SlideShare a Scribd company logo
1 of 33
Download to read offline
Unearth the Journey of
Implementing Vision based
Deep Learning Solution using
Open Source Tools
Neethu Elizabeth Simon
Software Engineer
Intel Corporation
Phoenix, AZ
Java
Developer
C#
Developer
IoT Solutions Engineer
My Journey
Smart IoT Systems
Video – Eye of IoT
https://www.youtube.com/watch?v=nwPtcqcqz00
Dried Chili Peppers Cinnamon Sticks
Solution Data Set
Implementation
• Build Model
• Improve Models
to meet
performance &
accuracy
objectives
• Deploy Model
• Make Predictions
by applying the
Trained Model
on new Data
• Label Data• Data
Acquisition &
Organization
• Preprocessing
Collect Annotation
TrainingInference
Implementation – Data Collection
Collect Annotation
TrainingInference
 Data Acquisition & Organization
• Web Crawling/Scraping - Download pictures &
videos
Python Packages Request, Parcel, Beautiful Soup
Frameworks/Tools Scrapy, Pyspider, Mechanical Soup
Implementation – Data Collection
Collect Annotation
TrainingInference
 Data Acquisition & Organization
• Web Crawling/Scraping - Download pictures &
videos
• Live Video Recording/Image Capture from actual
retail store
Python Packages Request, Parcel, Beautiful Soup
Frameworks/Tools Scrapy, Pyspider, Mechanical Soup
Implementation – Data Collection
Collect Annotation
TrainingInference
 Pre-processing
• Data Cleanup
• Remove Bad/Unrelated Data
Implementation – Data Annotation
Collect Annotation
TrainingInference
 Label Data
• Establish ground truth
Implementation – Data Annotation
Collect Annotation
TrainingInference
 Open Source Tools
• Bbox - https://github.com/puzzledqs/BBox-Label-Tool
• LabelImg - https://github.com/tzutalin/labelImg
• VGG Image Annotator (VIA)
http://www.robots.ox.ac.uk/~vgg/software/via/
• Computer Vision Annotation Tool (CVAT)
https://github.com/opencv/cvat
 Paid Tools – Labelbox, Dataloop, RectLabel
Dried_chilli_pepper
Implementation – Training
Collect Annotation
TrainingInference
dried_chilli_pepper
cinnamon_stick
Lots of
Labeled Data !!!
Model Weights
Forward
Backward
Implementation – Training
Collect Annotation
TrainingInference
Data
• Type/Amount – Visual, Sensors
• Labeled Datasets – PASCAL VOC, COCO, KITTI, ImageNet or
ILSVRC, SUN
• Custom Data
Implementation – Training
Collect Annotation
TrainingInference
Compute Platform
• Complexity Determined by Data
• Cloud vs Edge
• Intel® Xeon vs Core vs Atom
Data
• Type/Amount – Visual, Sensors
• Labeled Datasets – PASCAL VOC, COCO, KITTI, ImageNet or
ILSVRC, SUN
• Custom Data
Intel
Core +
GPU
Implementation – Training
Collect Annotation
TrainingInference
Object Detection
Algorithm
• YOLO – You Only Look Once
• SSD, R-CNN, Fast R-CNN, Faster RCNN, R-FCN
Compute Platform
• Complexity Determined by Data
• Cloud vs Edge
• Intel® Xeon vs Core vs Atom
Data
• Type/Amount – Visual, Sensors
• Labeled Datasets – PASCAL VOC, COCO, KITTI, ImageNet or
ILSVRC, SUN
• Custom Data
Intel
Core +
GPU
YOLO
Implementation – Training
Collect Annotation
TrainingInference
Deep Learning
Framework
• DarkNet - C & CUDA - https://pjreddie.com/darknet/
• DarkFlow - Tensorflow implementation of DarkNet
https://github.com/thtrieu/darkflow
• Others – PyTorch, Caffe
Object Detection
Algorithm
• YOLO – You Only Look Once
• SSD, R-CNN, Fast R-CNN, Faster RCNN, R-FCN
Compute Platform
• Complexity Determined by Data
• Cloud vs Edge
• Intel® Xeon vs Core vs Atom
Data
• Type/Amount – Visual, Sensors
• Labeled Datasets – PASCAL VOC, COCO, KITTI, ImageNet or
ILSVRC, SUN
• Custom Data
YOLO
Intel
Core +
GPU
DarkFlow
Implementation – Training
Collect Annotation
TrainingInference
Deep Learning
Framework
• DarkNet - C & CUDA - https://pjreddie.com/darknet/
• DarkFlow - Tensorflow implementation of DarkNet
https://github.com/thtrieu/darkflow
• Others – PyTorch, Caffe
Object Detection
Algorithm
• YOLO – You Only Look Once
• SSD, R-CNN, Fast R-CNN, Faster RCNN, R-FCN
Compute Platform
• Complexity Determined by Data
• Cloud vs Edge
• Intel® Xeon vs Core vs Atom
Data
• Type/Amount – Visual, Sensors
• Labeled Datasets – PASCAL VOC, COCO, KITTI, ImageNet or
ILSVRC, SUN
• Custom Data
YOLO
Intel
Core +
GPU
DarkFlow
Images - 1000 Images
Training Time - 10 hrs
Model Size – 800 MB
Average loss – ~0.9
Implementation – Training
Collect Annotation
TrainingInferencedried_chilli_pepper
cinnamon_stick
Lots of
Labeled Data !!!
Model Weights
Forward
Backward
python flow --train --model cfg/yolov2-2c.cfg --load bin/yolov2.weights --annotation all_Spice_Annotations/ --dataset all_Spice_Images/
DarkFlow
Implementation – Inference
Collect Annotation
TrainingInference
Model Weights
Forward
dried_chilli_pepper
Edge Device – Laptop/NUC/Mobile
????
Intel® OpenVINO Toolkit
https://software.intel.com/en-
us/openvino-toolkit
Open Source Tool for Inference
Support Different Models
(Caffe/Tensorflow/MXNet/ONNX/Kaldi)
Model Optimization
Speed Deploymentpython flow --imgdir all_test_images/ --model cfg/yolov2-2c.cfg --load new_model_weights
DarkFlow
Results
[{"label": "cinnamon_stick", "confidence": 0.87, "topleft":
{"x": 42, "y": 22}, "bottomright": {"x": 208, "y": 113}}]
[{"label": "cinnamon_stick", "confidence": 0.06, "topleft":
{"x": 93, "y": 216}, "bottomright": {"x": 384, "y": 402}}]
Results
[{"label": "dried_chilli_pepper", "confidence": 0.95, "topleft": {"x": 0,
"y": 25}, "bottomright": {"x": 761, "y": 352}}]
[{"label": "dried_chilli_pepper", "confidence": 0.02, "topleft":
{"x": 357, "y": 93}, "bottomright": {"x": 513, "y": 446}}, {"label":
"cinnamon_stick", "confidence": 0.03, "topleft": {"x": 21, "y":
6}, "bottomright": {"x": 568, "y": 472}}]
Implementation Challenges
• More is Better
• Object Orientation, angle,
lighting
• Remove Bad/Unrelated
Data
• Require SME knowledge
to define bad data &
remove unrelated data
Collect Annotation
TrainingInference
[{"label": "cinnamon_stick", "confidence": 0.87, "topleft": {"x": 42, "y": 22}, "bottomright": {"x": 208, "y": 113}}]
Implementation Challenges
• Manual vs
Automation
• Time consuming
• Expensive
• Difficult to Scale
• More is Better
• Object Orientation,
angle, lighting
• Remove Bad/Unrelated
Data
• Require SME knowledge
to define bad data &
remove unrelated data
Collect Annotation
TrainingInference
Implementation Challenges
• Iterative Learning
• Varied Data
parameters &
tuning
• Hyperparamter
optimization & data
overfitting
• High Compute
• Manual vs
Automation
• Time consuming
• Expensive
• Difficult to Scale
• More is Better
• Object Orientation,
angle, lighting
• Remove Bad/Unrelated
Data
• Require SME knowledge
to define bad data &
remove unrelated data
Collect Annotation
TrainingInference
Implementation Challenges
• Iterative Learning
• Varied Data
parameters &
tuning
• Hyperparamter
optimization & data
overfitting
• High Compute
• Constantly Update
Models
• Complexity in Hardware
design – combination of
CPU/GPU/Accelerators
• Data Privacy & Security
• Manual vs
Automation
• Time consuming
• Expensive
• Difficult to Scale
• More is Better
• Object Orientation,
angle, lighting
• Remove Bad/Unrelated
Data
• Require SME knowledge
to define bad data &
remove unrelated data
Collect Annotation
TrainingInference
Neethu Elizabeth Simon
Software Engineer
Intel Corporation
Phoenix, AZ

More Related Content

What's hot

집단지성 프로그래밍 08-가격모델링
집단지성 프로그래밍 08-가격모델링집단지성 프로그래밍 08-가격모델링
집단지성 프로그래밍 08-가격모델링Kwang Woo NAM
 
Modularized Persistence - B Zsoldos
Modularized Persistence - B ZsoldosModularized Persistence - B Zsoldos
Modularized Persistence - B Zsoldosmfrancis
 
Firebase not really_yohoho
Firebase not really_yohohoFirebase not really_yohoho
Firebase not really_yohohoDA-14
 
Firebase_not_really_yohoho
Firebase_not_really_yohohoFirebase_not_really_yohoho
Firebase_not_really_yohohoRoman Sachenko
 
Managing a shared_mysql_farm_phpday2011
Managing a shared_mysql_farm_phpday2011Managing a shared_mysql_farm_phpday2011
Managing a shared_mysql_farm_phpday2011Combell NV
 
MongoDB World 2018: Time for a Change Stream - Using MongoDB Change Streams t...
MongoDB World 2018: Time for a Change Stream - Using MongoDB Change Streams t...MongoDB World 2018: Time for a Change Stream - Using MongoDB Change Streams t...
MongoDB World 2018: Time for a Change Stream - Using MongoDB Change Streams t...MongoDB
 
Whats New for WPF in .NET 4.5
Whats New for WPF in .NET 4.5Whats New for WPF in .NET 4.5
Whats New for WPF in .NET 4.5Rainer Stropek
 
Accelerating distributed joins in Apache Hive: Runtime filtering enhancements
Accelerating distributed joins in Apache Hive: Runtime filtering enhancementsAccelerating distributed joins in Apache Hive: Runtime filtering enhancements
Accelerating distributed joins in Apache Hive: Runtime filtering enhancementsPanagiotis Garefalakis
 
[WLDN] Supercharging word press development in 2018
[WLDN] Supercharging word press development in 2018[WLDN] Supercharging word press development in 2018
[WLDN] Supercharging word press development in 2018Adam Tomat
 
MongoDB World 2018: Using Change Streams to Keep Up with Your Data
MongoDB World 2018: Using Change Streams to Keep Up with Your DataMongoDB World 2018: Using Change Streams to Keep Up with Your Data
MongoDB World 2018: Using Change Streams to Keep Up with Your DataMongoDB
 
MongoDB World 2018: Keynote
MongoDB World 2018: KeynoteMongoDB World 2018: Keynote
MongoDB World 2018: KeynoteMongoDB
 
After max+phonegap
After max+phonegapAfter max+phonegap
After max+phonegapyangdj
 
Vielseitiges In-Memory Computing mit Apache Ignite und Kubernetes
Vielseitiges In-Memory Computing mit Apache Ignite und KubernetesVielseitiges In-Memory Computing mit Apache Ignite und Kubernetes
Vielseitiges In-Memory Computing mit Apache Ignite und KubernetesQAware GmbH
 
Introduction to Data Modeling with Apache Cassandra
Introduction to Data Modeling with Apache CassandraIntroduction to Data Modeling with Apache Cassandra
Introduction to Data Modeling with Apache CassandraDataStax Academy
 

What's hot (19)

집단지성 프로그래밍 08-가격모델링
집단지성 프로그래밍 08-가격모델링집단지성 프로그래밍 08-가격모델링
집단지성 프로그래밍 08-가격모델링
 
Letgo Data Platform: A global overview
Letgo Data Platform: A global overviewLetgo Data Platform: A global overview
Letgo Data Platform: A global overview
 
Modularized Persistence - B Zsoldos
Modularized Persistence - B ZsoldosModularized Persistence - B Zsoldos
Modularized Persistence - B Zsoldos
 
Firebase not really_yohoho
Firebase not really_yohohoFirebase not really_yohoho
Firebase not really_yohoho
 
Firebase_not_really_yohoho
Firebase_not_really_yohohoFirebase_not_really_yohoho
Firebase_not_really_yohoho
 
Managing a shared_mysql_farm_phpday2011
Managing a shared_mysql_farm_phpday2011Managing a shared_mysql_farm_phpday2011
Managing a shared_mysql_farm_phpday2011
 
MongoDB World 2018: Time for a Change Stream - Using MongoDB Change Streams t...
MongoDB World 2018: Time for a Change Stream - Using MongoDB Change Streams t...MongoDB World 2018: Time for a Change Stream - Using MongoDB Change Streams t...
MongoDB World 2018: Time for a Change Stream - Using MongoDB Change Streams t...
 
Whats New for WPF in .NET 4.5
Whats New for WPF in .NET 4.5Whats New for WPF in .NET 4.5
Whats New for WPF in .NET 4.5
 
Internet of things
Internet of thingsInternet of things
Internet of things
 
Advance MySQL Docstore Features
Advance MySQL Docstore FeaturesAdvance MySQL Docstore Features
Advance MySQL Docstore Features
 
RicoLiveGrid
RicoLiveGridRicoLiveGrid
RicoLiveGrid
 
Accelerating distributed joins in Apache Hive: Runtime filtering enhancements
Accelerating distributed joins in Apache Hive: Runtime filtering enhancementsAccelerating distributed joins in Apache Hive: Runtime filtering enhancements
Accelerating distributed joins in Apache Hive: Runtime filtering enhancements
 
[WLDN] Supercharging word press development in 2018
[WLDN] Supercharging word press development in 2018[WLDN] Supercharging word press development in 2018
[WLDN] Supercharging word press development in 2018
 
YouDrup_in_Drupal
YouDrup_in_DrupalYouDrup_in_Drupal
YouDrup_in_Drupal
 
MongoDB World 2018: Using Change Streams to Keep Up with Your Data
MongoDB World 2018: Using Change Streams to Keep Up with Your DataMongoDB World 2018: Using Change Streams to Keep Up with Your Data
MongoDB World 2018: Using Change Streams to Keep Up with Your Data
 
MongoDB World 2018: Keynote
MongoDB World 2018: KeynoteMongoDB World 2018: Keynote
MongoDB World 2018: Keynote
 
After max+phonegap
After max+phonegapAfter max+phonegap
After max+phonegap
 
Vielseitiges In-Memory Computing mit Apache Ignite und Kubernetes
Vielseitiges In-Memory Computing mit Apache Ignite und KubernetesVielseitiges In-Memory Computing mit Apache Ignite und Kubernetes
Vielseitiges In-Memory Computing mit Apache Ignite und Kubernetes
 
Introduction to Data Modeling with Apache Cassandra
Introduction to Data Modeling with Apache CassandraIntroduction to Data Modeling with Apache Cassandra
Introduction to Data Modeling with Apache Cassandra
 

Similar to Implementing Vision DL Solution using Open Source Tools

Smart Data Conference: DL4J and DataVec
Smart Data Conference: DL4J and DataVecSmart Data Conference: DL4J and DataVec
Smart Data Conference: DL4J and DataVecJosh Patterson
 
Josh Patterson, Advisor, Skymind – Deep learning for Industry at MLconf ATL 2016
Josh Patterson, Advisor, Skymind – Deep learning for Industry at MLconf ATL 2016Josh Patterson, Advisor, Skymind – Deep learning for Industry at MLconf ATL 2016
Josh Patterson, Advisor, Skymind – Deep learning for Industry at MLconf ATL 2016MLconf
 
Deep Learning: DL4J and DataVec
Deep Learning: DL4J and DataVecDeep Learning: DL4J and DataVec
Deep Learning: DL4J and DataVecJosh Patterson
 
Jonathon Wright - Intelligent Performance Cognitive Learning (AIOps)
Jonathon Wright - Intelligent Performance Cognitive Learning (AIOps)Jonathon Wright - Intelligent Performance Cognitive Learning (AIOps)
Jonathon Wright - Intelligent Performance Cognitive Learning (AIOps)Neotys_Partner
 
Deep Learning Workflows: Training and Inference
Deep Learning Workflows: Training and InferenceDeep Learning Workflows: Training and Inference
Deep Learning Workflows: Training and InferenceNVIDIA
 
Dsdt meetup 2017 11-21
Dsdt meetup 2017 11-21Dsdt meetup 2017 11-21
Dsdt meetup 2017 11-21JDA Labs MTL
 
DSDT Meetup Nov 2017
DSDT Meetup Nov 2017DSDT Meetup Nov 2017
DSDT Meetup Nov 2017DSDT_MTL
 
Building Google Cloud ML Engine From Scratch on AWS with PipelineAI - ODSC Lo...
Building Google Cloud ML Engine From Scratch on AWS with PipelineAI - ODSC Lo...Building Google Cloud ML Engine From Scratch on AWS with PipelineAI - ODSC Lo...
Building Google Cloud ML Engine From Scratch on AWS with PipelineAI - ODSC Lo...Chris Fregly
 
ISS Capstone - Martinez Technology Consulting and Cedar Hills Church Security...
ISS Capstone - Martinez Technology Consulting and Cedar Hills Church Security...ISS Capstone - Martinez Technology Consulting and Cedar Hills Church Security...
ISS Capstone - Martinez Technology Consulting and Cedar Hills Church Security...Robert Conti Jr.
 
Automated ML Workflow for Distributed Big Data Using Analytics Zoo (CVPR2020 ...
Automated ML Workflow for Distributed Big Data Using Analytics Zoo (CVPR2020 ...Automated ML Workflow for Distributed Big Data Using Analytics Zoo (CVPR2020 ...
Automated ML Workflow for Distributed Big Data Using Analytics Zoo (CVPR2020 ...Jason Dai
 
From 0 to hero adf cicd pass mdpug oslo feb 2020
From 0 to hero adf cicd pass mdpug oslo feb 2020From 0 to hero adf cicd pass mdpug oslo feb 2020
From 0 to hero adf cicd pass mdpug oslo feb 2020Halvar Trøyel Nerbø
 
"Scaling ML from 0 to millions of users", Julien Simon, AWS Dev Day Kyiv 2019
"Scaling ML from 0 to millions of users", Julien Simon, AWS Dev Day Kyiv 2019"Scaling ML from 0 to millions of users", Julien Simon, AWS Dev Day Kyiv 2019
"Scaling ML from 0 to millions of users", Julien Simon, AWS Dev Day Kyiv 2019Provectus
 
Advertising Fraud Detection at Scale at T-Mobile
Advertising Fraud Detection at Scale at T-MobileAdvertising Fraud Detection at Scale at T-Mobile
Advertising Fraud Detection at Scale at T-MobileDatabricks
 
Cloud patterns applied
Cloud patterns appliedCloud patterns applied
Cloud patterns appliedLars Fronius
 
Online Security Analytics on Large Scale Video Surveillance System by Yu Cao ...
Online Security Analytics on Large Scale Video Surveillance System by Yu Cao ...Online Security Analytics on Large Scale Video Surveillance System by Yu Cao ...
Online Security Analytics on Large Scale Video Surveillance System by Yu Cao ...Spark Summit
 
Simpler, faster, cheaper Enterprise Apps using only Spring Boot on GCP
Simpler, faster, cheaper Enterprise Apps using only Spring Boot on GCPSimpler, faster, cheaper Enterprise Apps using only Spring Boot on GCP
Simpler, faster, cheaper Enterprise Apps using only Spring Boot on GCPDaniel Zivkovic
 
Possibilities of generative models
Possibilities of generative modelsPossibilities of generative models
Possibilities of generative modelsAlison B. Lowndes
 

Similar to Implementing Vision DL Solution using Open Source Tools (20)

Smart Data Conference: DL4J and DataVec
Smart Data Conference: DL4J and DataVecSmart Data Conference: DL4J and DataVec
Smart Data Conference: DL4J and DataVec
 
Josh Patterson, Advisor, Skymind – Deep learning for Industry at MLconf ATL 2016
Josh Patterson, Advisor, Skymind – Deep learning for Industry at MLconf ATL 2016Josh Patterson, Advisor, Skymind – Deep learning for Industry at MLconf ATL 2016
Josh Patterson, Advisor, Skymind – Deep learning for Industry at MLconf ATL 2016
 
Deep Learning: DL4J and DataVec
Deep Learning: DL4J and DataVecDeep Learning: DL4J and DataVec
Deep Learning: DL4J and DataVec
 
Iac d.damyanov 4.pptx
Iac d.damyanov 4.pptxIac d.damyanov 4.pptx
Iac d.damyanov 4.pptx
 
Jonathon Wright - Intelligent Performance Cognitive Learning (AIOps)
Jonathon Wright - Intelligent Performance Cognitive Learning (AIOps)Jonathon Wright - Intelligent Performance Cognitive Learning (AIOps)
Jonathon Wright - Intelligent Performance Cognitive Learning (AIOps)
 
Deep Learning Workflows: Training and Inference
Deep Learning Workflows: Training and InferenceDeep Learning Workflows: Training and Inference
Deep Learning Workflows: Training and Inference
 
Dsdt meetup 2017 11-21
Dsdt meetup 2017 11-21Dsdt meetup 2017 11-21
Dsdt meetup 2017 11-21
 
DSDT Meetup Nov 2017
DSDT Meetup Nov 2017DSDT Meetup Nov 2017
DSDT Meetup Nov 2017
 
Building Google Cloud ML Engine From Scratch on AWS with PipelineAI - ODSC Lo...
Building Google Cloud ML Engine From Scratch on AWS with PipelineAI - ODSC Lo...Building Google Cloud ML Engine From Scratch on AWS with PipelineAI - ODSC Lo...
Building Google Cloud ML Engine From Scratch on AWS with PipelineAI - ODSC Lo...
 
ISS Capstone - Martinez Technology Consulting and Cedar Hills Church Security...
ISS Capstone - Martinez Technology Consulting and Cedar Hills Church Security...ISS Capstone - Martinez Technology Consulting and Cedar Hills Church Security...
ISS Capstone - Martinez Technology Consulting and Cedar Hills Church Security...
 
Automated ML Workflow for Distributed Big Data Using Analytics Zoo (CVPR2020 ...
Automated ML Workflow for Distributed Big Data Using Analytics Zoo (CVPR2020 ...Automated ML Workflow for Distributed Big Data Using Analytics Zoo (CVPR2020 ...
Automated ML Workflow for Distributed Big Data Using Analytics Zoo (CVPR2020 ...
 
From 0 to hero adf cicd pass mdpug oslo feb 2020
From 0 to hero adf cicd pass mdpug oslo feb 2020From 0 to hero adf cicd pass mdpug oslo feb 2020
From 0 to hero adf cicd pass mdpug oslo feb 2020
 
"Scaling ML from 0 to millions of users", Julien Simon, AWS Dev Day Kyiv 2019
"Scaling ML from 0 to millions of users", Julien Simon, AWS Dev Day Kyiv 2019"Scaling ML from 0 to millions of users", Julien Simon, AWS Dev Day Kyiv 2019
"Scaling ML from 0 to millions of users", Julien Simon, AWS Dev Day Kyiv 2019
 
Advertising Fraud Detection at Scale at T-Mobile
Advertising Fraud Detection at Scale at T-MobileAdvertising Fraud Detection at Scale at T-Mobile
Advertising Fraud Detection at Scale at T-Mobile
 
Why Startups Are Still On AWS
Why Startups Are Still On AWSWhy Startups Are Still On AWS
Why Startups Are Still On AWS
 
TIAD : Automating the modern datacenter
TIAD : Automating the modern datacenterTIAD : Automating the modern datacenter
TIAD : Automating the modern datacenter
 
Cloud patterns applied
Cloud patterns appliedCloud patterns applied
Cloud patterns applied
 
Online Security Analytics on Large Scale Video Surveillance System by Yu Cao ...
Online Security Analytics on Large Scale Video Surveillance System by Yu Cao ...Online Security Analytics on Large Scale Video Surveillance System by Yu Cao ...
Online Security Analytics on Large Scale Video Surveillance System by Yu Cao ...
 
Simpler, faster, cheaper Enterprise Apps using only Spring Boot on GCP
Simpler, faster, cheaper Enterprise Apps using only Spring Boot on GCPSimpler, faster, cheaper Enterprise Apps using only Spring Boot on GCP
Simpler, faster, cheaper Enterprise Apps using only Spring Boot on GCP
 
Possibilities of generative models
Possibilities of generative modelsPossibilities of generative models
Possibilities of generative models
 

More from Women in Analytics Conference

WIA 2019 - Enhancing AI in Wearable Devices using Topological Data Analysis
WIA 2019 - Enhancing AI in Wearable Devices using Topological Data AnalysisWIA 2019 - Enhancing AI in Wearable Devices using Topological Data Analysis
WIA 2019 - Enhancing AI in Wearable Devices using Topological Data AnalysisWomen in Analytics Conference
 
WIA 2019 - Using Embeddings to Understand the Variance and Evolution of Data ...
WIA 2019 - Using Embeddings to Understand the Variance and Evolution of Data ...WIA 2019 - Using Embeddings to Understand the Variance and Evolution of Data ...
WIA 2019 - Using Embeddings to Understand the Variance and Evolution of Data ...Women in Analytics Conference
 
WIA 2019 - Ethics in Algorithms Panel (Emily Schlesinger)
WIA 2019 - Ethics in Algorithms Panel (Emily Schlesinger)WIA 2019 - Ethics in Algorithms Panel (Emily Schlesinger)
WIA 2019 - Ethics in Algorithms Panel (Emily Schlesinger)Women in Analytics Conference
 
WIA 2019 - Ethics in Algorithms Panel (Nicole Alexander)
WIA 2019 - Ethics in Algorithms Panel (Nicole Alexander)WIA 2019 - Ethics in Algorithms Panel (Nicole Alexander)
WIA 2019 - Ethics in Algorithms Panel (Nicole Alexander)Women in Analytics Conference
 
2019 WIA - Setting Expectations on Data Science Projects
2019 WIA - Setting Expectations on Data Science Projects2019 WIA - Setting Expectations on Data Science Projects
2019 WIA - Setting Expectations on Data Science ProjectsWomen in Analytics Conference
 
2019 WIA - Analytics Maturity: The Power to be Incredible
2019 WIA - Analytics Maturity: The Power to be Incredible2019 WIA - Analytics Maturity: The Power to be Incredible
2019 WIA - Analytics Maturity: The Power to be IncredibleWomen in Analytics Conference
 
WIA 2019 - Steering Model Selection with Visual Diagnostics
WIA 2019 - Steering Model Selection with Visual DiagnosticsWIA 2019 - Steering Model Selection with Visual Diagnostics
WIA 2019 - Steering Model Selection with Visual DiagnosticsWomen in Analytics Conference
 

More from Women in Analytics Conference (10)

WIA 2019 - Enhancing AI in Wearable Devices using Topological Data Analysis
WIA 2019 - Enhancing AI in Wearable Devices using Topological Data AnalysisWIA 2019 - Enhancing AI in Wearable Devices using Topological Data Analysis
WIA 2019 - Enhancing AI in Wearable Devices using Topological Data Analysis
 
WIA 2019 - Using Embeddings to Understand the Variance and Evolution of Data ...
WIA 2019 - Using Embeddings to Understand the Variance and Evolution of Data ...WIA 2019 - Using Embeddings to Understand the Variance and Evolution of Data ...
WIA 2019 - Using Embeddings to Understand the Variance and Evolution of Data ...
 
WIA 2019 - Ethics in Algorithms Panel (Emily Schlesinger)
WIA 2019 - Ethics in Algorithms Panel (Emily Schlesinger)WIA 2019 - Ethics in Algorithms Panel (Emily Schlesinger)
WIA 2019 - Ethics in Algorithms Panel (Emily Schlesinger)
 
2019 WIA - The Importance of Ethics in Data Science
2019 WIA - The Importance of Ethics in Data Science2019 WIA - The Importance of Ethics in Data Science
2019 WIA - The Importance of Ethics in Data Science
 
WIA 2019 - Ethics in Algorithms Panel (Nicole Alexander)
WIA 2019 - Ethics in Algorithms Panel (Nicole Alexander)WIA 2019 - Ethics in Algorithms Panel (Nicole Alexander)
WIA 2019 - Ethics in Algorithms Panel (Nicole Alexander)
 
2019 WIA - Setting Expectations on Data Science Projects
2019 WIA - Setting Expectations on Data Science Projects2019 WIA - Setting Expectations on Data Science Projects
2019 WIA - Setting Expectations on Data Science Projects
 
2019 WIA - Analytics Maturity: The Power to be Incredible
2019 WIA - Analytics Maturity: The Power to be Incredible2019 WIA - Analytics Maturity: The Power to be Incredible
2019 WIA - Analytics Maturity: The Power to be Incredible
 
2019 WIA - Diversity in Analytics
2019 WIA - Diversity in Analytics2019 WIA - Diversity in Analytics
2019 WIA - Diversity in Analytics
 
WIA 2019 - Steering Model Selection with Visual Diagnostics
WIA 2019 - Steering Model Selection with Visual DiagnosticsWIA 2019 - Steering Model Selection with Visual Diagnostics
WIA 2019 - Steering Model Selection with Visual Diagnostics
 
WIA 2019 - From Academia to Industry
WIA 2019 - From Academia to IndustryWIA 2019 - From Academia to Industry
WIA 2019 - From Academia to Industry
 

Recently uploaded

Mature dropshipping via API with DroFx.pptx
Mature dropshipping via API with DroFx.pptxMature dropshipping via API with DroFx.pptx
Mature dropshipping via API with DroFx.pptxolyaivanovalion
 
B2 Creative Industry Response Evaluation.docx
B2 Creative Industry Response Evaluation.docxB2 Creative Industry Response Evaluation.docx
B2 Creative Industry Response Evaluation.docxStephen266013
 
Introduction-to-Machine-Learning (1).pptx
Introduction-to-Machine-Learning (1).pptxIntroduction-to-Machine-Learning (1).pptx
Introduction-to-Machine-Learning (1).pptxfirstjob4
 
VIP Call Girls in Amravati Aarohi 8250192130 Independent Escort Service Amravati
VIP Call Girls in Amravati Aarohi 8250192130 Independent Escort Service AmravatiVIP Call Girls in Amravati Aarohi 8250192130 Independent Escort Service Amravati
VIP Call Girls in Amravati Aarohi 8250192130 Independent Escort Service AmravatiSuhani Kapoor
 
Week-01-2.ppt BBB human Computer interaction
Week-01-2.ppt BBB human Computer interactionWeek-01-2.ppt BBB human Computer interaction
Week-01-2.ppt BBB human Computer interactionfulawalesam
 
Brighton SEO | April 2024 | Data Storytelling
Brighton SEO | April 2024 | Data StorytellingBrighton SEO | April 2024 | Data Storytelling
Brighton SEO | April 2024 | Data StorytellingNeil Barnes
 
Midocean dropshipping via API with DroFx
Midocean dropshipping via API with DroFxMidocean dropshipping via API with DroFx
Midocean dropshipping via API with DroFxolyaivanovalion
 
EMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM TRACKING WITH GOOGLE ANALYTICS.pptx
EMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM  TRACKING WITH GOOGLE ANALYTICS.pptxEMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM  TRACKING WITH GOOGLE ANALYTICS.pptx
EMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM TRACKING WITH GOOGLE ANALYTICS.pptxthyngster
 
定制英国白金汉大学毕业证(UCB毕业证书) 成绩单原版一比一
定制英国白金汉大学毕业证(UCB毕业证书)																			成绩单原版一比一定制英国白金汉大学毕业证(UCB毕业证书)																			成绩单原版一比一
定制英国白金汉大学毕业证(UCB毕业证书) 成绩单原版一比一ffjhghh
 
Market Analysis in the 5 Largest Economic Countries in Southeast Asia.pdf
Market Analysis in the 5 Largest Economic Countries in Southeast Asia.pdfMarket Analysis in the 5 Largest Economic Countries in Southeast Asia.pdf
Market Analysis in the 5 Largest Economic Countries in Southeast Asia.pdfRachmat Ramadhan H
 
Al Barsha Escorts $#$ O565212860 $#$ Escort Service In Al Barsha
Al Barsha Escorts $#$ O565212860 $#$ Escort Service In Al BarshaAl Barsha Escorts $#$ O565212860 $#$ Escort Service In Al Barsha
Al Barsha Escorts $#$ O565212860 $#$ Escort Service In Al BarshaAroojKhan71
 
FESE Capital Markets Fact Sheet 2024 Q1.pdf
FESE Capital Markets Fact Sheet 2024 Q1.pdfFESE Capital Markets Fact Sheet 2024 Q1.pdf
FESE Capital Markets Fact Sheet 2024 Q1.pdfMarinCaroMartnezBerg
 
Invezz.com - Grow your wealth with trading signals
Invezz.com - Grow your wealth with trading signalsInvezz.com - Grow your wealth with trading signals
Invezz.com - Grow your wealth with trading signalsInvezz1
 
VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130
VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130
VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130Suhani Kapoor
 
Smarteg dropshipping via API with DroFx.pptx
Smarteg dropshipping via API with DroFx.pptxSmarteg dropshipping via API with DroFx.pptx
Smarteg dropshipping via API with DroFx.pptxolyaivanovalion
 
Carero dropshipping via API with DroFx.pptx
Carero dropshipping via API with DroFx.pptxCarero dropshipping via API with DroFx.pptx
Carero dropshipping via API with DroFx.pptxolyaivanovalion
 
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /WhatsappsBeautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsappssapnasaifi408
 
Unveiling Insights: The Role of a Data Analyst
Unveiling Insights: The Role of a Data AnalystUnveiling Insights: The Role of a Data Analyst
Unveiling Insights: The Role of a Data AnalystSamantha Rae Coolbeth
 
dokumen.tips_chapter-4-transient-heat-conduction-mehmet-kanoglu.ppt
dokumen.tips_chapter-4-transient-heat-conduction-mehmet-kanoglu.pptdokumen.tips_chapter-4-transient-heat-conduction-mehmet-kanoglu.ppt
dokumen.tips_chapter-4-transient-heat-conduction-mehmet-kanoglu.pptSonatrach
 

Recently uploaded (20)

Mature dropshipping via API with DroFx.pptx
Mature dropshipping via API with DroFx.pptxMature dropshipping via API with DroFx.pptx
Mature dropshipping via API with DroFx.pptx
 
B2 Creative Industry Response Evaluation.docx
B2 Creative Industry Response Evaluation.docxB2 Creative Industry Response Evaluation.docx
B2 Creative Industry Response Evaluation.docx
 
Introduction-to-Machine-Learning (1).pptx
Introduction-to-Machine-Learning (1).pptxIntroduction-to-Machine-Learning (1).pptx
Introduction-to-Machine-Learning (1).pptx
 
VIP Call Girls in Amravati Aarohi 8250192130 Independent Escort Service Amravati
VIP Call Girls in Amravati Aarohi 8250192130 Independent Escort Service AmravatiVIP Call Girls in Amravati Aarohi 8250192130 Independent Escort Service Amravati
VIP Call Girls in Amravati Aarohi 8250192130 Independent Escort Service Amravati
 
Week-01-2.ppt BBB human Computer interaction
Week-01-2.ppt BBB human Computer interactionWeek-01-2.ppt BBB human Computer interaction
Week-01-2.ppt BBB human Computer interaction
 
Brighton SEO | April 2024 | Data Storytelling
Brighton SEO | April 2024 | Data StorytellingBrighton SEO | April 2024 | Data Storytelling
Brighton SEO | April 2024 | Data Storytelling
 
Midocean dropshipping via API with DroFx
Midocean dropshipping via API with DroFxMidocean dropshipping via API with DroFx
Midocean dropshipping via API with DroFx
 
EMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM TRACKING WITH GOOGLE ANALYTICS.pptx
EMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM  TRACKING WITH GOOGLE ANALYTICS.pptxEMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM  TRACKING WITH GOOGLE ANALYTICS.pptx
EMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM TRACKING WITH GOOGLE ANALYTICS.pptx
 
定制英国白金汉大学毕业证(UCB毕业证书) 成绩单原版一比一
定制英国白金汉大学毕业证(UCB毕业证书)																			成绩单原版一比一定制英国白金汉大学毕业证(UCB毕业证书)																			成绩单原版一比一
定制英国白金汉大学毕业证(UCB毕业证书) 成绩单原版一比一
 
Market Analysis in the 5 Largest Economic Countries in Southeast Asia.pdf
Market Analysis in the 5 Largest Economic Countries in Southeast Asia.pdfMarket Analysis in the 5 Largest Economic Countries in Southeast Asia.pdf
Market Analysis in the 5 Largest Economic Countries in Southeast Asia.pdf
 
Al Barsha Escorts $#$ O565212860 $#$ Escort Service In Al Barsha
Al Barsha Escorts $#$ O565212860 $#$ Escort Service In Al BarshaAl Barsha Escorts $#$ O565212860 $#$ Escort Service In Al Barsha
Al Barsha Escorts $#$ O565212860 $#$ Escort Service In Al Barsha
 
FESE Capital Markets Fact Sheet 2024 Q1.pdf
FESE Capital Markets Fact Sheet 2024 Q1.pdfFESE Capital Markets Fact Sheet 2024 Q1.pdf
FESE Capital Markets Fact Sheet 2024 Q1.pdf
 
Invezz.com - Grow your wealth with trading signals
Invezz.com - Grow your wealth with trading signalsInvezz.com - Grow your wealth with trading signals
Invezz.com - Grow your wealth with trading signals
 
꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...
꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...
꧁❤ Aerocity Call Girls Service Aerocity Delhi ❤꧂ 9999965857 ☎️ Hard And Sexy ...
 
VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130
VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130
VIP Call Girls Service Miyapur Hyderabad Call +91-8250192130
 
Smarteg dropshipping via API with DroFx.pptx
Smarteg dropshipping via API with DroFx.pptxSmarteg dropshipping via API with DroFx.pptx
Smarteg dropshipping via API with DroFx.pptx
 
Carero dropshipping via API with DroFx.pptx
Carero dropshipping via API with DroFx.pptxCarero dropshipping via API with DroFx.pptx
Carero dropshipping via API with DroFx.pptx
 
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /WhatsappsBeautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsapps
 
Unveiling Insights: The Role of a Data Analyst
Unveiling Insights: The Role of a Data AnalystUnveiling Insights: The Role of a Data Analyst
Unveiling Insights: The Role of a Data Analyst
 
dokumen.tips_chapter-4-transient-heat-conduction-mehmet-kanoglu.ppt
dokumen.tips_chapter-4-transient-heat-conduction-mehmet-kanoglu.pptdokumen.tips_chapter-4-transient-heat-conduction-mehmet-kanoglu.ppt
dokumen.tips_chapter-4-transient-heat-conduction-mehmet-kanoglu.ppt
 

Implementing Vision DL Solution using Open Source Tools

  • 1. Unearth the Journey of Implementing Vision based Deep Learning Solution using Open Source Tools Neethu Elizabeth Simon Software Engineer Intel Corporation Phoenix, AZ
  • 2.
  • 3.
  • 5.
  • 7. Video – Eye of IoT
  • 9. Dried Chili Peppers Cinnamon Sticks Solution Data Set
  • 10. Implementation • Build Model • Improve Models to meet performance & accuracy objectives • Deploy Model • Make Predictions by applying the Trained Model on new Data • Label Data• Data Acquisition & Organization • Preprocessing Collect Annotation TrainingInference
  • 11. Implementation – Data Collection Collect Annotation TrainingInference  Data Acquisition & Organization • Web Crawling/Scraping - Download pictures & videos Python Packages Request, Parcel, Beautiful Soup Frameworks/Tools Scrapy, Pyspider, Mechanical Soup
  • 12. Implementation – Data Collection Collect Annotation TrainingInference  Data Acquisition & Organization • Web Crawling/Scraping - Download pictures & videos • Live Video Recording/Image Capture from actual retail store Python Packages Request, Parcel, Beautiful Soup Frameworks/Tools Scrapy, Pyspider, Mechanical Soup
  • 13. Implementation – Data Collection Collect Annotation TrainingInference  Pre-processing • Data Cleanup • Remove Bad/Unrelated Data
  • 14. Implementation – Data Annotation Collect Annotation TrainingInference  Label Data • Establish ground truth
  • 15. Implementation – Data Annotation Collect Annotation TrainingInference  Open Source Tools • Bbox - https://github.com/puzzledqs/BBox-Label-Tool • LabelImg - https://github.com/tzutalin/labelImg • VGG Image Annotator (VIA) http://www.robots.ox.ac.uk/~vgg/software/via/ • Computer Vision Annotation Tool (CVAT) https://github.com/opencv/cvat  Paid Tools – Labelbox, Dataloop, RectLabel
  • 17. Implementation – Training Collect Annotation TrainingInference dried_chilli_pepper cinnamon_stick Lots of Labeled Data !!! Model Weights Forward Backward
  • 18. Implementation – Training Collect Annotation TrainingInference Data • Type/Amount – Visual, Sensors • Labeled Datasets – PASCAL VOC, COCO, KITTI, ImageNet or ILSVRC, SUN • Custom Data
  • 19. Implementation – Training Collect Annotation TrainingInference Compute Platform • Complexity Determined by Data • Cloud vs Edge • Intel® Xeon vs Core vs Atom Data • Type/Amount – Visual, Sensors • Labeled Datasets – PASCAL VOC, COCO, KITTI, ImageNet or ILSVRC, SUN • Custom Data Intel Core + GPU
  • 20. Implementation – Training Collect Annotation TrainingInference Object Detection Algorithm • YOLO – You Only Look Once • SSD, R-CNN, Fast R-CNN, Faster RCNN, R-FCN Compute Platform • Complexity Determined by Data • Cloud vs Edge • Intel® Xeon vs Core vs Atom Data • Type/Amount – Visual, Sensors • Labeled Datasets – PASCAL VOC, COCO, KITTI, ImageNet or ILSVRC, SUN • Custom Data Intel Core + GPU YOLO
  • 21. Implementation – Training Collect Annotation TrainingInference Deep Learning Framework • DarkNet - C & CUDA - https://pjreddie.com/darknet/ • DarkFlow - Tensorflow implementation of DarkNet https://github.com/thtrieu/darkflow • Others – PyTorch, Caffe Object Detection Algorithm • YOLO – You Only Look Once • SSD, R-CNN, Fast R-CNN, Faster RCNN, R-FCN Compute Platform • Complexity Determined by Data • Cloud vs Edge • Intel® Xeon vs Core vs Atom Data • Type/Amount – Visual, Sensors • Labeled Datasets – PASCAL VOC, COCO, KITTI, ImageNet or ILSVRC, SUN • Custom Data YOLO Intel Core + GPU DarkFlow
  • 22. Implementation – Training Collect Annotation TrainingInference Deep Learning Framework • DarkNet - C & CUDA - https://pjreddie.com/darknet/ • DarkFlow - Tensorflow implementation of DarkNet https://github.com/thtrieu/darkflow • Others – PyTorch, Caffe Object Detection Algorithm • YOLO – You Only Look Once • SSD, R-CNN, Fast R-CNN, Faster RCNN, R-FCN Compute Platform • Complexity Determined by Data • Cloud vs Edge • Intel® Xeon vs Core vs Atom Data • Type/Amount – Visual, Sensors • Labeled Datasets – PASCAL VOC, COCO, KITTI, ImageNet or ILSVRC, SUN • Custom Data YOLO Intel Core + GPU DarkFlow Images - 1000 Images Training Time - 10 hrs Model Size – 800 MB Average loss – ~0.9
  • 23. Implementation – Training Collect Annotation TrainingInferencedried_chilli_pepper cinnamon_stick Lots of Labeled Data !!! Model Weights Forward Backward python flow --train --model cfg/yolov2-2c.cfg --load bin/yolov2.weights --annotation all_Spice_Annotations/ --dataset all_Spice_Images/ DarkFlow
  • 24. Implementation – Inference Collect Annotation TrainingInference Model Weights Forward dried_chilli_pepper Edge Device – Laptop/NUC/Mobile ???? Intel® OpenVINO Toolkit https://software.intel.com/en- us/openvino-toolkit Open Source Tool for Inference Support Different Models (Caffe/Tensorflow/MXNet/ONNX/Kaldi) Model Optimization Speed Deploymentpython flow --imgdir all_test_images/ --model cfg/yolov2-2c.cfg --load new_model_weights DarkFlow
  • 25. Results [{"label": "cinnamon_stick", "confidence": 0.87, "topleft": {"x": 42, "y": 22}, "bottomright": {"x": 208, "y": 113}}] [{"label": "cinnamon_stick", "confidence": 0.06, "topleft": {"x": 93, "y": 216}, "bottomright": {"x": 384, "y": 402}}]
  • 26. Results [{"label": "dried_chilli_pepper", "confidence": 0.95, "topleft": {"x": 0, "y": 25}, "bottomright": {"x": 761, "y": 352}}] [{"label": "dried_chilli_pepper", "confidence": 0.02, "topleft": {"x": 357, "y": 93}, "bottomright": {"x": 513, "y": 446}}, {"label": "cinnamon_stick", "confidence": 0.03, "topleft": {"x": 21, "y": 6}, "bottomright": {"x": 568, "y": 472}}]
  • 27.
  • 28. Implementation Challenges • More is Better • Object Orientation, angle, lighting • Remove Bad/Unrelated Data • Require SME knowledge to define bad data & remove unrelated data Collect Annotation TrainingInference
  • 29. [{"label": "cinnamon_stick", "confidence": 0.87, "topleft": {"x": 42, "y": 22}, "bottomright": {"x": 208, "y": 113}}]
  • 30. Implementation Challenges • Manual vs Automation • Time consuming • Expensive • Difficult to Scale • More is Better • Object Orientation, angle, lighting • Remove Bad/Unrelated Data • Require SME knowledge to define bad data & remove unrelated data Collect Annotation TrainingInference
  • 31. Implementation Challenges • Iterative Learning • Varied Data parameters & tuning • Hyperparamter optimization & data overfitting • High Compute • Manual vs Automation • Time consuming • Expensive • Difficult to Scale • More is Better • Object Orientation, angle, lighting • Remove Bad/Unrelated Data • Require SME knowledge to define bad data & remove unrelated data Collect Annotation TrainingInference
  • 32. Implementation Challenges • Iterative Learning • Varied Data parameters & tuning • Hyperparamter optimization & data overfitting • High Compute • Constantly Update Models • Complexity in Hardware design – combination of CPU/GPU/Accelerators • Data Privacy & Security • Manual vs Automation • Time consuming • Expensive • Difficult to Scale • More is Better • Object Orientation, angle, lighting • Remove Bad/Unrelated Data • Require SME knowledge to define bad data & remove unrelated data Collect Annotation TrainingInference
  • 33. Neethu Elizabeth Simon Software Engineer Intel Corporation Phoenix, AZ