and Cars
Emotion Recognition in Images
Emotion Recognition in Images
"Deep Learning Beyond Cats and Cars: Developing a Real-life DNN-based Embedde...
"Deep Learning Beyond Cats and Cars: Developing a Real-life DNN-based Embedde...
Issues in AI product development and practices in audio applications
Issues in AI product development and practices in audio applications
Build Deep Learning Applications Using Apache MXNet - Featuring Chick-fil-A (...
Build Deep Learning Applications Using Apache MXNet - Featuring Chick-fil-A (...
Deep Learning Workshop
Deep Learning Workshop
Amazon SageMaker (December 2018)
Amazon SageMaker (December 2018)
Julien Simon, Principal Technical Evangelist at Amazon - Machine Learning: Fr...
Julien Simon, Principal Technical Evangelist at Amazon - Machine Learning: Fr...
Recently uploaded
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
Delhi Call girls
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
carlostorres15106
Slack Application Development 101 Slides
Slack Application Development 101 Slides
praypatel2
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
Pooja Nehwal
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
Delhi Call girls
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
CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):
comworks
Key Features Of Token Development (1).pptx
Key Features Of Token Development (1).pptx
LBM Solutions
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men
Delhi Call girls
Install Stable Diffusion in windows machine
Install Stable Diffusion in windows machine
Padma Pradeep
Maximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptx
OnBoard
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR Systems
Mark Billinghurst
Pigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping Elbows
Pigging Solutions
My Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 Presentation
Ridwan Fadjar
Pigging Solutions in Pet Food Manufacturing
Pigging Solutions in Pet Food Manufacturing
Pigging Solutions
How to Remove Document Management Hurdles with X-Docs?
How to Remove Document Management Hurdles with X-Docs?
XfilesPro
Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024
Scott Keck-Warren
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
BookNet Canada
The Codex of Business Writing Software for Real-World Solutions 2.pptx
The Codex of Business Writing Software for Real-World Solutions 2.pptx
Malak Abu Hammad
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Safe Software
Recently uploaded
(20)
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Slack Application Development 101 Slides
Slack Application Development 101 Slides
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
WhatsApp 9892124323 ✓Call Girls In Kalyan ( Mumbai ) secure service
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
08448380779 Call Girls In Diplomatic Enclave Women Seeking Men
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
Integration and Automation in Practice: CI/CD in Mule Integration and Automat...
CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):
Key Features Of Token Development (1).pptx
Key Features Of Token Development (1).pptx
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men
Install Stable Diffusion in windows machine
Install Stable Diffusion in windows machine
Maximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptx
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR Systems
Pigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping Elbows
My Hashitalk Indonesia April 2024 Presentation
My Hashitalk Indonesia April 2024 Presentation
Pigging Solutions in Pet Food Manufacturing
Pigging Solutions in Pet Food Manufacturing
How to Remove Document Management Hurdles with X-Docs?
How to Remove Document Management Hurdles with X-Docs?
Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
The Codex of Business Writing Software for Real-World Solutions 2.pptx
The Codex of Business Writing Software for Real-World Solutions 2.pptx
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
GluonCV
1.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. Soji Adeshina, Machine Learning Engineer, Amazon AI Computer Vision 101 - Gluon CV
2.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. Computer Vision Architectures for Image Classification : A brief Timeline
3.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. Convolution • Ideal for picking up on spatial patterns in data • Applied over and over again (layer after layer), you can create more abstracted spatial features • Inspired by experiments on visual cortex of a cat. • Can be run in parallel for really fast computations http://colah.github.io/posts/2014-07- Understanding-Convolutions/
4.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. LeNet 1995 • Challenge: Multiple convolutions blow up dimensionality • Solution: Pooling • AvgPooling/Subsampling - average over patches (works OK) • MaxPooling - pick the maximum over patches (much better)
5.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. AlexNet (Krizhevsky et al., 2012) • More convolutional layers • More channels • More filters • More data More computation
6.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. VGG (2014) + vs. • Want to reach receptive field of size k • Use one large filter (linear mix of many, then nonlinearity) • Use several small filters (many linear mixes of few) - has fewer parameters • Simonyan & Zisserman, 2014 find that deep and narrow wins Deep and Narrow or Wide and Shallow?
7.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. Fancy structures - Networks of networks • Compute different filters • Compose one big vector from all of them • Layer them iteratively Szegedy et al. arxiv.org/pdf/1409.4842v1.pdf Inception (2014)
8.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. Batch Norm (Ioffe et al., 2015) loss data • Loss occurs at last layer • Last layers learn quickly • Data is inserted at bottom layer • Bottom layers change - everything changes • Last layers need to relearn many times • Slow convergence • This is like covariate shift Can we avoid changing last layers while learning first layers?
9.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. Batch Norm (Ioffe et al., 2015) • Can we avoid changing last layers while learning first layers? • Fix mean and variance and adjust it separately mean variance
10.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. ResNet (He et al., 2015) • In regular layer simple function is given by f(x) = 0 • Key idea - ‘Taylor expansion’
11.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. DenseNet (Huang et al., 2016) • Simple Function • In ResNet ‘Taylor expansion’ ends after one term • In DenseNet use multiple steps
12.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. Gluon CV: Deep Learning Toolkit for Computer Vision
13.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. Why GluonCV? What is the biggest challenge you have ever encountered with deep learning?
14.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. Why GluonCV? What is the biggest challenge you have ever encountered with deep learning? “reproducing the best claimed results”
15.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. Real-world Stories Back to a period in 2016, the same ImageNet models trained by MXNet achieves on average 1% worse accuracy compared to Torch. Tried almost everything to debug, even developed a plugin to run Torch code inside MXNet so that it is easier to compare the results. Transcoding training images using 95 JPEG quality rather than 85 solved the problem.
16.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. Real-world Stories Using another open source DL framework, a similar problem happened: trained model accuracies cannot match previous internal version. Spent months to figure out why, with no clue. The order of data augmentation is different from previous version.
17.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. Starting from scratch can be hard • Even the most talented researchers will get blocked by trivial things. • Experiences and instincts might be your enemies in certain circumstances. • Training is time-consuming, initialization and augmentation is randomized, and tons of implementation details need to be taken care of. Debugging deep models is extremely difficult.
18.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. • Qualities of open-source implementations vary. • Languages, code styles, project structures, DL frameworks are mixed. • Personal projects tend to focusing on a specific task with specific datasets. It requires significant engineering efforts to adapt to your use case. • Community projects can be abandoned frequently. Embracing open source solutions can be difficult
19.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. What does GluonCV provide Reproduction of important papers in recent years Training scripts (as well as tuned hyper- parameters) to reproduce the results Considerate APIs and modules that are easy to follow and understand, so that experiments based on existing algorithms are less frustrating Community support, feel free to ask and discuss
20.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. What’s in GluonCV Image Classification • More than 20+ pre-trained ImageNet models(ResNet, MobileNet…) • We achieved the best accuracy using some of the most popular models(e.g., ResNet), compared with other frameworks
21.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. What’s in GluonCV • Object Detection • SSD and YOLOv3: fastest solution • Faster-RCNN, RFCN and FPN: slower but more accurate, especially for tiny objects • Mask-RCNN: simultaneous object detection and semantic segmentation
22.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. What’s in GluonCV Semantic Segmentation • FCN • PSPNet • Mask-RCNN • DeepLab Instance Segmentation • Mask-RCNN
23.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. What’s in GluonCV • Style Transfer • MSGNet • Generative Adversarial Networks (GAN) • CycleGAN
24.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved. Like GluonCV? https://gluon-cv.mxnet.io https://github.com/dmlc/gluon-cv
25.
© 2017, Amazon
Web Services, Inc. or its Affiliates. All rights reserved.
Editor's Notes
Won a nobel prize for this in 19
Download now