SlideShare a Scribd company logo
1 of 49
Visual Attention: Detecting Saliency on Images Vicente Ordonez Department of Computer Science State University of New York Stony Brook, NY 11790
I will be working mainly on the following paper Learning to Detect a Salient Object. T. Liu, J. Sun, N. Zheng, X. Tang, H. Shum. (Xian Jiaotong University and Microsoft Research Asia) from CVPR 2007.  http://research.microsoft.com/en-us/um/people/jiansun/papers/SalientDetection_CVPR07.pdf
What is Saliency? What is Visual Attention? “Everyone knows what attention is...” —William James, 1890
This is a problem of… Arbitrary object detection? Background / Foreground segmentation? Modeling Visual Attention?
The Method Features:  Multiscale Contrast    (Done!) Center surround histogram   (Mostly Done!) (Done!) Color spatial distribution (Done!) Supervised learning using Conditional Random Fields to determine the parameters to combine the features obtained above.  (Done!) [I will use a labeled dataset of 5000 images provided by Microsoft Research Asia!]
Multiscale Contrast Function Generate the Gaussian Pyramid for the input image. For each level in the pyramid  Do gaussian blurring Do resampling I’m using a 6 levels Gaussian pyramid for each RGB channel.
How a Gaussian pyramid looks like Figure from David Forsyth
Generate contrast maps for each level of the Pyramid. Sum all of the results to produce the final multiscale contrast map. The two steps mentioned above are described in this formula: Multiscale Contrast Function
Input image
Contrast maps
Contrast maps Original image Contrast map at level 1 Contrast map at level 4 Contrast map at level 6
Multiscale Contrast Map Output
Center Surround Histogram Feature ,[object Object]
For each possible rectangle with a reasonable size and aspect ratio
Create a surrounding rectangle and calculate the histogram of the rectangle and the surrounding area.
Pick and record the rectangle that maximizes the Chi-Square distance between the two histograms calculated above and also record the Chi-Square distance.,[object Object]
Center Surround Histogram Feature The algorithm as described before is computationally expensive…  It is required to use a technique called Integral Histogram. It allows you fast calculation of the histogram of any given rectangular region of an image. The algorithm was introduced in: “Integral Histogram: A Fast Way to Extract Histograms in Cartesian Spaces” by FatihPorikli, Mitsubishi Electric Research Lab in CVPR 2005.
Center Surround Histogram Feature Use the Chi Square Distances Map and the Map of Most Salient Rectangle Regions per pixel to generate the Center Surround Histogram Feature using the next formula:
Center Surround Histogram Results Using my Implementation        (15.2 sec, size = 245x384) Results Reported in the Paper
Center Surround Histogram Results Using my Implementation        (13.6 sec, size = 247x346) Results Reported in the Paper
Center Surround Histogram Results Using my Implementation        (10.2 sec, size = 248x277)
More Results
More Results
More results
More Results
More Results
More Results
More Results
More Results
More Results
More Results
More Results
Color Spatial Distribution
Color Spatial Distribution Make an initial clustering of the colors in the image using k-means.  Further refine the clusters by using Gaussian Mixture Models. The Gaussian Mixture Model parameters are calculated using the EM algorithm. I am using 5 clusters (5 colors) per image. And the results look similar to those presented in the paper with an execution time of around 17 seconds per image.
Color Spatial Distribution Calculate the vertical variance of the horizontal positions of the pixels for each cluster. And then the same for the vertical positions.  Sum the variances and use this value to weight more those clusters with less spatial variance. Penalize the clusters that contain the majority of its pixels away from the center of the image.
Color Spatial Distribution
Color Spatial Distribution
Color Spatial Distribution
Color Spatial Distribution
Color Spatial Distribution
Color Spatial Distribution
Color Spatial Distribution
Color Spatial Distribution
Combine Features Together
Conditional Random Field Training and Inference Accelerated Training of Conditional Random Fields with Stochastic Meta-Descent S Vishwanathan, N. Schraudolph, M. Schmidt, K. Murphy. ICML'06 (Intl Conf on Machine Learning).  I did the training using this toolbox from the above paper: http://people.cs.ubc.ca/~murphyk/Software/CRF/crf.html
Mask outputs using CRF inference Input                  M-Contrast-map         Center Surr. Hist.       Color Spatial Var. Input                      Combined features                    Ground truth
Mask outputs using CRF inference Input                  M-Contrast-map         Center Surr. Hist.       Color Spatial Var. Input                      Combined features                    Ground truth
Mask outputs using CRF inference Input                  M-Contrast-map         Center Surr. Hist.       Color Spatial Var. Input                 Combined features        Ground truth
Mask outputs using CRF inference Input                  M-Contrast-map         Center Surr. Hist.       Color Spatial Var. Input                 Combined features        Ground truth

More Related Content

What's hot

Lecture9 camera calibration
Lecture9 camera calibrationLecture9 camera calibration
Lecture9 camera calibration
zukun
 

What's hot (20)

NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis
NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis
NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis
 
Bayes Classification
Bayes ClassificationBayes Classification
Bayes Classification
 
3.2 partitioning methods
3.2 partitioning methods3.2 partitioning methods
3.2 partitioning methods
 
Deep Semi-supervised Learning methods
Deep Semi-supervised Learning methodsDeep Semi-supervised Learning methods
Deep Semi-supervised Learning methods
 
Parametric & Non-Parametric Machine Learning (Supervised ML)
Parametric & Non-Parametric Machine Learning (Supervised ML)Parametric & Non-Parametric Machine Learning (Supervised ML)
Parametric & Non-Parametric Machine Learning (Supervised ML)
 
Image segmentation with deep learning
Image segmentation with deep learningImage segmentation with deep learning
Image segmentation with deep learning
 
Image Enhancement - Point Processing
Image Enhancement - Point ProcessingImage Enhancement - Point Processing
Image Enhancement - Point Processing
 
Introduction to OpenCV
Introduction to OpenCVIntroduction to OpenCV
Introduction to OpenCV
 
Handwritten Digit Recognition(Convolutional Neural Network) PPT
Handwritten Digit Recognition(Convolutional Neural Network) PPTHandwritten Digit Recognition(Convolutional Neural Network) PPT
Handwritten Digit Recognition(Convolutional Neural Network) PPT
 
Machine Learning - Dataset Preparation
Machine Learning - Dataset PreparationMachine Learning - Dataset Preparation
Machine Learning - Dataset Preparation
 
Artificial Intelligence -- Search Algorithms
Artificial Intelligence-- Search Algorithms Artificial Intelligence-- Search Algorithms
Artificial Intelligence -- Search Algorithms
 
5 csp
5 csp5 csp
5 csp
 
Lecture9 camera calibration
Lecture9 camera calibrationLecture9 camera calibration
Lecture9 camera calibration
 
Cloud applications - Protein Structure Predication and gene expression data...
Cloud applications - Protein Structure Predication  and  gene expression data...Cloud applications - Protein Structure Predication  and  gene expression data...
Cloud applications - Protein Structure Predication and gene expression data...
 
Brain Tumour Detection.pptx
Brain Tumour Detection.pptxBrain Tumour Detection.pptx
Brain Tumour Detection.pptx
 
Computer graphics notes
Computer graphics notesComputer graphics notes
Computer graphics notes
 
Crime rate analysis using k nn in python
Crime rate analysis using k nn in python Crime rate analysis using k nn in python
Crime rate analysis using k nn in python
 
Introduction to Grad-CAM (complete version)
Introduction to Grad-CAM (complete version)Introduction to Grad-CAM (complete version)
Introduction to Grad-CAM (complete version)
 
Research Trends in Editing image using GAN (TAGAN, Editable GAN)
Research Trends in Editing image using GAN (TAGAN, Editable GAN)Research Trends in Editing image using GAN (TAGAN, Editable GAN)
Research Trends in Editing image using GAN (TAGAN, Editable GAN)
 
AlexNet(ImageNet Classification with Deep Convolutional Neural Networks)
AlexNet(ImageNet Classification with Deep Convolutional Neural Networks)AlexNet(ImageNet Classification with Deep Convolutional Neural Networks)
AlexNet(ImageNet Classification with Deep Convolutional Neural Networks)
 

Viewers also liked (6)

Iccv11 salientobjectdetection
Iccv11 salientobjectdetectionIccv11 salientobjectdetection
Iccv11 salientobjectdetection
 
Salient Point Detection
Salient Point DetectionSalient Point Detection
Salient Point Detection
 
Visual attention
Visual attentionVisual attention
Visual attention
 
Visual Attention & Processing with Visual-Only IM
Visual Attention & Processing with Visual-Only IMVisual Attention & Processing with Visual-Only IM
Visual Attention & Processing with Visual-Only IM
 
Chris Atherton at TCUK09
Chris Atherton at TCUK09Chris Atherton at TCUK09
Chris Atherton at TCUK09
 
Visual attention: models and performance
Visual attention: models and performanceVisual attention: models and performance
Visual attention: models and performance
 

Similar to Visual Saliency: Learning to Detect Salient Objects

Miniproject final group 14
Miniproject final group 14Miniproject final group 14
Miniproject final group 14
Ashish Mundhra
 
Urban 3D Semantic Modelling Using Stereo Vision, ICRA 2013
Urban 3D Semantic Modelling Using Stereo Vision, ICRA 2013Urban 3D Semantic Modelling Using Stereo Vision, ICRA 2013
Urban 3D Semantic Modelling Using Stereo Vision, ICRA 2013
Sunando Sengupta
 
Automatic Detection of Window Regions in Indoor Point Clouds Using R-CNN
Automatic Detection of Window Regions in Indoor Point Clouds Using R-CNNAutomatic Detection of Window Regions in Indoor Point Clouds Using R-CNN
Automatic Detection of Window Regions in Indoor Point Clouds Using R-CNN
Zihao(Gerald) Zhang
 
Super Resolution of Image
Super Resolution of ImageSuper Resolution of Image
Super Resolution of Image
Satheesh K
 
A STUDY AND ANALYSIS OF DIFFERENT EDGE DETECTION TECHNIQUES
A STUDY AND ANALYSIS OF DIFFERENT EDGE DETECTION TECHNIQUESA STUDY AND ANALYSIS OF DIFFERENT EDGE DETECTION TECHNIQUES
A STUDY AND ANALYSIS OF DIFFERENT EDGE DETECTION TECHNIQUES
cscpconf
 

Similar to Visual Saliency: Learning to Detect Salient Objects (20)

Praseed Pai
Praseed PaiPraseed Pai
Praseed Pai
 
Mirko Lucchese - Deep Image Processing
Mirko Lucchese - Deep Image ProcessingMirko Lucchese - Deep Image Processing
Mirko Lucchese - Deep Image Processing
 
Conception_et_realisation_dun_site_Web_d.pdf
Conception_et_realisation_dun_site_Web_d.pdfConception_et_realisation_dun_site_Web_d.pdf
Conception_et_realisation_dun_site_Web_d.pdf
 
Miniproject final group 14
Miniproject final group 14Miniproject final group 14
Miniproject final group 14
 
Unsupervised Building Extraction from High Resolution Satellite Images Irresp...
Unsupervised Building Extraction from High Resolution Satellite Images Irresp...Unsupervised Building Extraction from High Resolution Satellite Images Irresp...
Unsupervised Building Extraction from High Resolution Satellite Images Irresp...
 
Lw3620362041
Lw3620362041Lw3620362041
Lw3620362041
 
Currency recognition on mobile phones
Currency recognition on mobile phonesCurrency recognition on mobile phones
Currency recognition on mobile phones
 
Fisheye Omnidirectional View in Autonomous Driving
Fisheye Omnidirectional View in Autonomous DrivingFisheye Omnidirectional View in Autonomous Driving
Fisheye Omnidirectional View in Autonomous Driving
 
Introduction to Binocular Stereo in Computer Vision
Introduction to Binocular Stereo in Computer VisionIntroduction to Binocular Stereo in Computer Vision
Introduction to Binocular Stereo in Computer Vision
 
Normal Mapping / Computer Graphics - IK
Normal Mapping / Computer Graphics - IKNormal Mapping / Computer Graphics - IK
Normal Mapping / Computer Graphics - IK
 
Urban 3D Semantic Modelling Using Stereo Vision, ICRA 2013
Urban 3D Semantic Modelling Using Stereo Vision, ICRA 2013Urban 3D Semantic Modelling Using Stereo Vision, ICRA 2013
Urban 3D Semantic Modelling Using Stereo Vision, ICRA 2013
 
IEEE ICAPR 2009
IEEE ICAPR 2009IEEE ICAPR 2009
IEEE ICAPR 2009
 
Automatic Detection of Window Regions in Indoor Point Clouds Using R-CNN
Automatic Detection of Window Regions in Indoor Point Clouds Using R-CNNAutomatic Detection of Window Regions in Indoor Point Clouds Using R-CNN
Automatic Detection of Window Regions in Indoor Point Clouds Using R-CNN
 
A Survey on Exemplar-Based Image Inpainting Techniques
A Survey on Exemplar-Based Image Inpainting TechniquesA Survey on Exemplar-Based Image Inpainting Techniques
A Survey on Exemplar-Based Image Inpainting Techniques
 
Video Stitching using Improved RANSAC and SIFT
Video Stitching using Improved RANSAC and SIFTVideo Stitching using Improved RANSAC and SIFT
Video Stitching using Improved RANSAC and SIFT
 
Design and implementation of video tracking system based on camera field of view
Design and implementation of video tracking system based on camera field of viewDesign and implementation of video tracking system based on camera field of view
Design and implementation of video tracking system based on camera field of view
 
Super Resolution of Image
Super Resolution of ImageSuper Resolution of Image
Super Resolution of Image
 
Remotely sensed image segmentation using multiphase level set acm
Remotely sensed image segmentation using multiphase level set acmRemotely sensed image segmentation using multiphase level set acm
Remotely sensed image segmentation using multiphase level set acm
 
A STUDY AND ANALYSIS OF DIFFERENT EDGE DETECTION TECHNIQUES
A STUDY AND ANALYSIS OF DIFFERENT EDGE DETECTION TECHNIQUESA STUDY AND ANALYSIS OF DIFFERENT EDGE DETECTION TECHNIQUES
A STUDY AND ANALYSIS OF DIFFERENT EDGE DETECTION TECHNIQUES
 
Av4301248253
Av4301248253Av4301248253
Av4301248253
 

More from Vicente Ordonez

From Large Scale Image Categorization to Entry-Level Categories
From Large Scale Image Categorization to Entry-Level CategoriesFrom Large Scale Image Categorization to Entry-Level Categories
From Large Scale Image Categorization to Entry-Level Categories
Vicente Ordonez
 
Data-driven Generation of Image Descriptions
Data-driven Generation of Image DescriptionsData-driven Generation of Image Descriptions
Data-driven Generation of Image Descriptions
Vicente Ordonez
 
Google Earth Maps Api Barcamp Quito 2009
Google Earth Maps Api Barcamp Quito 2009Google Earth Maps Api Barcamp Quito 2009
Google Earth Maps Api Barcamp Quito 2009
Vicente Ordonez
 
Sistema de Recuperacion de Audio
Sistema de Recuperacion de AudioSistema de Recuperacion de Audio
Sistema de Recuperacion de Audio
Vicente Ordonez
 
Transmision de Vídeo por Red / Internet
Transmision de Vídeo por Red / InternetTransmision de Vídeo por Red / Internet
Transmision de Vídeo por Red / Internet
Vicente Ordonez
 
Portal Concepts and .NET Webparts
Portal Concepts and .NET WebpartsPortal Concepts and .NET Webparts
Portal Concepts and .NET Webparts
Vicente Ordonez
 

More from Vicente Ordonez (16)

From Large Scale Image Categorization to Entry-Level Categories
From Large Scale Image Categorization to Entry-Level CategoriesFrom Large Scale Image Categorization to Entry-Level Categories
From Large Scale Image Categorization to Entry-Level Categories
 
Data-driven Generation of Image Descriptions
Data-driven Generation of Image DescriptionsData-driven Generation of Image Descriptions
Data-driven Generation of Image Descriptions
 
Im2Text: Describing Images Using 1 Million Captioned Photographs
Im2Text: Describing Images Using 1 Million Captioned PhotographsIm2Text: Describing Images Using 1 Million Captioned Photographs
Im2Text: Describing Images Using 1 Million Captioned Photographs
 
Texture Synthesis
Texture SynthesisTexture Synthesis
Texture Synthesis
 
Contenido Generado Por Los Usuarios
Contenido Generado Por Los UsuariosContenido Generado Por Los Usuarios
Contenido Generado Por Los Usuarios
 
Pantallas Plasma vs LCD
Pantallas Plasma vs LCDPantallas Plasma vs LCD
Pantallas Plasma vs LCD
 
Google Earth Maps Api Barcamp Quito 2009
Google Earth Maps Api Barcamp Quito 2009Google Earth Maps Api Barcamp Quito 2009
Google Earth Maps Api Barcamp Quito 2009
 
Sistema de Recuperacion de Audio
Sistema de Recuperacion de AudioSistema de Recuperacion de Audio
Sistema de Recuperacion de Audio
 
Suenaemprendevive
SuenaemprendeviveSuenaemprendevive
Suenaemprendevive
 
MapReduce
MapReduceMapReduce
MapReduce
 
Robotica
RoboticaRobotica
Robotica
 
Transmision de Vídeo por Red / Internet
Transmision de Vídeo por Red / InternetTransmision de Vídeo por Red / Internet
Transmision de Vídeo por Red / Internet
 
Buscadores de Podcast en Internet
Buscadores de Podcast en InternetBuscadores de Podcast en Internet
Buscadores de Podcast en Internet
 
Sistemas Operativos 3D
Sistemas Operativos 3DSistemas Operativos 3D
Sistemas Operativos 3D
 
Ajax Atlas
Ajax AtlasAjax Atlas
Ajax Atlas
 
Portal Concepts and .NET Webparts
Portal Concepts and .NET WebpartsPortal Concepts and .NET Webparts
Portal Concepts and .NET Webparts
 

Recently uploaded

+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
?#DUbAI#??##{{(☎️+971_581248768%)**%*]'#abortion pills for sale in dubai@
 
Why Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire businessWhy Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire business
panagenda
 

Recently uploaded (20)

Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
Navigating the Deluge_ Dubai Floods and the Resilience of Dubai International...
 
AXA XL - Insurer Innovation Award Americas 2024
AXA XL - Insurer Innovation Award Americas 2024AXA XL - Insurer Innovation Award Americas 2024
AXA XL - Insurer Innovation Award Americas 2024
 
MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024
 
Ransomware_Q4_2023. The report. [EN].pdf
Ransomware_Q4_2023. The report. [EN].pdfRansomware_Q4_2023. The report. [EN].pdf
Ransomware_Q4_2023. The report. [EN].pdf
 
ICT role in 21st century education and its challenges
ICT role in 21st century education and its challengesICT role in 21st century education and its challenges
ICT role in 21st century education and its challenges
 
[BuildWithAI] Introduction to Gemini.pdf
[BuildWithAI] Introduction to Gemini.pdf[BuildWithAI] Introduction to Gemini.pdf
[BuildWithAI] Introduction to Gemini.pdf
 
CNIC Information System with Pakdata Cf In Pakistan
CNIC Information System with Pakdata Cf In PakistanCNIC Information System with Pakdata Cf In Pakistan
CNIC Information System with Pakdata Cf In Pakistan
 
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
 
DBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor PresentationDBX First Quarter 2024 Investor Presentation
DBX First Quarter 2024 Investor Presentation
 
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
 
Cyberprint. Dark Pink Apt Group [EN].pdf
Cyberprint. Dark Pink Apt Group [EN].pdfCyberprint. Dark Pink Apt Group [EN].pdf
Cyberprint. Dark Pink Apt Group [EN].pdf
 
Why Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire businessWhy Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire business
 
AWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of Terraform
 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
 
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWEREMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
 
Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processors
 
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
 
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
 
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ..."I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...
 

Visual Saliency: Learning to Detect Salient Objects

  • 1. Visual Attention: Detecting Saliency on Images Vicente Ordonez Department of Computer Science State University of New York Stony Brook, NY 11790
  • 2. I will be working mainly on the following paper Learning to Detect a Salient Object. T. Liu, J. Sun, N. Zheng, X. Tang, H. Shum. (Xian Jiaotong University and Microsoft Research Asia) from CVPR 2007. http://research.microsoft.com/en-us/um/people/jiansun/papers/SalientDetection_CVPR07.pdf
  • 3. What is Saliency? What is Visual Attention? “Everyone knows what attention is...” —William James, 1890
  • 4. This is a problem of… Arbitrary object detection? Background / Foreground segmentation? Modeling Visual Attention?
  • 5. The Method Features: Multiscale Contrast (Done!) Center surround histogram (Mostly Done!) (Done!) Color spatial distribution (Done!) Supervised learning using Conditional Random Fields to determine the parameters to combine the features obtained above. (Done!) [I will use a labeled dataset of 5000 images provided by Microsoft Research Asia!]
  • 6. Multiscale Contrast Function Generate the Gaussian Pyramid for the input image. For each level in the pyramid Do gaussian blurring Do resampling I’m using a 6 levels Gaussian pyramid for each RGB channel.
  • 7. How a Gaussian pyramid looks like Figure from David Forsyth
  • 8. Generate contrast maps for each level of the Pyramid. Sum all of the results to produce the final multiscale contrast map. The two steps mentioned above are described in this formula: Multiscale Contrast Function
  • 11. Contrast maps Original image Contrast map at level 1 Contrast map at level 4 Contrast map at level 6
  • 13.
  • 14. For each possible rectangle with a reasonable size and aspect ratio
  • 15. Create a surrounding rectangle and calculate the histogram of the rectangle and the surrounding area.
  • 16.
  • 17. Center Surround Histogram Feature The algorithm as described before is computationally expensive… It is required to use a technique called Integral Histogram. It allows you fast calculation of the histogram of any given rectangular region of an image. The algorithm was introduced in: “Integral Histogram: A Fast Way to Extract Histograms in Cartesian Spaces” by FatihPorikli, Mitsubishi Electric Research Lab in CVPR 2005.
  • 18. Center Surround Histogram Feature Use the Chi Square Distances Map and the Map of Most Salient Rectangle Regions per pixel to generate the Center Surround Histogram Feature using the next formula:
  • 19. Center Surround Histogram Results Using my Implementation (15.2 sec, size = 245x384) Results Reported in the Paper
  • 20. Center Surround Histogram Results Using my Implementation (13.6 sec, size = 247x346) Results Reported in the Paper
  • 21. Center Surround Histogram Results Using my Implementation (10.2 sec, size = 248x277)
  • 34. Color Spatial Distribution Make an initial clustering of the colors in the image using k-means. Further refine the clusters by using Gaussian Mixture Models. The Gaussian Mixture Model parameters are calculated using the EM algorithm. I am using 5 clusters (5 colors) per image. And the results look similar to those presented in the paper with an execution time of around 17 seconds per image.
  • 35. Color Spatial Distribution Calculate the vertical variance of the horizontal positions of the pixels for each cluster. And then the same for the vertical positions. Sum the variances and use this value to weight more those clusters with less spatial variance. Penalize the clusters that contain the majority of its pixels away from the center of the image.
  • 45. Conditional Random Field Training and Inference Accelerated Training of Conditional Random Fields with Stochastic Meta-Descent S Vishwanathan, N. Schraudolph, M. Schmidt, K. Murphy. ICML'06 (Intl Conf on Machine Learning).  I did the training using this toolbox from the above paper: http://people.cs.ubc.ca/~murphyk/Software/CRF/crf.html
  • 46. Mask outputs using CRF inference Input M-Contrast-map Center Surr. Hist. Color Spatial Var. Input Combined features Ground truth
  • 47. Mask outputs using CRF inference Input M-Contrast-map Center Surr. Hist. Color Spatial Var. Input Combined features Ground truth
  • 48. Mask outputs using CRF inference Input M-Contrast-map Center Surr. Hist. Color Spatial Var. Input Combined features Ground truth
  • 49. Mask outputs using CRF inference Input M-Contrast-map Center Surr. Hist. Color Spatial Var. Input Combined features Ground truth
  • 50. Precision / Recall obtained
  • 51. Some Conclusions The results of the original research paper on computing the visual features have been successfully replicated in a considerable extent. The Conditional Random Field framework used in this project turned out to perform well for this task. The center-surround histogram map turned out to be the feature that gave the higher precision. The amount of time required for computing the individual features is in the order of several seconds.

Editor's Notes

  1. Not so good result
  2. Good result
  3. Not so good result