SlideShare une entreprise Scribd logo
Project Presentation
How to detect vanishing points
on architectural scenes
PERNEY Benjamin - 20121
Contents
• Why compute vanishing points ?
• Application on architectural scenes
• What is a vanishing point ?
• Chasles-Steiner theorem
• Algorithm
• Results
• Summary & conclusion
PERNEY Benjamin - 20122
Vanishing points :
Why ?
PERNEY Benjamin - 20123
2D image : 3D information LOST• Real life : Parallel lines
• Environments created by humans contains many parallel lines
Vanishing points :
Why ?
• Providing strong information about 3D
structure of a scene (Best way)
• Applications :
– Camera calibration
– Augmented Reality
– Create 3D map
– Help land surveyor to
align buildings
PERNEY Benjamin - 20124
• Compute the photo orientation
Vertical of the scene
• Minimize the error
High numbers of segments
Purpose :
Application on architectural scene
PERNEY Benjamin - 20125
Allows to :
Vanishing points :
What is it ?
PERNEY Benjamin - 20126
2D image : Converging linesReal life : Parallel lines
Vanishing point
Vanishing line
Related works
• Different methods were proposed :
– Gaussian sphere : more accurate but complex
– Projective Geometry approach
• Chasles-Steiner theorem :
PERNEY Benjamin - 20127
« An homography between two
bundles of converging lines define
a conic section, and reciprocally »
Algorithm steps
PERNEY Benjamin - 20128
1. Extract segments
• Based on Canny-Deriche detection
2. Transform segments in points
• Apply the Chasles-Steiner theorem
3. Extract circles among previous points found
• RanSac method adapted
4. Compute coordinates of the vanishing points
2 steps :
-Smoothing
- Calculation of magnitude and gradient direction
- Non-maximum suppression
- Hysteresis thresholding
Extract segments
Canny-Deriche
detection
Local maxima
detection
Polygonization
PERNEY Benjamin - 20129
4th step : Thanks to the circle parameters (especially, center coordinates),
We can determine the vanishing points as the opposite point of the image origin
Chasles-Steiner Theorem
• Applied to vanishing points computing :
PERNEY Benjamin - 201210
O
P
C
Image
H1
H2
H3
S1
S2
S3
1st step : Thanks to the segments, compute the carrier lines2nd step : H points are computed : OH and Segments should make a 90° angle3rd step : A circle is found passing through the H points
Chasles-Steiner Theorem
PERNEY Benjamin - 201211
Extract circles
RanSac method
PERNEY Benjamin - 201212
O
2. Compute the circle which intersect the 2 points and O1. Two points H are chosen randomly among all3. Create a band of epsilon size and count the number of H points inside4. Repeat steps 1 to 3 many times5. Keep in memory the 2 H points and captured points
-> Remove them from the beginning ensemble and iterate
Compute the P coordinates
• XP = 2 x Xc
• YP = 2 x Yc
• We could compute uncertainty with the variance-covariance matrix
PERNEY Benjamin - 201213
O
P
C
Results
• On 100 different images :
PERNEY Benjamin - 201214
Pourcentage of correct
detection of the vertical
vanishing point
100%
Pourcentage of correct
detection of the horizontal
vanishing points
92%
Good performance
Results
• Issues :
– Segments near to the origin -> the H point
position will change a lot the circle
– Noisy image : edge detection not precise
– Complex architectures whith many curves
PERNEY Benjamin - 201215
Results & improvements
PERNEY Benjamin - 201216
Summary
PERNEY Benjamin - 201217
(a) (b)
(c) (d)
Bibliography
• Automatic detection of vanishing points and
their uncertainty based on projective
geometry, M. Kalantari, F. Jung, JP. Guédon, N.
Paparoditis
• Détéction entièrement automatique de points
de fuite dans des scènes architecturales
urbaines, M. Kalantari, F. Jung
• A new Approach to Vanishing Point Detection
in Architectural Environments, Carsten Rother
PERNEY Benjamin - 201218
Conclusion
• Interesting and contemporary subject
• What’s next : Smartphone applications etc.
SOME QUESTIONS ?
PERNEY Benjamin - 201219

Contenu connexe

Similaire à How to detect vanishing points on architectural scenes ?

Two marks with answers ME6501 CAD
Two marks with answers ME6501 CADTwo marks with answers ME6501 CAD
Two marks with answers ME6501 CADPriscilla CPG
 
Computer Vision - Single View
Computer Vision - Single ViewComputer Vision - Single View
Computer Vision - Single ViewWael Badawy
 
Lecture 6-computer vision features descriptors matching
Lecture 6-computer vision features descriptors matchingLecture 6-computer vision features descriptors matching
Lecture 6-computer vision features descriptors matchingcairo university
 
Computer Vision panoramas
Computer Vision  panoramasComputer Vision  panoramas
Computer Vision panoramasWael Badawy
 
Angel6E25educationmaterial for reference
Angel6E25educationmaterial for referenceAngel6E25educationmaterial for reference
Angel6E25educationmaterial for referencekanchang3684
 
Fisheye Omnidirectional View in Autonomous Driving II
Fisheye Omnidirectional View in Autonomous Driving IIFisheye Omnidirectional View in Autonomous Driving II
Fisheye Omnidirectional View in Autonomous Driving IIYu Huang
 
Stixel based real time object detection for ADAS using surface normal
Stixel based real time object detection for ADAS using surface normalStixel based real time object detection for ADAS using surface normal
Stixel based real time object detection for ADAS using surface normalTaeKang Woo
 
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...grssieee
 
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...grssieee
 
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...grssieee
 
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...grssieee
 
Cameras
CamerasCameras
Camerasgayu r
 
Lec13 stereo converted
Lec13 stereo convertedLec13 stereo converted
Lec13 stereo convertedBaliThorat1
 
Passive stereo vision with deep learning
Passive stereo vision with deep learningPassive stereo vision with deep learning
Passive stereo vision with deep learningYu Huang
 

Similaire à How to detect vanishing points on architectural scenes ? (20)

Two marks with answers ME6501 CAD
Two marks with answers ME6501 CADTwo marks with answers ME6501 CAD
Two marks with answers ME6501 CAD
 
Computer Vision - Single View
Computer Vision - Single ViewComputer Vision - Single View
Computer Vision - Single View
 
Lecture 6-computer vision features descriptors matching
Lecture 6-computer vision features descriptors matchingLecture 6-computer vision features descriptors matching
Lecture 6-computer vision features descriptors matching
 
Computer Vision panoramas
Computer Vision  panoramasComputer Vision  panoramas
Computer Vision panoramas
 
05_optics.ppt
05_optics.ppt05_optics.ppt
05_optics.ppt
 
05_optics.ppt
05_optics.ppt05_optics.ppt
05_optics.ppt
 
05_optics.ppt
05_optics.ppt05_optics.ppt
05_optics.ppt
 
Angel6E25educationmaterial for reference
Angel6E25educationmaterial for referenceAngel6E25educationmaterial for reference
Angel6E25educationmaterial for reference
 
Fisheye Omnidirectional View in Autonomous Driving II
Fisheye Omnidirectional View in Autonomous Driving IIFisheye Omnidirectional View in Autonomous Driving II
Fisheye Omnidirectional View in Autonomous Driving II
 
Stixel based real time object detection for ADAS using surface normal
Stixel based real time object detection for ADAS using surface normalStixel based real time object detection for ADAS using surface normal
Stixel based real time object detection for ADAS using surface normal
 
cv1.ppt
cv1.pptcv1.ppt
cv1.ppt
 
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...
 
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...
 
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...
 
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...
A SELF-ADJUSTIVE GEOMETRIC CORRECTION METHOD FOR SERIOUSLY OBLIQUE AERO IMAGE...
 
Lecture24
Lecture24Lecture24
Lecture24
 
Cameras
CamerasCameras
Cameras
 
Lec13 stereo converted
Lec13 stereo convertedLec13 stereo converted
Lec13 stereo converted
 
Schematic eye and cardinal points
Schematic eye and cardinal pointsSchematic eye and cardinal points
Schematic eye and cardinal points
 
Passive stereo vision with deep learning
Passive stereo vision with deep learningPassive stereo vision with deep learning
Passive stereo vision with deep learning
 

Dernier

Choosing the Right FDO Deployment Model for Your Application _ Geoffrey at In...
Choosing the Right FDO Deployment Model for Your Application _ Geoffrey at In...Choosing the Right FDO Deployment Model for Your Application _ Geoffrey at In...
Choosing the Right FDO Deployment Model for Your Application _ Geoffrey at In...FIDO Alliance
 
The Metaverse: Are We There Yet?
The  Metaverse:    Are   We  There  Yet?The  Metaverse:    Are   We  There  Yet?
The Metaverse: Are We There Yet?Mark Billinghurst
 
Intro in Product Management - Коротко про професію продакт менеджера
Intro in Product Management - Коротко про професію продакт менеджераIntro in Product Management - Коротко про професію продакт менеджера
Intro in Product Management - Коротко про професію продакт менеджераMark Opanasiuk
 
Buy Epson EcoTank L3210 Colour Printer Online.pptx
Buy Epson EcoTank L3210 Colour Printer Online.pptxBuy Epson EcoTank L3210 Colour Printer Online.pptx
Buy Epson EcoTank L3210 Colour Printer Online.pptxEasyPrinterHelp
 
What's New in Teams Calling, Meetings and Devices April 2024
What's New in Teams Calling, Meetings and Devices April 2024What's New in Teams Calling, Meetings and Devices April 2024
What's New in Teams Calling, Meetings and Devices April 2024Stephanie Beckett
 
Agentic RAG What it is its types applications and implementation.pdf
Agentic RAG What it is its types applications and implementation.pdfAgentic RAG What it is its types applications and implementation.pdf
Agentic RAG What it is its types applications and implementation.pdfChristopherTHyatt
 
PLAI - Acceleration Program for Generative A.I. Startups
PLAI - Acceleration Program for Generative A.I. StartupsPLAI - Acceleration Program for Generative A.I. Startups
PLAI - Acceleration Program for Generative A.I. StartupsStefano
 
Custom Approval Process: A New Perspective, Pavel Hrbacek & Anindya Halder
Custom Approval Process: A New Perspective, Pavel Hrbacek & Anindya HalderCustom Approval Process: A New Perspective, Pavel Hrbacek & Anindya Halder
Custom Approval Process: A New Perspective, Pavel Hrbacek & Anindya HalderCzechDreamin
 
Top 10 Symfony Development Companies 2024
Top 10 Symfony Development Companies 2024Top 10 Symfony Development Companies 2024
Top 10 Symfony Development Companies 2024TopCSSGallery
 
Introduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdf
Introduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdfIntroduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdf
Introduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdfFIDO Alliance
 
FDO for Camera, Sensor and Networking Device – Commercial Solutions from VinC...
FDO for Camera, Sensor and Networking Device – Commercial Solutions from VinC...FDO for Camera, Sensor and Networking Device – Commercial Solutions from VinC...
FDO for Camera, Sensor and Networking Device – Commercial Solutions from VinC...FIDO Alliance
 
ASRock Industrial FDO Solutions in Action for Industrial Edge AI _ Kenny at A...
ASRock Industrial FDO Solutions in Action for Industrial Edge AI _ Kenny at A...ASRock Industrial FDO Solutions in Action for Industrial Edge AI _ Kenny at A...
ASRock Industrial FDO Solutions in Action for Industrial Edge AI _ Kenny at A...FIDO Alliance
 
IOS-PENTESTING-BEGINNERS-PRACTICAL-GUIDE-.pptx
IOS-PENTESTING-BEGINNERS-PRACTICAL-GUIDE-.pptxIOS-PENTESTING-BEGINNERS-PRACTICAL-GUIDE-.pptx
IOS-PENTESTING-BEGINNERS-PRACTICAL-GUIDE-.pptxAbida Shariff
 
Where to Learn More About FDO _ Richard at FIDO Alliance.pdf
Where to Learn More About FDO _ Richard at FIDO Alliance.pdfWhere to Learn More About FDO _ Richard at FIDO Alliance.pdf
Where to Learn More About FDO _ Richard at FIDO Alliance.pdfFIDO Alliance
 
Optimizing NoSQL Performance Through Observability
Optimizing NoSQL Performance Through ObservabilityOptimizing NoSQL Performance Through Observability
Optimizing NoSQL Performance Through ObservabilityScyllaDB
 
AI presentation and introduction - Retrieval Augmented Generation RAG 101
AI presentation and introduction - Retrieval Augmented Generation RAG 101AI presentation and introduction - Retrieval Augmented Generation RAG 101
AI presentation and introduction - Retrieval Augmented Generation RAG 101vincent683379
 
Buy Epson EcoTank L3210 Colour Printer Online.pdf
Buy Epson EcoTank L3210 Colour Printer Online.pdfBuy Epson EcoTank L3210 Colour Printer Online.pdf
Buy Epson EcoTank L3210 Colour Printer Online.pdfEasyPrinterHelp
 
Simplified FDO Manufacturing Flow with TPMs _ Liam at Infineon.pdf
Simplified FDO Manufacturing Flow with TPMs _ Liam at Infineon.pdfSimplified FDO Manufacturing Flow with TPMs _ Liam at Infineon.pdf
Simplified FDO Manufacturing Flow with TPMs _ Liam at Infineon.pdfFIDO Alliance
 
Designing for Hardware Accessibility at Comcast
Designing for Hardware Accessibility at ComcastDesigning for Hardware Accessibility at Comcast
Designing for Hardware Accessibility at ComcastUXDXConf
 
Linux Foundation Edge _ Overview of FDO Software Components _ Randy at Intel.pdf
Linux Foundation Edge _ Overview of FDO Software Components _ Randy at Intel.pdfLinux Foundation Edge _ Overview of FDO Software Components _ Randy at Intel.pdf
Linux Foundation Edge _ Overview of FDO Software Components _ Randy at Intel.pdfFIDO Alliance
 

Dernier (20)

Choosing the Right FDO Deployment Model for Your Application _ Geoffrey at In...
Choosing the Right FDO Deployment Model for Your Application _ Geoffrey at In...Choosing the Right FDO Deployment Model for Your Application _ Geoffrey at In...
Choosing the Right FDO Deployment Model for Your Application _ Geoffrey at In...
 
The Metaverse: Are We There Yet?
The  Metaverse:    Are   We  There  Yet?The  Metaverse:    Are   We  There  Yet?
The Metaverse: Are We There Yet?
 
Intro in Product Management - Коротко про професію продакт менеджера
Intro in Product Management - Коротко про професію продакт менеджераIntro in Product Management - Коротко про професію продакт менеджера
Intro in Product Management - Коротко про професію продакт менеджера
 
Buy Epson EcoTank L3210 Colour Printer Online.pptx
Buy Epson EcoTank L3210 Colour Printer Online.pptxBuy Epson EcoTank L3210 Colour Printer Online.pptx
Buy Epson EcoTank L3210 Colour Printer Online.pptx
 
What's New in Teams Calling, Meetings and Devices April 2024
What's New in Teams Calling, Meetings and Devices April 2024What's New in Teams Calling, Meetings and Devices April 2024
What's New in Teams Calling, Meetings and Devices April 2024
 
Agentic RAG What it is its types applications and implementation.pdf
Agentic RAG What it is its types applications and implementation.pdfAgentic RAG What it is its types applications and implementation.pdf
Agentic RAG What it is its types applications and implementation.pdf
 
PLAI - Acceleration Program for Generative A.I. Startups
PLAI - Acceleration Program for Generative A.I. StartupsPLAI - Acceleration Program for Generative A.I. Startups
PLAI - Acceleration Program for Generative A.I. Startups
 
Custom Approval Process: A New Perspective, Pavel Hrbacek & Anindya Halder
Custom Approval Process: A New Perspective, Pavel Hrbacek & Anindya HalderCustom Approval Process: A New Perspective, Pavel Hrbacek & Anindya Halder
Custom Approval Process: A New Perspective, Pavel Hrbacek & Anindya Halder
 
Top 10 Symfony Development Companies 2024
Top 10 Symfony Development Companies 2024Top 10 Symfony Development Companies 2024
Top 10 Symfony Development Companies 2024
 
Introduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdf
Introduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdfIntroduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdf
Introduction to FDO and How It works Applications _ Richard at FIDO Alliance.pdf
 
FDO for Camera, Sensor and Networking Device – Commercial Solutions from VinC...
FDO for Camera, Sensor and Networking Device – Commercial Solutions from VinC...FDO for Camera, Sensor and Networking Device – Commercial Solutions from VinC...
FDO for Camera, Sensor and Networking Device – Commercial Solutions from VinC...
 
ASRock Industrial FDO Solutions in Action for Industrial Edge AI _ Kenny at A...
ASRock Industrial FDO Solutions in Action for Industrial Edge AI _ Kenny at A...ASRock Industrial FDO Solutions in Action for Industrial Edge AI _ Kenny at A...
ASRock Industrial FDO Solutions in Action for Industrial Edge AI _ Kenny at A...
 
IOS-PENTESTING-BEGINNERS-PRACTICAL-GUIDE-.pptx
IOS-PENTESTING-BEGINNERS-PRACTICAL-GUIDE-.pptxIOS-PENTESTING-BEGINNERS-PRACTICAL-GUIDE-.pptx
IOS-PENTESTING-BEGINNERS-PRACTICAL-GUIDE-.pptx
 
Where to Learn More About FDO _ Richard at FIDO Alliance.pdf
Where to Learn More About FDO _ Richard at FIDO Alliance.pdfWhere to Learn More About FDO _ Richard at FIDO Alliance.pdf
Where to Learn More About FDO _ Richard at FIDO Alliance.pdf
 
Optimizing NoSQL Performance Through Observability
Optimizing NoSQL Performance Through ObservabilityOptimizing NoSQL Performance Through Observability
Optimizing NoSQL Performance Through Observability
 
AI presentation and introduction - Retrieval Augmented Generation RAG 101
AI presentation and introduction - Retrieval Augmented Generation RAG 101AI presentation and introduction - Retrieval Augmented Generation RAG 101
AI presentation and introduction - Retrieval Augmented Generation RAG 101
 
Buy Epson EcoTank L3210 Colour Printer Online.pdf
Buy Epson EcoTank L3210 Colour Printer Online.pdfBuy Epson EcoTank L3210 Colour Printer Online.pdf
Buy Epson EcoTank L3210 Colour Printer Online.pdf
 
Simplified FDO Manufacturing Flow with TPMs _ Liam at Infineon.pdf
Simplified FDO Manufacturing Flow with TPMs _ Liam at Infineon.pdfSimplified FDO Manufacturing Flow with TPMs _ Liam at Infineon.pdf
Simplified FDO Manufacturing Flow with TPMs _ Liam at Infineon.pdf
 
Designing for Hardware Accessibility at Comcast
Designing for Hardware Accessibility at ComcastDesigning for Hardware Accessibility at Comcast
Designing for Hardware Accessibility at Comcast
 
Linux Foundation Edge _ Overview of FDO Software Components _ Randy at Intel.pdf
Linux Foundation Edge _ Overview of FDO Software Components _ Randy at Intel.pdfLinux Foundation Edge _ Overview of FDO Software Components _ Randy at Intel.pdf
Linux Foundation Edge _ Overview of FDO Software Components _ Randy at Intel.pdf
 

How to detect vanishing points on architectural scenes ?

  • 1. Project Presentation How to detect vanishing points on architectural scenes PERNEY Benjamin - 20121
  • 2. Contents • Why compute vanishing points ? • Application on architectural scenes • What is a vanishing point ? • Chasles-Steiner theorem • Algorithm • Results • Summary & conclusion PERNEY Benjamin - 20122
  • 3. Vanishing points : Why ? PERNEY Benjamin - 20123 2D image : 3D information LOST• Real life : Parallel lines • Environments created by humans contains many parallel lines
  • 4. Vanishing points : Why ? • Providing strong information about 3D structure of a scene (Best way) • Applications : – Camera calibration – Augmented Reality – Create 3D map – Help land surveyor to align buildings PERNEY Benjamin - 20124
  • 5. • Compute the photo orientation Vertical of the scene • Minimize the error High numbers of segments Purpose : Application on architectural scene PERNEY Benjamin - 20125 Allows to :
  • 6. Vanishing points : What is it ? PERNEY Benjamin - 20126 2D image : Converging linesReal life : Parallel lines Vanishing point Vanishing line
  • 7. Related works • Different methods were proposed : – Gaussian sphere : more accurate but complex – Projective Geometry approach • Chasles-Steiner theorem : PERNEY Benjamin - 20127 « An homography between two bundles of converging lines define a conic section, and reciprocally »
  • 8. Algorithm steps PERNEY Benjamin - 20128 1. Extract segments • Based on Canny-Deriche detection 2. Transform segments in points • Apply the Chasles-Steiner theorem 3. Extract circles among previous points found • RanSac method adapted 4. Compute coordinates of the vanishing points
  • 9. 2 steps : -Smoothing - Calculation of magnitude and gradient direction - Non-maximum suppression - Hysteresis thresholding Extract segments Canny-Deriche detection Local maxima detection Polygonization PERNEY Benjamin - 20129
  • 10. 4th step : Thanks to the circle parameters (especially, center coordinates), We can determine the vanishing points as the opposite point of the image origin Chasles-Steiner Theorem • Applied to vanishing points computing : PERNEY Benjamin - 201210 O P C Image H1 H2 H3 S1 S2 S3 1st step : Thanks to the segments, compute the carrier lines2nd step : H points are computed : OH and Segments should make a 90° angle3rd step : A circle is found passing through the H points
  • 12. Extract circles RanSac method PERNEY Benjamin - 201212 O 2. Compute the circle which intersect the 2 points and O1. Two points H are chosen randomly among all3. Create a band of epsilon size and count the number of H points inside4. Repeat steps 1 to 3 many times5. Keep in memory the 2 H points and captured points -> Remove them from the beginning ensemble and iterate
  • 13. Compute the P coordinates • XP = 2 x Xc • YP = 2 x Yc • We could compute uncertainty with the variance-covariance matrix PERNEY Benjamin - 201213 O P C
  • 14. Results • On 100 different images : PERNEY Benjamin - 201214 Pourcentage of correct detection of the vertical vanishing point 100% Pourcentage of correct detection of the horizontal vanishing points 92% Good performance
  • 15. Results • Issues : – Segments near to the origin -> the H point position will change a lot the circle – Noisy image : edge detection not precise – Complex architectures whith many curves PERNEY Benjamin - 201215
  • 16. Results & improvements PERNEY Benjamin - 201216
  • 17. Summary PERNEY Benjamin - 201217 (a) (b) (c) (d)
  • 18. Bibliography • Automatic detection of vanishing points and their uncertainty based on projective geometry, M. Kalantari, F. Jung, JP. Guédon, N. Paparoditis • Détéction entièrement automatique de points de fuite dans des scènes architecturales urbaines, M. Kalantari, F. Jung • A new Approach to Vanishing Point Detection in Architectural Environments, Carsten Rother PERNEY Benjamin - 201218
  • 19. Conclusion • Interesting and contemporary subject • What’s next : Smartphone applications etc. SOME QUESTIONS ? PERNEY Benjamin - 201219