SlideShare a Scribd company logo
Information Technology

Robust Background Subtraction Based on
Perceptual Mixture-of-Gaussians with
Dynamic Adaptation Speed
Mahfuzul Haque and Manzur Murshed
Agenda
 Background Subtraction
 Statistical Background Subtraction
 Perception Inspired Background Subtraction

 Dynamic Adaptation Speed
 Experiments

 Summary
 Q&A

Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

2
Background Subtraction
Input

Output

Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

3
Background Subtraction: Challenges
Basic Background Subtraction (e.g., BBS)
-

Current frame

Challenges

=

Background

Foreground Blob

Dynamic Background Subtraction(e.g., MOG)
Background






Illumination variation
Local background motion
Camera displacement
Shadow and reflection

Model
Current frame

Foreground Blob

Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

4
Statistical Background Subtraction
ω1

ω2

ω3

σ12

σ22

σ32

µ1

µ2

µ3

road

car

shadow

65%

20%

Statistical Approaches

x

x

P(x)

Our Hypothesis (Perception Inspired)

x

BBS: x = c
MOG: x = c1σ

μ

15%

x

x = c2b

P(x)

Te b
Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

5
Perception Inspired Background Subtraction
x = c2 b

Current
Frame

Detection with
Low x

Detection with
High x

x

x
P(x)

b

Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

6
Weber’s Law
How human visual system perceives noticeable intensity
deviation from the background?

Ernst Weber, an experimental psychologist in the 19th
century, observed that the just-noticeable increment ΔI

is linearly proportional to the background intensity I.

ΔI = c2I
Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

7
Weber’s Law
Ernst Weber, an experimental psychologist in the
19th century, observed that the just-noticeable
increment ΔI is linearly proportional to the
background intensity I.

?

x

x

ΔI = c2I
x

x = c2 b

P(x)

b

Te

b

Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

8
Perceptual tolerance of HVS
What is the perceptual tolerance level in distinguishing
distorted intensity measures?

p dB

Method 1

Reference

q dB

Method 2

Image
Distorted
Images

|p – q| < 0.5 dB
Not perceivable
by human visual
system

Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

9
Our Problem: c2 = ?
x = c2 b
x

x
P(x)

Weber’s Law

x = c2b
Perceptual Threshold, TP (0.5 dB)

 255
20 log10 
 bx




  20 log  255
10  b  x





 1
 2TP


b

Te

Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

10
Linear Relationship

x

b
Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

11
Rod and Cone
 Rods and Cones are two different types of
photoreceptor cells in the retina of human eye
 Rods

– Operate in less intense light
– Responsible for scotopic vision (night vision)
 Cones
– Operate in relatively bright light
– Responsible for photopic (color vision)
Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

12
Error Sensitivity in Darker Background

Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

13
Piece-wise Liner Relationship

Scotopic Vision (R)

Photopic Vision (C)
Te
Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

14
Dynamic Adaptation Speed
•Sleeping person problem
•Walking person problem

Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

15
Dynamic Adaptation Speed

Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

16
Dynamic Adaptation Speed

Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

17
Dynamic Adaptation Speed

Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

18
Experiments
Test Sequences
 Total 50 test sequences from 8 different sources
 Scenario distribution






Indoor
Outdoor
Multimodal
Shadow and Reflection
Low background-foreground contrast

False Classification

Evaluation
 Qualitative and quantitative comparison:
 MOG (S&G) (TPAMI, 2000)

False Positive (FP)
False Negative (FN)

 MOG (Lee) (TPAMI, 2005)

 ViBe (TIP, 2011)
Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

19
Test Sequences

PETS (9) Wallflower (7) UCF (7)

IBM (11)

CAVIAR (7)

Te

VSSN06 (7)

Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

Other (2)

December 30, 2013

20
Experiments

Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

21
Experiments

Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

22
Experiments

Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

23
First
Frame

Test
Frame

Ground
Truth

MOG
(S&G)

MOG
(Lee)

ViBe

Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

Proposed

December 30, 2013

24
Summary
 Realistic background value prediction: high model agility
and superior detection quality at fast learning rate.
 No context related information: high stability across
changing scenarios.
 Perception based detection threshold: superior detection
quality in terms of shadow, noise, and reflection.
 Perceptual model similarity: optimal number of models
throughout the system life cycle.
 Parameter-less background subtraction: ideal for realtime video analytics.
Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

25
Q&A

Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed

December 30, 2013

26

More Related Content

Similar to Talk 2012-icmew-perception

Gtti 10032021
Gtti 10032021Gtti 10032021
Gtti 10032021
Valentin De Bortoli
 
Lucas Theis - Compressing Images with Neural Networks - Creative AI meetup
Lucas Theis - Compressing Images with Neural Networks - Creative AI meetupLucas Theis - Compressing Images with Neural Networks - Creative AI meetup
Lucas Theis - Compressing Images with Neural Networks - Creative AI meetup
Luba Elliott
 
Consensual gene co-expression network inference with multiple samples
Consensual gene co-expression network inference with multiple samplesConsensual gene co-expression network inference with multiple samples
Consensual gene co-expression network inference with multiple samples
tuxette
 
Regression and Classification: An Artificial Neural Network Approach
Regression and Classification: An Artificial Neural Network ApproachRegression and Classification: An Artificial Neural Network Approach
Regression and Classification: An Artificial Neural Network Approach
Khulna University
 
20130722
2013072220130722
20130722
Hayaru SHOUNO
 
Multiple Person Tracking with Shadow Removal Using Adaptive Gaussian Mixture ...
Multiple Person Tracking with Shadow Removal Using Adaptive Gaussian Mixture ...Multiple Person Tracking with Shadow Removal Using Adaptive Gaussian Mixture ...
Multiple Person Tracking with Shadow Removal Using Adaptive Gaussian Mixture ...
IJSRD
 
Human detection in hours of
Human detection in hours ofHuman detection in hours of
Human detection in hours of
ijistjournal
 
Senseapp13 keynote
Senseapp13 keynoteSenseapp13 keynote
Senseapp13 keynote
Raja Jurdak
 

Similar to Talk 2012-icmew-perception (8)

Gtti 10032021
Gtti 10032021Gtti 10032021
Gtti 10032021
 
Lucas Theis - Compressing Images with Neural Networks - Creative AI meetup
Lucas Theis - Compressing Images with Neural Networks - Creative AI meetupLucas Theis - Compressing Images with Neural Networks - Creative AI meetup
Lucas Theis - Compressing Images with Neural Networks - Creative AI meetup
 
Consensual gene co-expression network inference with multiple samples
Consensual gene co-expression network inference with multiple samplesConsensual gene co-expression network inference with multiple samples
Consensual gene co-expression network inference with multiple samples
 
Regression and Classification: An Artificial Neural Network Approach
Regression and Classification: An Artificial Neural Network ApproachRegression and Classification: An Artificial Neural Network Approach
Regression and Classification: An Artificial Neural Network Approach
 
20130722
2013072220130722
20130722
 
Multiple Person Tracking with Shadow Removal Using Adaptive Gaussian Mixture ...
Multiple Person Tracking with Shadow Removal Using Adaptive Gaussian Mixture ...Multiple Person Tracking with Shadow Removal Using Adaptive Gaussian Mixture ...
Multiple Person Tracking with Shadow Removal Using Adaptive Gaussian Mixture ...
 
Human detection in hours of
Human detection in hours ofHuman detection in hours of
Human detection in hours of
 
Senseapp13 keynote
Senseapp13 keynoteSenseapp13 keynote
Senseapp13 keynote
 

More from Mahfuzul Haque

Dependency inversion using ports and adapters
Dependency inversion using ports and adaptersDependency inversion using ports and adapters
Dependency inversion using ports and adapters
Mahfuzul Haque
 
Resilient machine learning systems for health analytics
Resilient machine learning systems for health analyticsResilient machine learning systems for health analytics
Resilient machine learning systems for health analytics
Mahfuzul Haque
 
Talk 2012-icmew-event
Talk 2012-icmew-eventTalk 2012-icmew-event
Talk 2012-icmew-event
Mahfuzul Haque
 
Talk 2011-buet-perception-event
Talk 2011-buet-perception-eventTalk 2011-buet-perception-event
Talk 2011-buet-perception-event
Mahfuzul Haque
 
Talk 2010-monash-seminar-panic-driven-event-detection
Talk 2010-monash-seminar-panic-driven-event-detectionTalk 2010-monash-seminar-panic-driven-event-detection
Talk 2010-monash-seminar-panic-driven-event-detection
Mahfuzul Haque
 
Talk 2009-monash-seminar-perception
Talk 2009-monash-seminar-perceptionTalk 2009-monash-seminar-perception
Talk 2009-monash-seminar-perception
Mahfuzul Haque
 
Talk 2009-monash-seminar-intelligent-video-surveillance
Talk 2009-monash-seminar-intelligent-video-surveillanceTalk 2009-monash-seminar-intelligent-video-surveillance
Talk 2009-monash-seminar-intelligent-video-surveillance
Mahfuzul Haque
 
Talk 2009-monash-open-day-surveillance
Talk 2009-monash-open-day-surveillanceTalk 2009-monash-open-day-surveillance
Talk 2009-monash-open-day-surveillance
Mahfuzul Haque
 
Talk 2007-monash-seminar-behavior-recognition-framework
Talk 2007-monash-seminar-behavior-recognition-frameworkTalk 2007-monash-seminar-behavior-recognition-framework
Talk 2007-monash-seminar-behavior-recognition-framework
Mahfuzul Haque
 
Kb hmm
Kb hmmKb hmm
Kb gait-recognition
Kb gait-recognitionKb gait-recognition
Kb gait-recognition
Mahfuzul Haque
 
Kb behaviour-recognition
Kb behaviour-recognitionKb behaviour-recognition
Kb behaviour-recognition
Mahfuzul Haque
 
Poster: Monash Research Month 2009
Poster: Monash Research Month 2009Poster: Monash Research Month 2009
Poster: Monash Research Month 2009
Mahfuzul Haque
 
Poster: Monash Research Month 2008
Poster: Monash Research Month 2008Poster: Monash Research Month 2008
Poster: Monash Research Month 2008
Mahfuzul Haque
 
Poster: Monash Research Month 2007
Poster: Monash Research Month 2007Poster: Monash Research Month 2007
Poster: Monash Research Month 2007
Mahfuzul Haque
 
Poster: ICPR 2008
Poster: ICPR 2008Poster: ICPR 2008
Poster: ICPR 2008
Mahfuzul Haque
 
Poster: ICME 2010
Poster: ICME 2010Poster: ICME 2010
Poster: ICME 2010
Mahfuzul Haque
 
Poster: EII Workshop 2007
Poster: EII Workshop 2007Poster: EII Workshop 2007
Poster: EII Workshop 2007
Mahfuzul Haque
 
Poster: EII Winter School 2007
Poster: EII Winter School 2007Poster: EII Winter School 2007
Poster: EII Winter School 2007
Mahfuzul Haque
 
Poster: AVSS 2012
Poster: AVSS 2012Poster: AVSS 2012
Poster: AVSS 2012
Mahfuzul Haque
 

More from Mahfuzul Haque (20)

Dependency inversion using ports and adapters
Dependency inversion using ports and adaptersDependency inversion using ports and adapters
Dependency inversion using ports and adapters
 
Resilient machine learning systems for health analytics
Resilient machine learning systems for health analyticsResilient machine learning systems for health analytics
Resilient machine learning systems for health analytics
 
Talk 2012-icmew-event
Talk 2012-icmew-eventTalk 2012-icmew-event
Talk 2012-icmew-event
 
Talk 2011-buet-perception-event
Talk 2011-buet-perception-eventTalk 2011-buet-perception-event
Talk 2011-buet-perception-event
 
Talk 2010-monash-seminar-panic-driven-event-detection
Talk 2010-monash-seminar-panic-driven-event-detectionTalk 2010-monash-seminar-panic-driven-event-detection
Talk 2010-monash-seminar-panic-driven-event-detection
 
Talk 2009-monash-seminar-perception
Talk 2009-monash-seminar-perceptionTalk 2009-monash-seminar-perception
Talk 2009-monash-seminar-perception
 
Talk 2009-monash-seminar-intelligent-video-surveillance
Talk 2009-monash-seminar-intelligent-video-surveillanceTalk 2009-monash-seminar-intelligent-video-surveillance
Talk 2009-monash-seminar-intelligent-video-surveillance
 
Talk 2009-monash-open-day-surveillance
Talk 2009-monash-open-day-surveillanceTalk 2009-monash-open-day-surveillance
Talk 2009-monash-open-day-surveillance
 
Talk 2007-monash-seminar-behavior-recognition-framework
Talk 2007-monash-seminar-behavior-recognition-frameworkTalk 2007-monash-seminar-behavior-recognition-framework
Talk 2007-monash-seminar-behavior-recognition-framework
 
Kb hmm
Kb hmmKb hmm
Kb hmm
 
Kb gait-recognition
Kb gait-recognitionKb gait-recognition
Kb gait-recognition
 
Kb behaviour-recognition
Kb behaviour-recognitionKb behaviour-recognition
Kb behaviour-recognition
 
Poster: Monash Research Month 2009
Poster: Monash Research Month 2009Poster: Monash Research Month 2009
Poster: Monash Research Month 2009
 
Poster: Monash Research Month 2008
Poster: Monash Research Month 2008Poster: Monash Research Month 2008
Poster: Monash Research Month 2008
 
Poster: Monash Research Month 2007
Poster: Monash Research Month 2007Poster: Monash Research Month 2007
Poster: Monash Research Month 2007
 
Poster: ICPR 2008
Poster: ICPR 2008Poster: ICPR 2008
Poster: ICPR 2008
 
Poster: ICME 2010
Poster: ICME 2010Poster: ICME 2010
Poster: ICME 2010
 
Poster: EII Workshop 2007
Poster: EII Workshop 2007Poster: EII Workshop 2007
Poster: EII Workshop 2007
 
Poster: EII Winter School 2007
Poster: EII Winter School 2007Poster: EII Winter School 2007
Poster: EII Winter School 2007
 
Poster: AVSS 2012
Poster: AVSS 2012Poster: AVSS 2012
Poster: AVSS 2012
 

Recently uploaded

Columbus Data & Analytics Wednesdays - June 2024
Columbus Data & Analytics Wednesdays - June 2024Columbus Data & Analytics Wednesdays - June 2024
Columbus Data & Analytics Wednesdays - June 2024
Jason Packer
 
leewayhertz.com-AI in predictive maintenance Use cases technologies benefits ...
leewayhertz.com-AI in predictive maintenance Use cases technologies benefits ...leewayhertz.com-AI in predictive maintenance Use cases technologies benefits ...
leewayhertz.com-AI in predictive maintenance Use cases technologies benefits ...
alexjohnson7307
 
FREE A4 Cyber Security Awareness Posters-Social Engineering part 3
FREE A4 Cyber Security Awareness  Posters-Social Engineering part 3FREE A4 Cyber Security Awareness  Posters-Social Engineering part 3
FREE A4 Cyber Security Awareness Posters-Social Engineering part 3
Data Hops
 
Energy Efficient Video Encoding for Cloud and Edge Computing Instances
Energy Efficient Video Encoding for Cloud and Edge Computing InstancesEnergy Efficient Video Encoding for Cloud and Edge Computing Instances
Energy Efficient Video Encoding for Cloud and Edge Computing Instances
Alpen-Adria-Universität
 
Azure API Management to expose backend services securely
Azure API Management to expose backend services securelyAzure API Management to expose backend services securely
Azure API Management to expose backend services securely
Dinusha Kumarasiri
 
Choosing The Best AWS Service For Your Website + API.pptx
Choosing The Best AWS Service For Your Website + API.pptxChoosing The Best AWS Service For Your Website + API.pptx
Choosing The Best AWS Service For Your Website + API.pptx
Brandon Minnick, MBA
 
Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...
Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...
Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...
saastr
 
SAP S/4 HANA sourcing and procurement to Public cloud
SAP S/4 HANA sourcing and procurement to Public cloudSAP S/4 HANA sourcing and procurement to Public cloud
SAP S/4 HANA sourcing and procurement to Public cloud
maazsz111
 
Fueling AI with Great Data with Airbyte Webinar
Fueling AI with Great Data with Airbyte WebinarFueling AI with Great Data with Airbyte Webinar
Fueling AI with Great Data with Airbyte Webinar
Zilliz
 
Let's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with Slack
Let's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with SlackLet's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with Slack
Let's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with Slack
shyamraj55
 
JavaLand 2024: Application Development Green Masterplan
JavaLand 2024: Application Development Green MasterplanJavaLand 2024: Application Development Green Masterplan
JavaLand 2024: Application Development Green Masterplan
Miro Wengner
 
Building Production Ready Search Pipelines with Spark and Milvus
Building Production Ready Search Pipelines with Spark and MilvusBuilding Production Ready Search Pipelines with Spark and Milvus
Building Production Ready Search Pipelines with Spark and Milvus
Zilliz
 
Your One-Stop Shop for Python Success: Top 10 US Python Development Providers
Your One-Stop Shop for Python Success: Top 10 US Python Development ProvidersYour One-Stop Shop for Python Success: Top 10 US Python Development Providers
Your One-Stop Shop for Python Success: Top 10 US Python Development Providers
akankshawande
 
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
Edge AI and Vision Alliance
 
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
Jeffrey Haguewood
 
Monitoring and Managing Anomaly Detection on OpenShift.pdf
Monitoring and Managing Anomaly Detection on OpenShift.pdfMonitoring and Managing Anomaly Detection on OpenShift.pdf
Monitoring and Managing Anomaly Detection on OpenShift.pdf
Tosin Akinosho
 
Introduction of Cybersecurity with OSS at Code Europe 2024
Introduction of Cybersecurity with OSS  at Code Europe 2024Introduction of Cybersecurity with OSS  at Code Europe 2024
Introduction of Cybersecurity with OSS at Code Europe 2024
Hiroshi SHIBATA
 
Deep Dive: AI-Powered Marketing to Get More Leads and Customers with HyperGro...
Deep Dive: AI-Powered Marketing to Get More Leads and Customers with HyperGro...Deep Dive: AI-Powered Marketing to Get More Leads and Customers with HyperGro...
Deep Dive: AI-Powered Marketing to Get More Leads and Customers with HyperGro...
saastr
 
June Patch Tuesday
June Patch TuesdayJune Patch Tuesday
June Patch Tuesday
Ivanti
 
Trusted Execution Environment for Decentralized Process Mining
Trusted Execution Environment for Decentralized Process MiningTrusted Execution Environment for Decentralized Process Mining
Trusted Execution Environment for Decentralized Process Mining
LucaBarbaro3
 

Recently uploaded (20)

Columbus Data & Analytics Wednesdays - June 2024
Columbus Data & Analytics Wednesdays - June 2024Columbus Data & Analytics Wednesdays - June 2024
Columbus Data & Analytics Wednesdays - June 2024
 
leewayhertz.com-AI in predictive maintenance Use cases technologies benefits ...
leewayhertz.com-AI in predictive maintenance Use cases technologies benefits ...leewayhertz.com-AI in predictive maintenance Use cases technologies benefits ...
leewayhertz.com-AI in predictive maintenance Use cases technologies benefits ...
 
FREE A4 Cyber Security Awareness Posters-Social Engineering part 3
FREE A4 Cyber Security Awareness  Posters-Social Engineering part 3FREE A4 Cyber Security Awareness  Posters-Social Engineering part 3
FREE A4 Cyber Security Awareness Posters-Social Engineering part 3
 
Energy Efficient Video Encoding for Cloud and Edge Computing Instances
Energy Efficient Video Encoding for Cloud and Edge Computing InstancesEnergy Efficient Video Encoding for Cloud and Edge Computing Instances
Energy Efficient Video Encoding for Cloud and Edge Computing Instances
 
Azure API Management to expose backend services securely
Azure API Management to expose backend services securelyAzure API Management to expose backend services securely
Azure API Management to expose backend services securely
 
Choosing The Best AWS Service For Your Website + API.pptx
Choosing The Best AWS Service For Your Website + API.pptxChoosing The Best AWS Service For Your Website + API.pptx
Choosing The Best AWS Service For Your Website + API.pptx
 
Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...
Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...
Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...
 
SAP S/4 HANA sourcing and procurement to Public cloud
SAP S/4 HANA sourcing and procurement to Public cloudSAP S/4 HANA sourcing and procurement to Public cloud
SAP S/4 HANA sourcing and procurement to Public cloud
 
Fueling AI with Great Data with Airbyte Webinar
Fueling AI with Great Data with Airbyte WebinarFueling AI with Great Data with Airbyte Webinar
Fueling AI with Great Data with Airbyte Webinar
 
Let's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with Slack
Let's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with SlackLet's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with Slack
Let's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with Slack
 
JavaLand 2024: Application Development Green Masterplan
JavaLand 2024: Application Development Green MasterplanJavaLand 2024: Application Development Green Masterplan
JavaLand 2024: Application Development Green Masterplan
 
Building Production Ready Search Pipelines with Spark and Milvus
Building Production Ready Search Pipelines with Spark and MilvusBuilding Production Ready Search Pipelines with Spark and Milvus
Building Production Ready Search Pipelines with Spark and Milvus
 
Your One-Stop Shop for Python Success: Top 10 US Python Development Providers
Your One-Stop Shop for Python Success: Top 10 US Python Development ProvidersYour One-Stop Shop for Python Success: Top 10 US Python Development Providers
Your One-Stop Shop for Python Success: Top 10 US Python Development Providers
 
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
“Temporal Event Neural Networks: A More Efficient Alternative to the Transfor...
 
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
Salesforce Integration for Bonterra Impact Management (fka Social Solutions A...
 
Monitoring and Managing Anomaly Detection on OpenShift.pdf
Monitoring and Managing Anomaly Detection on OpenShift.pdfMonitoring and Managing Anomaly Detection on OpenShift.pdf
Monitoring and Managing Anomaly Detection on OpenShift.pdf
 
Introduction of Cybersecurity with OSS at Code Europe 2024
Introduction of Cybersecurity with OSS  at Code Europe 2024Introduction of Cybersecurity with OSS  at Code Europe 2024
Introduction of Cybersecurity with OSS at Code Europe 2024
 
Deep Dive: AI-Powered Marketing to Get More Leads and Customers with HyperGro...
Deep Dive: AI-Powered Marketing to Get More Leads and Customers with HyperGro...Deep Dive: AI-Powered Marketing to Get More Leads and Customers with HyperGro...
Deep Dive: AI-Powered Marketing to Get More Leads and Customers with HyperGro...
 
June Patch Tuesday
June Patch TuesdayJune Patch Tuesday
June Patch Tuesday
 
Trusted Execution Environment for Decentralized Process Mining
Trusted Execution Environment for Decentralized Process MiningTrusted Execution Environment for Decentralized Process Mining
Trusted Execution Environment for Decentralized Process Mining
 

Talk 2012-icmew-perception

  • 1. Information Technology Robust Background Subtraction Based on Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed Mahfuzul Haque and Manzur Murshed
  • 2. Agenda  Background Subtraction  Statistical Background Subtraction  Perception Inspired Background Subtraction  Dynamic Adaptation Speed  Experiments  Summary  Q&A Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 2
  • 3. Background Subtraction Input Output Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 3
  • 4. Background Subtraction: Challenges Basic Background Subtraction (e.g., BBS) - Current frame Challenges = Background Foreground Blob Dynamic Background Subtraction(e.g., MOG) Background     Illumination variation Local background motion Camera displacement Shadow and reflection Model Current frame Foreground Blob Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 4
  • 5. Statistical Background Subtraction ω1 ω2 ω3 σ12 σ22 σ32 µ1 µ2 µ3 road car shadow 65% 20% Statistical Approaches x x P(x) Our Hypothesis (Perception Inspired) x BBS: x = c MOG: x = c1σ μ 15% x x = c2b P(x) Te b Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 5
  • 6. Perception Inspired Background Subtraction x = c2 b Current Frame Detection with Low x Detection with High x x x P(x) b Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 6
  • 7. Weber’s Law How human visual system perceives noticeable intensity deviation from the background? Ernst Weber, an experimental psychologist in the 19th century, observed that the just-noticeable increment ΔI is linearly proportional to the background intensity I. ΔI = c2I Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 7
  • 8. Weber’s Law Ernst Weber, an experimental psychologist in the 19th century, observed that the just-noticeable increment ΔI is linearly proportional to the background intensity I. ? x x ΔI = c2I x x = c2 b P(x) b Te b Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 8
  • 9. Perceptual tolerance of HVS What is the perceptual tolerance level in distinguishing distorted intensity measures? p dB Method 1 Reference q dB Method 2 Image Distorted Images |p – q| < 0.5 dB Not perceivable by human visual system Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 9
  • 10. Our Problem: c2 = ? x = c2 b x x P(x) Weber’s Law x = c2b Perceptual Threshold, TP (0.5 dB)  255 20 log10   bx      20 log  255 10  b  x      1  2TP  b Te Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 10
  • 11. Linear Relationship x b Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 11
  • 12. Rod and Cone  Rods and Cones are two different types of photoreceptor cells in the retina of human eye  Rods – Operate in less intense light – Responsible for scotopic vision (night vision)  Cones – Operate in relatively bright light – Responsible for photopic (color vision) Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 12
  • 13. Error Sensitivity in Darker Background Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 13
  • 14. Piece-wise Liner Relationship Scotopic Vision (R) Photopic Vision (C) Te Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 14
  • 15. Dynamic Adaptation Speed •Sleeping person problem •Walking person problem Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 15
  • 16. Dynamic Adaptation Speed Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 16
  • 17. Dynamic Adaptation Speed Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 17
  • 18. Dynamic Adaptation Speed Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 18
  • 19. Experiments Test Sequences  Total 50 test sequences from 8 different sources  Scenario distribution      Indoor Outdoor Multimodal Shadow and Reflection Low background-foreground contrast False Classification Evaluation  Qualitative and quantitative comparison:  MOG (S&G) (TPAMI, 2000) False Positive (FP) False Negative (FN)  MOG (Lee) (TPAMI, 2005)  ViBe (TIP, 2011) Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 19
  • 20. Test Sequences PETS (9) Wallflower (7) UCF (7) IBM (11) CAVIAR (7) Te VSSN06 (7) Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed Other (2) December 30, 2013 20
  • 21. Experiments Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 21
  • 22. Experiments Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 22
  • 23. Experiments Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 23
  • 25. Summary  Realistic background value prediction: high model agility and superior detection quality at fast learning rate.  No context related information: high stability across changing scenarios.  Perception based detection threshold: superior detection quality in terms of shadow, noise, and reflection.  Perceptual model similarity: optimal number of models throughout the system life cycle.  Parameter-less background subtraction: ideal for realtime video analytics. Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 25
  • 26. Q&A Perceptual Mixture-of-Gaussians with Dynamic Adaptation Speed December 30, 2013 26