SlideShare una empresa de Scribd logo
1 de 53
Descargar para leer sin conexión
REAL-TIME EMOTION
RECOGNITION DURING
PRESENTATION
17 October 2019
2
Jogging may be boring
RUNKEEPER FOR PRESENTATIONS
Jogging with Runkeeper
is much more interesting
3
Jogging may be boring
RUNKEEPER FOR PRESENTATIONS
Presentations may be
boring
4
Jogging may be boring
RUNKEEPER FOR PRESENTATIONS
Jogging with Runkeeper
is much more interesting
Presentations may be
boring
How about
presentations with
Runkeeper?5
Jogging may be boring
RUNKEEPER FOR PRESENTATIONS
Jogging with Runkeeper
is much more interesting
• Task description: what are we doing and what do we need
• Face detection: Amazon solution, MTCNN and collecting data
• Emotion recognition: Amazon Rekogniton and mini-Xception
• DeepLens: working pipeline and what’s wrong with it
• Realization: splitting tasks and watchdogs
• Results visualization: Flask and Bokeh
• Data analysis: what one can get from results
• Plans: what else I can do
AGENDA
Machine learning part
Engineering part
Data part
7
HOW CAN WE EVALUATE PRESENTATION?
1. Record a video of presenter during presentation
and evaluate/measure performance later
2. Recognise pose and gesture of presenter during
presentation with machine learning
3. Take information from audience, not from
presenter
8
1. Photos are uploaded in the cloud
2. Photos are stored locally
3. Personal information
WHAT WERE THE MAIN PROBLEMS?
9
1. No photos are uploaded in the cloud
2. No photos are stored locally
3. Only aggregated information
PRIVACY CONCERN
10
AMAZON DEEPLENS
HD camera with GPU and CPU inside
FACE DETECTION ALGORITHM
11
WHAT DO WE NEED?
FACE DETECTION ALGORITHM EMOTION RECOGNITION ALGORITHM
12
WHAT DO WE NEED?
• Preconfigured project in
Amazon library
13
FACE DETECTION
Amazon solution
• Preconfigured project in
Amazon library
• MXNet model
• 300*300 input reshape
14
FACE DETECTION
Amazon solution
• Preconfigured project in
Amazon library
• MXNet model
• 300*300 input reshape
• 4-5 faces maximum
15
FACE DETECTION
Amazon solution
• Preconfigured project in
Amazon library
• MXNet model
• 300*300 input reshape
• 4-5 faces maximum
16
FACE DETECTION
Amazon solution
CAN WE DO SOMETHING?
17
CROWD DETECTION PROBLEM
Resize -> Window -> Resize back
18
CROWD DETECTION PROBLEM
No resize -> Frame -> Face
19
CROWD DETECTION PROBLEM
Resize or not resize?
20
CROWD DETECTION PROBLEM
Why not both?
1. Building “Image pyramid”
2. Proposal Network (P-Net)
3. Refinement Network (R-Net)
4. Output Network (O-Net)
+ Non Maximum Supression
21
MTCNN…
..to the rescue!
https://arxiv.org/ftp/arxiv/papers/1604/1604.02878.pdf
22
MTCNN RESULTS
1 face
Image size Total pixels Process time FPS Image size Total pixels
Process
time
FPS
460x259 119,140 0.118 seconds 8.5 474x224 106,176
0.185
seconds
5.4
561x561 314,721 0.227 seconds 4.5 736x348 256,128
0.290
seconds
3.4
667x1000 667,000 0.456 seconds 2.2 2100x994 2,087,400
1.286
seconds
0.7
1920x1200 2,304,000 1.093 seconds 0.9
4799x3599 17,271,601 8.798 seconds 0.1
10 faces
23
MTCNN RESULTS
3872 x 2592
24
MTCNN RESULTS
3872 x 2592
25
MTCNN RESULTS
1280 x 1040
26
MTCNN RESULTS
1280 x 1040 300 x 300
27
AMAZON RECOGNITION API
Test launch
Crop faces
Recognize emotions
• SAD: 0.01
• HAPPY: 0.34
• CONFUSED: 0.14
• DISGUSTED: 0.05
• CALM: 0.22
• ANGRY: 0.11
• SURPRISED: 0.05
28
AMAZON REKOGNITION
Test launch
Crop faces
Recognize emotions
• SAD: 0.01
• HAPPY: 0.34
• CONFUSED: 0.14
• DISGUSTED: 0.05
• CALM: 0.22
• ANGRY: 0.11
• SURPRISED: 0.05
What can be wrong here?
29
• Privacy reasons
PROBLEMS
30
• Privacy reasons
• Time consumption
PROBLEMS
31
• Privacy reasons
• Time consumption
• Calm bias
PROBLEMS
32
FER+ DATASET
Even better than FER
33
MINI-XCEPTION…
Fast and small
https://arxiv.org/pdf/1710.07557.pdf
Resnet50: ~25 million weights
Inception: ~56 million weights
Xception: ~23 million weights
Mini-Xception: ~60k weights
Munich -> Airport ~ 40 km Length of equator ~40k km
34
MINI-XCEPTION…
Fast and small
https://arxiv.org/pdf/1710.07557.pdf
Resnet50: ~25 million weights
Inception: ~56 million weights
Xception: ~23 million weights
Mini-Xception: ~60k weights
35
DEPTHWISE SEPARABLE CONVOLUTION
https://towardsdatascience.com/review-xception-with-depthwise-separable-convolution-better-than-inception-v3-image-dc967dd42568
Why is it so fast?
Regular convolution:
Input: M*M*C1
Output: (M-K)*(M-K)*C2
Weights: C1*C2*K*K
Separable convolution:
Input: M*M*C1
Output: M*M*C2
Weights: C1*C2*1*1 + C2*K*K
Example:
Input: 100*100*16
Output: 97*97*32
Kernel: 3*3
Regular: 16*32*3*3 = 4608
Separable: 16*32*1*1 + 32*3*3 = 288
36
MINI-XCEPTION
Results
COMPARISON WITH AMAZON REKOGNITION
37
MINI-XCEPTION
Results
COMPARISON WITH AMAZON REKOGNITION
LOSS AND VAL-LOSS
ONE MODEL AT A
TIME
Fix: save the second
model in device
memory
38
CHALLENGES WITH DEEPLENS
Really, why?
ONE MODEL AT A
TIME
Fix: save the second
model in device
memory
Fix: check in a code if
there is a folder ->
create this folder via
ssh
39
CAN’T TURN ON /
TURN OFF
CHALLENGES WITH DEEPLENS
Really, why?
ONE MODEL AT A
TIME
Fix: save the second
model in device
memory
Fix: check in a code if
there is a folder ->
create this folder via
ssh
Fix: improvise, adapt,
overcome
40
CAN’T TURN ON /
TURN OFF
CAN’T STACK
OVERFLOWING
CHALLENGES WITH DEEPLENS
Really, why?
ONE MODEL AT A
TIME
Fix: save the second
model in device
memory
Fix: check in a code if
there is a folder ->
create this folder via
ssh
Fix: improvise, adapt,
overcome
41
CAN’T TURN ON /
TURN OFF
CAN’T STACK
OVERFLOWING
SETUP (INITIAL AND
EVERYTIME WIFI)
Fix: suffer
CHALLENGES WITH DEEPLENS
Really, why?
ONE MODEL AT A
TIME
Fix: save the second
model in device
memory
Fix: check in a code if
there is a folder ->
create this folder via
ssh
Fix: improvise, adapt,
overcome
42
CAN’T TURN ON /
TURN OFF
CAN’T STACK
OVERFLOWING
SETUP (INITIAL AND
EVERYTIME WIFI)
Fix: suffer
PYTHON3 IS NOT
SUPPORTED
Fix: micro-architecture
CHALLENGES WITH DEEPLENS
Really, why?
ALGORITHM 1
Send result to step 2 Send result to step 3 Send results to output
Send confirmation to
step 1
Do something
43
MICRO-ARCHITECTURE APPROACH
ALGORITHM 2
Do something
ALGORITHM 3
Do something
44
WATCHDOG
Inotify and threads
New file created
Function
• Four independent modules
• Unified watchdog algorithm
• Easy to modify
45
REALIZATION
MTCNN
Mini Xception
Photo
Visualization
• Running on device
• Can be reach by Wi-Fi from laptops and mobile devices
• Update graphics every 3 seconds on-the-go via Ajax
46
ONLINE VISUALIZATION
Flask and Bokeh
GIF
47
AFTERWARD DATA ANALYSIS
Emotions
48
AFTERWARD DATA ANALYSIS
Emotions
49
AFTERWARD DATA ANALYSIS
People interested
• Headpose detection
• Runnable on a mobile device
• Psychology research
50
POSSIBLE EXTENSIONS
Data Reply:
Dmitrii Azarnykh
Wadim Pessin
Gavin Perrie
51
THANKS
TUM:
Prof. Dr. Christoph Ungemach
BTW, I AM
LOOKING FOR
А THESIS
Artemii Frolov
arfr@live.ru
https://www.linkedin.com/in/artemii-frolov/
THANK YOU

Más contenido relacionado

Similar a Real time emotion recognition during presentations - Artemii Frolov

The Art Of Performance Tuning - with presenter notes!
The Art Of Performance Tuning - with presenter notes!The Art Of Performance Tuning - with presenter notes!
The Art Of Performance Tuning - with presenter notes!Jonathan Ross
 
Writing Performant Front-end Code
Writing Performant Front-end CodeWriting Performant Front-end Code
Writing Performant Front-end CodeLyubomir Bozhinov
 
Sensepost assessment automation
Sensepost assessment automationSensepost assessment automation
Sensepost assessment automationSensePost
 
iOS Scroll Performance
iOS Scroll PerformanceiOS Scroll Performance
iOS Scroll PerformanceKyle Sherman
 
COSCUP 2017 FACE OFF
COSCUP 2017 FACE OFFCOSCUP 2017 FACE OFF
COSCUP 2017 FACE OFFPRADA Hsiung
 
STAMP, or Test Amplification to DevTestOps service, OW2con'18, June 7-8, 2018...
STAMP, or Test Amplification to DevTestOps service, OW2con'18, June 7-8, 2018...STAMP, or Test Amplification to DevTestOps service, OW2con'18, June 7-8, 2018...
STAMP, or Test Amplification to DevTestOps service, OW2con'18, June 7-8, 2018...OW2
 
Has responsive had it's day? : Amplience Customer Day 2014
Has responsive had it's day? : Amplience Customer Day 2014Has responsive had it's day? : Amplience Customer Day 2014
Has responsive had it's day? : Amplience Customer Day 2014Ben Seymour
 
Advanced #4 GPU & Animations
Advanced #4   GPU & AnimationsAdvanced #4   GPU & Animations
Advanced #4 GPU & AnimationsVitali Pekelis
 
Getting Intimate with Images on Android with James Halpern
Getting Intimate with Images on Android with James HalpernGetting Intimate with Images on Android with James Halpern
Getting Intimate with Images on Android with James HalpernFITC
 
Scaling MongoDB for real time analytics
Scaling MongoDB for real time analyticsScaling MongoDB for real time analytics
Scaling MongoDB for real time analyticsDavid Tollmyr
 
Country domination - Causing chaos and wrecking havoc
Country domination - Causing chaos and wrecking havocCountry domination - Causing chaos and wrecking havoc
Country domination - Causing chaos and wrecking havocTiago Henriques
 
AWS DeepLens Workshop: Building Computer Vision Applications - BDA201 - Atlan...
AWS DeepLens Workshop: Building Computer Vision Applications - BDA201 - Atlan...AWS DeepLens Workshop: Building Computer Vision Applications - BDA201 - Atlan...
AWS DeepLens Workshop: Building Computer Vision Applications - BDA201 - Atlan...Amazon Web Services
 
Kostiantyn Yelisavenko "Mastering Macro Benchmarking in .NET"
Kostiantyn Yelisavenko "Mastering Macro Benchmarking in .NET"Kostiantyn Yelisavenko "Mastering Macro Benchmarking in .NET"
Kostiantyn Yelisavenko "Mastering Macro Benchmarking in .NET"LogeekNightUkraine
 
Os Leventhal
Os LeventhalOs Leventhal
Os Leventhaloscon2007
 
Us 16-subverting apple-graphics_practical_approaches_to_remotely_gaining_root...
Us 16-subverting apple-graphics_practical_approaches_to_remotely_gaining_root...Us 16-subverting apple-graphics_practical_approaches_to_remotely_gaining_root...
Us 16-subverting apple-graphics_practical_approaches_to_remotely_gaining_root...Liang Chen
 
深度學習在AOI的應用
深度學習在AOI的應用深度學習在AOI的應用
深度學習在AOI的應用CHENHuiMei
 
Basic Machinima Making, Part 3
Basic Machinima Making, Part 3Basic Machinima Making, Part 3
Basic Machinima Making, Part 3Barbara McQueen
 
Winning the Erlang Edit•Build•Test Cycle
Winning the Erlang Edit•Build•Test CycleWinning the Erlang Edit•Build•Test Cycle
Winning the Erlang Edit•Build•Test CycleRusty Klophaus
 
Browser Automation
Browser AutomationBrowser Automation
Browser AutomationEhren Foss
 
Using firefox internet browser
Using firefox internet browserUsing firefox internet browser
Using firefox internet browsertaylorr2
 

Similar a Real time emotion recognition during presentations - Artemii Frolov (20)

The Art Of Performance Tuning - with presenter notes!
The Art Of Performance Tuning - with presenter notes!The Art Of Performance Tuning - with presenter notes!
The Art Of Performance Tuning - with presenter notes!
 
Writing Performant Front-end Code
Writing Performant Front-end CodeWriting Performant Front-end Code
Writing Performant Front-end Code
 
Sensepost assessment automation
Sensepost assessment automationSensepost assessment automation
Sensepost assessment automation
 
iOS Scroll Performance
iOS Scroll PerformanceiOS Scroll Performance
iOS Scroll Performance
 
COSCUP 2017 FACE OFF
COSCUP 2017 FACE OFFCOSCUP 2017 FACE OFF
COSCUP 2017 FACE OFF
 
STAMP, or Test Amplification to DevTestOps service, OW2con'18, June 7-8, 2018...
STAMP, or Test Amplification to DevTestOps service, OW2con'18, June 7-8, 2018...STAMP, or Test Amplification to DevTestOps service, OW2con'18, June 7-8, 2018...
STAMP, or Test Amplification to DevTestOps service, OW2con'18, June 7-8, 2018...
 
Has responsive had it's day? : Amplience Customer Day 2014
Has responsive had it's day? : Amplience Customer Day 2014Has responsive had it's day? : Amplience Customer Day 2014
Has responsive had it's day? : Amplience Customer Day 2014
 
Advanced #4 GPU & Animations
Advanced #4   GPU & AnimationsAdvanced #4   GPU & Animations
Advanced #4 GPU & Animations
 
Getting Intimate with Images on Android with James Halpern
Getting Intimate with Images on Android with James HalpernGetting Intimate with Images on Android with James Halpern
Getting Intimate with Images on Android with James Halpern
 
Scaling MongoDB for real time analytics
Scaling MongoDB for real time analyticsScaling MongoDB for real time analytics
Scaling MongoDB for real time analytics
 
Country domination - Causing chaos and wrecking havoc
Country domination - Causing chaos and wrecking havocCountry domination - Causing chaos and wrecking havoc
Country domination - Causing chaos and wrecking havoc
 
AWS DeepLens Workshop: Building Computer Vision Applications - BDA201 - Atlan...
AWS DeepLens Workshop: Building Computer Vision Applications - BDA201 - Atlan...AWS DeepLens Workshop: Building Computer Vision Applications - BDA201 - Atlan...
AWS DeepLens Workshop: Building Computer Vision Applications - BDA201 - Atlan...
 
Kostiantyn Yelisavenko "Mastering Macro Benchmarking in .NET"
Kostiantyn Yelisavenko "Mastering Macro Benchmarking in .NET"Kostiantyn Yelisavenko "Mastering Macro Benchmarking in .NET"
Kostiantyn Yelisavenko "Mastering Macro Benchmarking in .NET"
 
Os Leventhal
Os LeventhalOs Leventhal
Os Leventhal
 
Us 16-subverting apple-graphics_practical_approaches_to_remotely_gaining_root...
Us 16-subverting apple-graphics_practical_approaches_to_remotely_gaining_root...Us 16-subverting apple-graphics_practical_approaches_to_remotely_gaining_root...
Us 16-subverting apple-graphics_practical_approaches_to_remotely_gaining_root...
 
深度學習在AOI的應用
深度學習在AOI的應用深度學習在AOI的應用
深度學習在AOI的應用
 
Basic Machinima Making, Part 3
Basic Machinima Making, Part 3Basic Machinima Making, Part 3
Basic Machinima Making, Part 3
 
Winning the Erlang Edit•Build•Test Cycle
Winning the Erlang Edit•Build•Test CycleWinning the Erlang Edit•Build•Test Cycle
Winning the Erlang Edit•Build•Test Cycle
 
Browser Automation
Browser AutomationBrowser Automation
Browser Automation
 
Using firefox internet browser
Using firefox internet browserUsing firefox internet browser
Using firefox internet browser
 

Último

Presentation of project of business person who are success
Presentation of project of business person who are successPresentation of project of business person who are success
Presentation of project of business person who are successPratikSingh115843
 
Decoding Movie Sentiments: Analyzing Reviews with Data Analysis model
Decoding Movie Sentiments: Analyzing Reviews with Data Analysis modelDecoding Movie Sentiments: Analyzing Reviews with Data Analysis model
Decoding Movie Sentiments: Analyzing Reviews with Data Analysis modelBoston Institute of Analytics
 
DATA ANALYSIS using various data sets like shoping data set etc
DATA ANALYSIS using various data sets like shoping data set etcDATA ANALYSIS using various data sets like shoping data set etc
DATA ANALYSIS using various data sets like shoping data set etclalithasri22
 
Role of Consumer Insights in business transformation
Role of Consumer Insights in business transformationRole of Consumer Insights in business transformation
Role of Consumer Insights in business transformationAnnie Melnic
 
6 Tips for Interpretable Topic Models _ by Nicha Ruchirawat _ Towards Data Sc...
6 Tips for Interpretable Topic Models _ by Nicha Ruchirawat _ Towards Data Sc...6 Tips for Interpretable Topic Models _ by Nicha Ruchirawat _ Towards Data Sc...
6 Tips for Interpretable Topic Models _ by Nicha Ruchirawat _ Towards Data Sc...Dr Arash Najmaei ( Phd., MBA, BSc)
 
Decision Making Under Uncertainty - Is It Better Off Joining a Partnership or...
Decision Making Under Uncertainty - Is It Better Off Joining a Partnership or...Decision Making Under Uncertainty - Is It Better Off Joining a Partnership or...
Decision Making Under Uncertainty - Is It Better Off Joining a Partnership or...ThinkInnovation
 
Data Analysis Project Presentation: Unveiling Your Ideal Customer, Bank Custo...
Data Analysis Project Presentation: Unveiling Your Ideal Customer, Bank Custo...Data Analysis Project Presentation: Unveiling Your Ideal Customer, Bank Custo...
Data Analysis Project Presentation: Unveiling Your Ideal Customer, Bank Custo...Boston Institute of Analytics
 
IBEF report on the Insurance market in India
IBEF report on the Insurance market in IndiaIBEF report on the Insurance market in India
IBEF report on the Insurance market in IndiaManalVerma4
 
why-transparency-and-traceability-are-essential-for-sustainable-supply-chains...
why-transparency-and-traceability-are-essential-for-sustainable-supply-chains...why-transparency-and-traceability-are-essential-for-sustainable-supply-chains...
why-transparency-and-traceability-are-essential-for-sustainable-supply-chains...Jack Cole
 
Digital Indonesia Report 2024 by We Are Social .pdf
Digital Indonesia Report 2024 by We Are Social .pdfDigital Indonesia Report 2024 by We Are Social .pdf
Digital Indonesia Report 2024 by We Are Social .pdfNicoChristianSunaryo
 
Statistics For Management by Richard I. Levin 8ed.pdf
Statistics For Management by Richard I. Levin 8ed.pdfStatistics For Management by Richard I. Levin 8ed.pdf
Statistics For Management by Richard I. Levin 8ed.pdfnikeshsingh56
 
Bank Loan Approval Analysis: A Comprehensive Data Analysis Project
Bank Loan Approval Analysis: A Comprehensive Data Analysis ProjectBank Loan Approval Analysis: A Comprehensive Data Analysis Project
Bank Loan Approval Analysis: A Comprehensive Data Analysis ProjectBoston Institute of Analytics
 
Predictive Analysis - Using Insight-informed Data to Plan Inventory in Next 6...
Predictive Analysis - Using Insight-informed Data to Plan Inventory in Next 6...Predictive Analysis - Using Insight-informed Data to Plan Inventory in Next 6...
Predictive Analysis - Using Insight-informed Data to Plan Inventory in Next 6...ThinkInnovation
 

Último (16)

Presentation of project of business person who are success
Presentation of project of business person who are successPresentation of project of business person who are success
Presentation of project of business person who are success
 
Decoding Movie Sentiments: Analyzing Reviews with Data Analysis model
Decoding Movie Sentiments: Analyzing Reviews with Data Analysis modelDecoding Movie Sentiments: Analyzing Reviews with Data Analysis model
Decoding Movie Sentiments: Analyzing Reviews with Data Analysis model
 
DATA ANALYSIS using various data sets like shoping data set etc
DATA ANALYSIS using various data sets like shoping data set etcDATA ANALYSIS using various data sets like shoping data set etc
DATA ANALYSIS using various data sets like shoping data set etc
 
Role of Consumer Insights in business transformation
Role of Consumer Insights in business transformationRole of Consumer Insights in business transformation
Role of Consumer Insights in business transformation
 
Data Analysis Project: Stroke Prediction
Data Analysis Project: Stroke PredictionData Analysis Project: Stroke Prediction
Data Analysis Project: Stroke Prediction
 
6 Tips for Interpretable Topic Models _ by Nicha Ruchirawat _ Towards Data Sc...
6 Tips for Interpretable Topic Models _ by Nicha Ruchirawat _ Towards Data Sc...6 Tips for Interpretable Topic Models _ by Nicha Ruchirawat _ Towards Data Sc...
6 Tips for Interpretable Topic Models _ by Nicha Ruchirawat _ Towards Data Sc...
 
Decision Making Under Uncertainty - Is It Better Off Joining a Partnership or...
Decision Making Under Uncertainty - Is It Better Off Joining a Partnership or...Decision Making Under Uncertainty - Is It Better Off Joining a Partnership or...
Decision Making Under Uncertainty - Is It Better Off Joining a Partnership or...
 
Data Analysis Project Presentation: Unveiling Your Ideal Customer, Bank Custo...
Data Analysis Project Presentation: Unveiling Your Ideal Customer, Bank Custo...Data Analysis Project Presentation: Unveiling Your Ideal Customer, Bank Custo...
Data Analysis Project Presentation: Unveiling Your Ideal Customer, Bank Custo...
 
IBEF report on the Insurance market in India
IBEF report on the Insurance market in IndiaIBEF report on the Insurance market in India
IBEF report on the Insurance market in India
 
2023 Survey Shows Dip in High School E-Cigarette Use
2023 Survey Shows Dip in High School E-Cigarette Use2023 Survey Shows Dip in High School E-Cigarette Use
2023 Survey Shows Dip in High School E-Cigarette Use
 
why-transparency-and-traceability-are-essential-for-sustainable-supply-chains...
why-transparency-and-traceability-are-essential-for-sustainable-supply-chains...why-transparency-and-traceability-are-essential-for-sustainable-supply-chains...
why-transparency-and-traceability-are-essential-for-sustainable-supply-chains...
 
Digital Indonesia Report 2024 by We Are Social .pdf
Digital Indonesia Report 2024 by We Are Social .pdfDigital Indonesia Report 2024 by We Are Social .pdf
Digital Indonesia Report 2024 by We Are Social .pdf
 
Statistics For Management by Richard I. Levin 8ed.pdf
Statistics For Management by Richard I. Levin 8ed.pdfStatistics For Management by Richard I. Levin 8ed.pdf
Statistics For Management by Richard I. Levin 8ed.pdf
 
Insurance Churn Prediction Data Analysis Project
Insurance Churn Prediction Data Analysis ProjectInsurance Churn Prediction Data Analysis Project
Insurance Churn Prediction Data Analysis Project
 
Bank Loan Approval Analysis: A Comprehensive Data Analysis Project
Bank Loan Approval Analysis: A Comprehensive Data Analysis ProjectBank Loan Approval Analysis: A Comprehensive Data Analysis Project
Bank Loan Approval Analysis: A Comprehensive Data Analysis Project
 
Predictive Analysis - Using Insight-informed Data to Plan Inventory in Next 6...
Predictive Analysis - Using Insight-informed Data to Plan Inventory in Next 6...Predictive Analysis - Using Insight-informed Data to Plan Inventory in Next 6...
Predictive Analysis - Using Insight-informed Data to Plan Inventory in Next 6...
 

Real time emotion recognition during presentations - Artemii Frolov