SlideShare a Scribd company logo
Learning visual similarity using classifiers Lior Wolf, The Blavatnik School of Computer Science, Tel-Aviv University Collaborators:                                   Students: YanivTaigman    Tal Hassner       Orit Klipper-Gross    Itay Maoz   face.com               Open U               Weizmann inst.         Tel-Aviv U  1
The Blavatnik School of Computer ScienceTel-Aviv University An example of higher education in Israel 2
A school in the Faculty of Exact Sciences that also includes: Mathematics, Physics, Chemistry, Geophysics and Planetary Sciences Originated in the 1970’s as part of the School of Math,  since 2000 a separate School 39 Faculty Members ~1000 undergrads ~200 MSc students ~70 PhD students Post Docs and other research personnel 3
School Ranking in the world TAU/CS Ranked #29 in number of citations - Thompson Scientific, (for the years 2000-2010).[Technion #33 , Weizmann #72, HebrewU #105] TAU/CS Ranked #28 by the Shanghai Academic Ranking of World Universities in Computer Science – 2011[Weizmann #12, Technion #15, HebrewU #21] TAU/CS Ranked #14 in the world in CS impact – Scientometrics, Vol. 76, No. 2, 2008. 12 TAU/CS faculty in positions 1-100 in “list of most central computer Scientists in Theory of Computer Science” (Kuhn – Wattenhofer, Sigact news, Dec ’07) 4
Computer vision in search Query Raw data: images,video, audio Information:objects,tags, IDs,context Searchresults Preprocessing 5
The pain: too many images On                 : Over 1,000,000,000 photos uploaded each month shared by 200,000,000+ users 10’s of billions served/week No tags  No Photos… “can I see all my photos?” “tagging takes hours, can you do that for me?” 6
The evolution of perceptual search Reranking bysimilarity Text-basedimage search Specializationin face identification Low-level vision Mid-level vision No vision With basicproperties Gist-based Image similarity Catalog basedsearch High-level vision: scene understanding 7
Photo Finder for facebook 8
THE 1st MOBILE APP TO FIND 3D ITEMS 9
WHAT MAKES IT SO HARD? 10
High-level vision: what is where? A happy couple walks in a fieldWhat kind of field? Where? Which season? How old are they? Gender? How attractive? What are they wearing? High-level vision: scene understanding 11
YaC, Moscow, September 19, 2011 Learning visual similarity using classifiers Lior Wolf, The Blavatnik School of Computer Science, Tel-Aviv University Collaborators:                                   Students: YanivTaigman    Tal Hassner       Orit Klipper-Gross    Itay Maoz   face.com               Open U               Weizmann inst.         Tel-Aviv U  12
The Pair-Matching Problem Training: 13
Modeling never before seen objects Natural setup for image retrieval with no categories Training: The Pair-Matching Problem 14
Instances Document Analysis Video Action Recogniti0n Face Recogniti0n Video Face Recogniti0n 15
The Pair-Matching Problem Training: 16
Labeled Faces in the Wild (LFW) 13,000 labeled images of faces collected from the web 5,749 individuals 1-150 images per individual Training: 17
Restricted Protocol 10-fold cross validation tests on randomly generated splits, each with: 300 same pairs 300 not same pairs 18
Pipeline (take 1)* Training. Note: no use of labels! =1 Sim (     ,    ) same =2 Sim(     ,    ) Classifier (e.g.SVM)  =i Sim (     ,    ) not same =i+1 Sim(     ,    ) Threshold * “Descriptor Based Methods in the Wild,” ECCVw’08 19
Pipeline (take 1)* Training – multiple descriptors similarities  (1,1,1,2,…,1,n) same (2,1,2,2,…,2,n) Classifier (e.g.SVM)  (i,1,i,2,…,i,n) not same (i+1,1,i+1,2,…,i+1,n) * “Descriptor Based Methods in the Wild,” ECCVw’08 20
Some Questions How to represent the images? Which similarity to use? Grayscales, Edge responces [Brunelli & Poggio’93], C1-Gabor [e.g., Riesenhuber & Poggio’99], SIFT [Lowe’04], LBP [e.g., Ojala & Pietikainen & Harwood’96],… Later on: How can subject IDs help improve pair-matching performance? L2, Correlation, Learned metrics [e.g., Bilenko etal.’04, Cristianini etal.’02, Hertz etal. 04, …], “hand-crafted” metrics [e.g., Belongie etal.’01] 21
One-Shot Similarity (OSS) Score* What: A measure of the similarity between two vectors Input: The two vectors A set of “Background samples” How: Use “One-Shot Learning” (classification with one positive example) * “Descriptor Based Methods in the Wild,” ECCVw’08   “The One-Shot Similarity Kernel”, ICCV’09 22
Computing the “One-Shot” Similarity Step a: Model1 = train (p, A) Step b: Score1 = classify(q, Model1) Step c: Model2 = train (q, A) Set “A” of background examples Step d: Score2 = classify(p, Model1) q One-Shot-Sim = (score1 + score2) /2 Similarity  p 23
Euclidean vs. One-Shot Visualized One-Shot Euclidean 24
Euclidean vs. One-Shot Visualized One-Shot Euclidean 25
Computing the “One-Shot” Similarity Using LDA as the underlying classifier : Where        is the mean of set A, and         is the pseudo-inverse of the intra-class cov. matrix.  * “The One-Shot Similarity Kernel”, ICCV’09 26
Computing the “One-Shot” Similarity Using Free-Scale LDA as the underlying classifier : Where        is the mean of set A, and         is the pseudo-inverse of the intra-class cov. matrix.  * “The One-Shot Similarity Kernel”, ICCV’09 27
Some Properties of the OSS* Uses unlabeled training data OSS based on Free-Scale LDA is a CPD Kernel May be efficiently computed Complexity:           is independent of the two vectors compared, and so computed only once. Also, repeated comparisons of a vector xi to different  xj may be performed in O(n) * “The One-Shot Similarity Kernel”, ICCV’09 28
Some Properties of the OSS* * “The One-Shot Similarity Kernel”, ICCV’09 29
Some Properties of the OSS* OSS based on Free-Scale LDA is a CPD Kernel * “The One-Shot Similarity Kernel”, ICCV’09 30
Metric learning for OSS* Instead of examples xi use Txi for some “optimal” T The transformation T is obtained by a gradient decent procedure that optimizes the score: *“One Shot Similarity Metric Learning for Action Recognition”, In submission. 31
The Unrestricted Protocol 10-fold cross validation tests on randomly generated splits, each with: 300 same pairs 300 not same pairs Training now includes subject labels 32
Multiple One-Shots* We now have IDs. How do we use them? Compute multiple OSS, each time using examples from a single class  * “Multiple One-Shots for Utilizing Class Label Information,” BMVC’09 33
Multiple One-Shots ID-based OSS 34
Multiple One-Shots We now have IDs. How do we use them? Compute multiple OSS, each time using examples from a single class  Discrimination based on different sources of variation: Subject ID, Pose, etc. 35
The Pose Issue Most confident wrong results* * “Descriptor Based Methods in the Wild,” ECCVw’08 36
Getting Poses To compute Pose based OSS, you need sets of images in the same pose… 7 fiducial points (eyes, mouth, nose) 14 x,y coordinates 14D vector of alignment errors (similarity trnsf.) Project to first Principal Component Bin to 10 classes 37
Multiple One-Shots Pose-based OSS 38
Multiple One-Shots - Examples  5 Id-based OSS and 5 Pose-based OSS scores Identity Pose 39
Multiple One-Shots - Examples  Identity Pose 40
Multiple One-Shots - Examples  Identity Pose 41
Pipeline* Input image pair Image alignment Commercial alignment  software by  * “Multiple One-Shots for Utilizing Class Label Information,” BMVC’09 42
Pipeline* Input image pair Image alignment Feature vectors Using: ,[object Object]
LBP [Ojala etal.’96, 01,02]
TPLBP, FPLBP [Wolf etal.’08]* “Multiple One-Shots for Utilizing Class Label Information,” BMVC’09 43
Pipeline* Input image pair Image alignment Feature vectors PCA+ITML Information Theoretic Metric Learning [Davis etal.’07] * “Multiple One-Shots for Utilizing Class Label Information,” BMVC’09 44
Pipeline* Input image pair Image alignment Feature vectors Multiple OSS scores PCA+ITML 20 Subjects 10 Poses  * “Multiple One-Shots for Utilizing Class Label Information,” BMVC’09 45
Pipeline* Input image pair Image alignment Feature vectors Multiple OSS scores SVM classifier Output PCA+ITML Same Not-same * “Multiple One-Shots for Utilizing Class Label Information,” BMVC’09 46
Pipeline – Multiple Descriptors* Feature vectors SIFT Multiple OSS scores PCA+ITML Image alignment SVM classifier Output Same Not-same Feature vectors LBP Multiple OSS scores PCA+ITML * “Multiple One-Shots for Utilizing Class Label Information,” BMVC’09 47
0.7847 ± 0.0051     [WHT’08] 0.8398 ± 0.0035  [WHT’08 + alignment] Results 0.8517 ± 0.0061   [this work, only LBP] 0.8950 ± 0.0051   [this work, multi-desc.] 0.9753  [Kumar etal. 09 - HUMAN] 48
Pair-Matching of Sets * Face Recognition in Unconstrained Videos with Matched B/G Similarity. CVPR 2011. 49
Pair-Matching of Sets Training: 50
Conventional methods ,[object Object]
pose based methods, comparing the two most frontal faces in each video or the two faces with the most similar pose.

More Related Content

What's hot

“Practical Guide to Implementing Deep Neural Network Inferencing at the Edge,...
“Practical Guide to Implementing Deep Neural Network Inferencing at the Edge,...“Practical Guide to Implementing Deep Neural Network Inferencing at the Edge,...
“Practical Guide to Implementing Deep Neural Network Inferencing at the Edge,...
Edge AI and Vision Alliance
 
Qualcomm research-imagenet2015
Qualcomm research-imagenet2015Qualcomm research-imagenet2015
Qualcomm research-imagenet2015
Bilkent University
 
Visual Object Tracking: review
Visual Object Tracking: reviewVisual Object Tracking: review
Visual Object Tracking: review
Dmytro Mishkin
 
Battista Biggio @ ECML PKDD 2013 - Evasion attacks against machine learning a...
Battista Biggio @ ECML PKDD 2013 - Evasion attacks against machine learning a...Battista Biggio @ ECML PKDD 2013 - Evasion attacks against machine learning a...
Battista Biggio @ ECML PKDD 2013 - Evasion attacks against machine learning a...Pluribus One
 
Battista Biggio @ ICML 2015 - "Is Feature Selection Secure against Training D...
Battista Biggio @ ICML 2015 - "Is Feature Selection Secure against Training D...Battista Biggio @ ICML 2015 - "Is Feature Selection Secure against Training D...
Battista Biggio @ ICML 2015 - "Is Feature Selection Secure against Training D...
Pluribus One
 
SSII2021 [OS2-03] 自己教師あり学習における対照学習の基礎と応用
SSII2021 [OS2-03] 自己教師あり学習における対照学習の基礎と応用SSII2021 [OS2-03] 自己教師あり学習における対照学習の基礎と応用
SSII2021 [OS2-03] 自己教師あり学習における対照学習の基礎と応用
SSII
 
MLIP - Chapter 6 - Generation, Super-Resolution, Style transfer
MLIP - Chapter 6 - Generation, Super-Resolution, Style transferMLIP - Chapter 6 - Generation, Super-Resolution, Style transfer
MLIP - Chapter 6 - Generation, Super-Resolution, Style transfer
Charles Deledalle
 
Final Presentation (REVISION 2)
Final Presentation (REVISION 2)Final Presentation (REVISION 2)
Final Presentation (REVISION 2)Chad Buckallew
 

What's hot (9)

“Practical Guide to Implementing Deep Neural Network Inferencing at the Edge,...
“Practical Guide to Implementing Deep Neural Network Inferencing at the Edge,...“Practical Guide to Implementing Deep Neural Network Inferencing at the Edge,...
“Practical Guide to Implementing Deep Neural Network Inferencing at the Edge,...
 
Presentation of Visual Tracking
Presentation of Visual TrackingPresentation of Visual Tracking
Presentation of Visual Tracking
 
Qualcomm research-imagenet2015
Qualcomm research-imagenet2015Qualcomm research-imagenet2015
Qualcomm research-imagenet2015
 
Visual Object Tracking: review
Visual Object Tracking: reviewVisual Object Tracking: review
Visual Object Tracking: review
 
Battista Biggio @ ECML PKDD 2013 - Evasion attacks against machine learning a...
Battista Biggio @ ECML PKDD 2013 - Evasion attacks against machine learning a...Battista Biggio @ ECML PKDD 2013 - Evasion attacks against machine learning a...
Battista Biggio @ ECML PKDD 2013 - Evasion attacks against machine learning a...
 
Battista Biggio @ ICML 2015 - "Is Feature Selection Secure against Training D...
Battista Biggio @ ICML 2015 - "Is Feature Selection Secure against Training D...Battista Biggio @ ICML 2015 - "Is Feature Selection Secure against Training D...
Battista Biggio @ ICML 2015 - "Is Feature Selection Secure against Training D...
 
SSII2021 [OS2-03] 自己教師あり学習における対照学習の基礎と応用
SSII2021 [OS2-03] 自己教師あり学習における対照学習の基礎と応用SSII2021 [OS2-03] 自己教師あり学習における対照学習の基礎と応用
SSII2021 [OS2-03] 自己教師あり学習における対照学習の基礎と応用
 
MLIP - Chapter 6 - Generation, Super-Resolution, Style transfer
MLIP - Chapter 6 - Generation, Super-Resolution, Style transferMLIP - Chapter 6 - Generation, Super-Resolution, Style transfer
MLIP - Chapter 6 - Generation, Super-Resolution, Style transfer
 
Final Presentation (REVISION 2)
Final Presentation (REVISION 2)Final Presentation (REVISION 2)
Final Presentation (REVISION 2)
 

Viewers also liked

C++11 (formerly known as C++0x) is the new C++ language standard. Dave Abraha...
C++11 (formerly known as C++0x) is the new C++ language standard. Dave Abraha...C++11 (formerly known as C++0x) is the new C++ language standard. Dave Abraha...
C++11 (formerly known as C++0x) is the new C++ language standard. Dave Abraha...
yaevents
 
Similarities between the crisis of 1930 and 2015 in brazil
Similarities between the crisis of 1930 and 2015 in brazilSimilarities between the crisis of 1930 and 2015 in brazil
Similarities between the crisis of 1930 and 2015 in brazil
Fernando Alcoforado
 
Контроль зверей: инструменты для управления и мониторинга распределенных сист...
Контроль зверей: инструменты для управления и мониторинга распределенных сист...Контроль зверей: инструменты для управления и мониторинга распределенных сист...
Контроль зверей: инструменты для управления и мониторинга распределенных сист...
yaevents
 
Масштабируемость Hadoop в Facebook. Дмитрий Мольков, Facebook
Масштабируемость Hadoop в Facebook. Дмитрий Мольков, FacebookМасштабируемость Hadoop в Facebook. Дмитрий Мольков, Facebook
Масштабируемость Hadoop в Facebook. Дмитрий Мольков, Facebook
yaevents
 
Поисковая технология "Спектр". Андрей Плахов, Яндекс
Поисковая технология "Спектр". Андрей Плахов, ЯндексПоисковая технология "Спектр". Андрей Плахов, Яндекс
Поисковая технология "Спектр". Андрей Плахов, Яндекс
yaevents
 
В поисках математики. Михаил Денисенко, Нигма
В поисках математики. Михаил Денисенко, НигмаВ поисках математики. Михаил Денисенко, Нигма
В поисках математики. Михаил Денисенко, Нигма
yaevents
 
Сложнейшие техники, применяемые буткитами и полиморфными вирусами. Вячеслав З...
Сложнейшие техники, применяемые буткитами и полиморфными вирусами. Вячеслав З...Сложнейшие техники, применяемые буткитами и полиморфными вирусами. Вячеслав З...
Сложнейшие техники, применяемые буткитами и полиморфными вирусами. Вячеслав З...
yaevents
 
Зачем обычному программисту знать языки, на которых почти никто не пишет. Але...
Зачем обычному программисту знать языки, на которых почти никто не пишет. Але...Зачем обычному программисту знать языки, на которых почти никто не пишет. Але...
Зачем обычному программисту знать языки, на которых почти никто не пишет. Але...
yaevents
 
Characteristics of Growth
Characteristics of GrowthCharacteristics of Growth
Characteristics of Growth
Angelika Solinap
 
Юнит-тестирование и Google Mock. Влад Лосев, Google
Юнит-тестирование и Google Mock. Влад Лосев, GoogleЮнит-тестирование и Google Mock. Влад Лосев, Google
Юнит-тестирование и Google Mock. Влад Лосев, Google
yaevents
 
Сканирование уязвимостей со вкусом Яндекса. Тарас Иващенко, Яндекс
Сканирование уязвимостей со вкусом Яндекса. Тарас Иващенко, ЯндексСканирование уязвимостей со вкусом Яндекса. Тарас Иващенко, Яндекс
Сканирование уязвимостей со вкусом Яндекса. Тарас Иващенко, Яндекс
yaevents
 
Philosophy ppt
Philosophy ppt Philosophy ppt
Philosophy ppt
Megha Mohan
 
Human Growth & Development: Developmental Psychology. By Theresa Lowry-Lehnen...
Human Growth & Development: Developmental Psychology. By Theresa Lowry-Lehnen...Human Growth & Development: Developmental Psychology. By Theresa Lowry-Lehnen...
Human Growth & Development: Developmental Psychology. By Theresa Lowry-Lehnen...
Theresa Lowry-Lehnen
 
Growth & development presentation
Growth & development presentationGrowth & development presentation
Growth & development presentation
Hazel Garin
 
Growth and development
Growth and developmentGrowth and development
Growth and development
Aruna Ap
 
Growth and development..ppt
Growth and development..pptGrowth and development..ppt
Growth and development..pptRahul Dhaker
 

Viewers also liked (16)

C++11 (formerly known as C++0x) is the new C++ language standard. Dave Abraha...
C++11 (formerly known as C++0x) is the new C++ language standard. Dave Abraha...C++11 (formerly known as C++0x) is the new C++ language standard. Dave Abraha...
C++11 (formerly known as C++0x) is the new C++ language standard. Dave Abraha...
 
Similarities between the crisis of 1930 and 2015 in brazil
Similarities between the crisis of 1930 and 2015 in brazilSimilarities between the crisis of 1930 and 2015 in brazil
Similarities between the crisis of 1930 and 2015 in brazil
 
Контроль зверей: инструменты для управления и мониторинга распределенных сист...
Контроль зверей: инструменты для управления и мониторинга распределенных сист...Контроль зверей: инструменты для управления и мониторинга распределенных сист...
Контроль зверей: инструменты для управления и мониторинга распределенных сист...
 
Масштабируемость Hadoop в Facebook. Дмитрий Мольков, Facebook
Масштабируемость Hadoop в Facebook. Дмитрий Мольков, FacebookМасштабируемость Hadoop в Facebook. Дмитрий Мольков, Facebook
Масштабируемость Hadoop в Facebook. Дмитрий Мольков, Facebook
 
Поисковая технология "Спектр". Андрей Плахов, Яндекс
Поисковая технология "Спектр". Андрей Плахов, ЯндексПоисковая технология "Спектр". Андрей Плахов, Яндекс
Поисковая технология "Спектр". Андрей Плахов, Яндекс
 
В поисках математики. Михаил Денисенко, Нигма
В поисках математики. Михаил Денисенко, НигмаВ поисках математики. Михаил Денисенко, Нигма
В поисках математики. Михаил Денисенко, Нигма
 
Сложнейшие техники, применяемые буткитами и полиморфными вирусами. Вячеслав З...
Сложнейшие техники, применяемые буткитами и полиморфными вирусами. Вячеслав З...Сложнейшие техники, применяемые буткитами и полиморфными вирусами. Вячеслав З...
Сложнейшие техники, применяемые буткитами и полиморфными вирусами. Вячеслав З...
 
Зачем обычному программисту знать языки, на которых почти никто не пишет. Але...
Зачем обычному программисту знать языки, на которых почти никто не пишет. Але...Зачем обычному программисту знать языки, на которых почти никто не пишет. Але...
Зачем обычному программисту знать языки, на которых почти никто не пишет. Але...
 
Characteristics of Growth
Characteristics of GrowthCharacteristics of Growth
Characteristics of Growth
 
Юнит-тестирование и Google Mock. Влад Лосев, Google
Юнит-тестирование и Google Mock. Влад Лосев, GoogleЮнит-тестирование и Google Mock. Влад Лосев, Google
Юнит-тестирование и Google Mock. Влад Лосев, Google
 
Сканирование уязвимостей со вкусом Яндекса. Тарас Иващенко, Яндекс
Сканирование уязвимостей со вкусом Яндекса. Тарас Иващенко, ЯндексСканирование уязвимостей со вкусом Яндекса. Тарас Иващенко, Яндекс
Сканирование уязвимостей со вкусом Яндекса. Тарас Иващенко, Яндекс
 
Philosophy ppt
Philosophy ppt Philosophy ppt
Philosophy ppt
 
Human Growth & Development: Developmental Psychology. By Theresa Lowry-Lehnen...
Human Growth & Development: Developmental Psychology. By Theresa Lowry-Lehnen...Human Growth & Development: Developmental Psychology. By Theresa Lowry-Lehnen...
Human Growth & Development: Developmental Psychology. By Theresa Lowry-Lehnen...
 
Growth & development presentation
Growth & development presentationGrowth & development presentation
Growth & development presentation
 
Growth and development
Growth and developmentGrowth and development
Growth and development
 
Growth and development..ppt
Growth and development..pptGrowth and development..ppt
Growth and development..ppt
 

Similar to Using classifiers to compute similarities between face images. Prof. Lior Wolf, Tel-Aviv University

Computational Biology, Part 4 Protein Coding Regions
Computational Biology, Part 4 Protein Coding RegionsComputational Biology, Part 4 Protein Coding Regions
Computational Biology, Part 4 Protein Coding Regionsbutest
 
Automatic Test Suite Generation for Key-Points Detection DNNs using Many-Obje...
Automatic Test Suite Generation for Key-Points Detection DNNs using Many-Obje...Automatic Test Suite Generation for Key-Points Detection DNNs using Many-Obje...
Automatic Test Suite Generation for Key-Points Detection DNNs using Many-Obje...
Lionel Briand
 
Deep Learning: concepts and use cases (October 2018)
Deep Learning: concepts and use cases (October 2018)Deep Learning: concepts and use cases (October 2018)
Deep Learning: concepts and use cases (October 2018)
Julien SIMON
 
Machine Learning ebook.pdf
Machine Learning ebook.pdfMachine Learning ebook.pdf
Machine Learning ebook.pdf
HODIT12
 
1_5_AI_edx_ml_51intro_240204_104838machine learning lecture 1
1_5_AI_edx_ml_51intro_240204_104838machine learning lecture 11_5_AI_edx_ml_51intro_240204_104838machine learning lecture 1
1_5_AI_edx_ml_51intro_240204_104838machine learning lecture 1
MostafaHazemMostafaa
 
Yulia Honcharenko "Application of metric learning for logo recognition"
Yulia Honcharenko "Application of metric learning for logo recognition"Yulia Honcharenko "Application of metric learning for logo recognition"
Yulia Honcharenko "Application of metric learning for logo recognition"
Fwdays
 
one shot15729752 Deep Learning for AI and DS
one shot15729752 Deep Learning for AI and DSone shot15729752 Deep Learning for AI and DS
one shot15729752 Deep Learning for AI and DS
ManiMaran230751
 
Crystallization classification semisupervised
Crystallization classification semisupervisedCrystallization classification semisupervised
Crystallization classification semisupervised
Madhav Sigdel
 
know Machine Learning Basic Concepts.pdf
know Machine Learning Basic Concepts.pdfknow Machine Learning Basic Concepts.pdf
know Machine Learning Basic Concepts.pdf
hemangppatel
 
HiPEAC2022_António Casimiro presentation
HiPEAC2022_António Casimiro presentationHiPEAC2022_António Casimiro presentation
HiPEAC2022_António Casimiro presentation
VEDLIoT Project
 
Learning where to look: focus and attention in deep vision
Learning where to look: focus and attention in deep visionLearning where to look: focus and attention in deep vision
Learning where to look: focus and attention in deep vision
Universitat Politècnica de Catalunya
 
Long-term Face Tracking in the Wild using Deep Learning
Long-term Face Tracking in the Wild using Deep LearningLong-term Face Tracking in the Wild using Deep Learning
Long-term Face Tracking in the Wild using Deep Learning
Elaheh Rashedi
 
Analytical study of feature extraction techniques in opinion mining
Analytical study of feature extraction techniques in opinion miningAnalytical study of feature extraction techniques in opinion mining
Analytical study of feature extraction techniques in opinion mining
csandit
 
ANALYTICAL STUDY OF FEATURE EXTRACTION TECHNIQUES IN OPINION MINING
ANALYTICAL STUDY OF FEATURE EXTRACTION TECHNIQUES IN OPINION MININGANALYTICAL STUDY OF FEATURE EXTRACTION TECHNIQUES IN OPINION MINING
ANALYTICAL STUDY OF FEATURE EXTRACTION TECHNIQUES IN OPINION MINING
csandit
 
Radial Basis Function Neural Network (RBFNN), Induction Motor, Vector control...
Radial Basis Function Neural Network (RBFNN), Induction Motor, Vector control...Radial Basis Function Neural Network (RBFNN), Induction Motor, Vector control...
Radial Basis Function Neural Network (RBFNN), Induction Motor, Vector control...
cscpconf
 
Automated Face Detection System
Automated Face Detection SystemAutomated Face Detection System
Automated Face Detection System
Abhiroop Ghatak
 
Ensemble Learning Featuring the Netflix Prize Competition and ...
Ensemble Learning Featuring the Netflix Prize Competition and ...Ensemble Learning Featuring the Netflix Prize Competition and ...
Ensemble Learning Featuring the Netflix Prize Competition and ...butest
 
L008.Eigenfaces And Nn Som
L008.Eigenfaces And Nn SomL008.Eigenfaces And Nn Som
L008.Eigenfaces And Nn Som
ramesh kumar
 
An Introduction to Computer Vision
An Introduction to Computer VisionAn Introduction to Computer Vision
An Introduction to Computer Vision
guestd1b1b5
 

Similar to Using classifiers to compute similarities between face images. Prof. Lior Wolf, Tel-Aviv University (20)

Computational Biology, Part 4 Protein Coding Regions
Computational Biology, Part 4 Protein Coding RegionsComputational Biology, Part 4 Protein Coding Regions
Computational Biology, Part 4 Protein Coding Regions
 
Automatic Test Suite Generation for Key-Points Detection DNNs using Many-Obje...
Automatic Test Suite Generation for Key-Points Detection DNNs using Many-Obje...Automatic Test Suite Generation for Key-Points Detection DNNs using Many-Obje...
Automatic Test Suite Generation for Key-Points Detection DNNs using Many-Obje...
 
Deep Learning: concepts and use cases (October 2018)
Deep Learning: concepts and use cases (October 2018)Deep Learning: concepts and use cases (October 2018)
Deep Learning: concepts and use cases (October 2018)
 
Machine Learning ebook.pdf
Machine Learning ebook.pdfMachine Learning ebook.pdf
Machine Learning ebook.pdf
 
1_5_AI_edx_ml_51intro_240204_104838machine learning lecture 1
1_5_AI_edx_ml_51intro_240204_104838machine learning lecture 11_5_AI_edx_ml_51intro_240204_104838machine learning lecture 1
1_5_AI_edx_ml_51intro_240204_104838machine learning lecture 1
 
Yulia Honcharenko "Application of metric learning for logo recognition"
Yulia Honcharenko "Application of metric learning for logo recognition"Yulia Honcharenko "Application of metric learning for logo recognition"
Yulia Honcharenko "Application of metric learning for logo recognition"
 
one shot15729752 Deep Learning for AI and DS
one shot15729752 Deep Learning for AI and DSone shot15729752 Deep Learning for AI and DS
one shot15729752 Deep Learning for AI and DS
 
Crystallization classification semisupervised
Crystallization classification semisupervisedCrystallization classification semisupervised
Crystallization classification semisupervised
 
know Machine Learning Basic Concepts.pdf
know Machine Learning Basic Concepts.pdfknow Machine Learning Basic Concepts.pdf
know Machine Learning Basic Concepts.pdf
 
inam
inaminam
inam
 
HiPEAC2022_António Casimiro presentation
HiPEAC2022_António Casimiro presentationHiPEAC2022_António Casimiro presentation
HiPEAC2022_António Casimiro presentation
 
Learning where to look: focus and attention in deep vision
Learning where to look: focus and attention in deep visionLearning where to look: focus and attention in deep vision
Learning where to look: focus and attention in deep vision
 
Long-term Face Tracking in the Wild using Deep Learning
Long-term Face Tracking in the Wild using Deep LearningLong-term Face Tracking in the Wild using Deep Learning
Long-term Face Tracking in the Wild using Deep Learning
 
Analytical study of feature extraction techniques in opinion mining
Analytical study of feature extraction techniques in opinion miningAnalytical study of feature extraction techniques in opinion mining
Analytical study of feature extraction techniques in opinion mining
 
ANALYTICAL STUDY OF FEATURE EXTRACTION TECHNIQUES IN OPINION MINING
ANALYTICAL STUDY OF FEATURE EXTRACTION TECHNIQUES IN OPINION MININGANALYTICAL STUDY OF FEATURE EXTRACTION TECHNIQUES IN OPINION MINING
ANALYTICAL STUDY OF FEATURE EXTRACTION TECHNIQUES IN OPINION MINING
 
Radial Basis Function Neural Network (RBFNN), Induction Motor, Vector control...
Radial Basis Function Neural Network (RBFNN), Induction Motor, Vector control...Radial Basis Function Neural Network (RBFNN), Induction Motor, Vector control...
Radial Basis Function Neural Network (RBFNN), Induction Motor, Vector control...
 
Automated Face Detection System
Automated Face Detection SystemAutomated Face Detection System
Automated Face Detection System
 
Ensemble Learning Featuring the Netflix Prize Competition and ...
Ensemble Learning Featuring the Netflix Prize Competition and ...Ensemble Learning Featuring the Netflix Prize Competition and ...
Ensemble Learning Featuring the Netflix Prize Competition and ...
 
L008.Eigenfaces And Nn Som
L008.Eigenfaces And Nn SomL008.Eigenfaces And Nn Som
L008.Eigenfaces And Nn Som
 
An Introduction to Computer Vision
An Introduction to Computer VisionAn Introduction to Computer Vision
An Introduction to Computer Vision
 

More from yaevents

Как научить роботов тестировать веб-интерфейсы. Артем Ерошенко, Илья Кацев, Я...
Как научить роботов тестировать веб-интерфейсы. Артем Ерошенко, Илья Кацев, Я...Как научить роботов тестировать веб-интерфейсы. Артем Ерошенко, Илья Кацев, Я...
Как научить роботов тестировать веб-интерфейсы. Артем Ерошенко, Илья Кацев, Я...
yaevents
 
Тема для WordPress в БЭМ. Владимир Гриненко, Яндекс
Тема для WordPress в БЭМ. Владимир Гриненко, ЯндексТема для WordPress в БЭМ. Владимир Гриненко, Яндекс
Тема для WordPress в БЭМ. Владимир Гриненко, Яндексyaevents
 
Построение сложносоставных блоков в шаблонизаторе bemhtml. Сергей Бережной, Я...
Построение сложносоставных блоков в шаблонизаторе bemhtml. Сергей Бережной, Я...Построение сложносоставных блоков в шаблонизаторе bemhtml. Сергей Бережной, Я...
Построение сложносоставных блоков в шаблонизаторе bemhtml. Сергей Бережной, Я...
yaevents
 
i-bem.js: JavaScript в БЭМ-терминах. Елена Глухова, Варвара Степанова, Яндекс
i-bem.js: JavaScript в БЭМ-терминах. Елена Глухова, Варвара Степанова, Яндексi-bem.js: JavaScript в БЭМ-терминах. Елена Глухова, Варвара Степанова, Яндекс
i-bem.js: JavaScript в БЭМ-терминах. Елена Глухова, Варвара Степанова, Яндекс
yaevents
 
Дом из готовых кирпичей. Библиотека блоков, тюнинг, инструменты. Елена Глухов...
Дом из готовых кирпичей. Библиотека блоков, тюнинг, инструменты. Елена Глухов...Дом из готовых кирпичей. Библиотека блоков, тюнинг, инструменты. Елена Глухов...
Дом из готовых кирпичей. Библиотека блоков, тюнинг, инструменты. Елена Глухов...
yaevents
 
Модели в профессиональной инженерии и тестировании программ. Александр Петрен...
Модели в профессиональной инженерии и тестировании программ. Александр Петрен...Модели в профессиональной инженерии и тестировании программ. Александр Петрен...
Модели в профессиональной инженерии и тестировании программ. Александр Петрен...
yaevents
 
Администрирование небольших сервисов или один за всех и 100 на одного. Роман ...
Администрирование небольших сервисов или один за всех и 100 на одного. Роман ...Администрирование небольших сервисов или один за всех и 100 на одного. Роман ...
Администрирование небольших сервисов или один за всех и 100 на одного. Роман ...
yaevents
 
Мониторинг со всех сторон. Алексей Симаков, Яндекс
Мониторинг со всех сторон. Алексей Симаков, ЯндексМониторинг со всех сторон. Алексей Симаков, Яндекс
Мониторинг со всех сторон. Алексей Симаков, Яндексyaevents
 
Истории про разработку сайтов. Сергей Бережной, Яндекс
Истории про разработку сайтов. Сергей Бережной, ЯндексИстории про разработку сайтов. Сергей Бережной, Яндекс
Истории про разработку сайтов. Сергей Бережной, Яндекс
yaevents
 
Разработка приложений для Android на С++. Юрий Береза, Shturmann
Разработка приложений для Android на С++. Юрий Береза, ShturmannРазработка приложений для Android на С++. Юрий Береза, Shturmann
Разработка приложений для Android на С++. Юрий Береза, Shturmann
yaevents
 
Кросс-платформенная разработка под мобильные устройства. Дмитрий Жестилевский...
Кросс-платформенная разработка под мобильные устройства. Дмитрий Жестилевский...Кросс-платформенная разработка под мобильные устройства. Дмитрий Жестилевский...
Кросс-платформенная разработка под мобильные устройства. Дмитрий Жестилевский...
yaevents
 
Julia Stoyanovich - Making interval-based clustering rank-aware
Julia Stoyanovich - Making interval-based clustering rank-awareJulia Stoyanovich - Making interval-based clustering rank-aware
Julia Stoyanovich - Making interval-based clustering rank-awareyaevents
 
Mike Thelwall - Sentiment strength detection for the social web: From YouTube...
Mike Thelwall - Sentiment strength detection for the social web: From YouTube...Mike Thelwall - Sentiment strength detection for the social web: From YouTube...
Mike Thelwall - Sentiment strength detection for the social web: From YouTube...yaevents
 
Evangelos Kanoulas — Advances in Information Retrieval Evaluation
Evangelos Kanoulas — Advances in Information Retrieval EvaluationEvangelos Kanoulas — Advances in Information Retrieval Evaluation
Evangelos Kanoulas — Advances in Information Retrieval Evaluationyaevents
 
Ben Carterett — Advances in Information Retrieval Evaluation
Ben Carterett — Advances in Information Retrieval EvaluationBen Carterett — Advances in Information Retrieval Evaluation
Ben Carterett — Advances in Information Retrieval Evaluationyaevents
 
Raffaele Perego "Efficient Query Suggestions in the Long Tail"
Raffaele Perego "Efficient Query Suggestions in the Long Tail"Raffaele Perego "Efficient Query Suggestions in the Long Tail"
Raffaele Perego "Efficient Query Suggestions in the Long Tail"yaevents
 
"Efficient Diversification of Web Search Results"
"Efficient Diversification of Web Search Results""Efficient Diversification of Web Search Results"
"Efficient Diversification of Web Search Results"yaevents
 
Salvatore_Orlando
Salvatore_OrlandoSalvatore_Orlando
Salvatore_Orlandoyaevents
 
Fast dynamic analysis, Kostya Serebryany
Fast dynamic analysis, Kostya SerebryanyFast dynamic analysis, Kostya Serebryany
Fast dynamic analysis, Kostya Serebryanyyaevents
 
Adapting Rankers Online, Maarten de Rijke
Adapting Rankers Online, Maarten de RijkeAdapting Rankers Online, Maarten de Rijke
Adapting Rankers Online, Maarten de Rijkeyaevents
 

More from yaevents (20)

Как научить роботов тестировать веб-интерфейсы. Артем Ерошенко, Илья Кацев, Я...
Как научить роботов тестировать веб-интерфейсы. Артем Ерошенко, Илья Кацев, Я...Как научить роботов тестировать веб-интерфейсы. Артем Ерошенко, Илья Кацев, Я...
Как научить роботов тестировать веб-интерфейсы. Артем Ерошенко, Илья Кацев, Я...
 
Тема для WordPress в БЭМ. Владимир Гриненко, Яндекс
Тема для WordPress в БЭМ. Владимир Гриненко, ЯндексТема для WordPress в БЭМ. Владимир Гриненко, Яндекс
Тема для WordPress в БЭМ. Владимир Гриненко, Яндекс
 
Построение сложносоставных блоков в шаблонизаторе bemhtml. Сергей Бережной, Я...
Построение сложносоставных блоков в шаблонизаторе bemhtml. Сергей Бережной, Я...Построение сложносоставных блоков в шаблонизаторе bemhtml. Сергей Бережной, Я...
Построение сложносоставных блоков в шаблонизаторе bemhtml. Сергей Бережной, Я...
 
i-bem.js: JavaScript в БЭМ-терминах. Елена Глухова, Варвара Степанова, Яндекс
i-bem.js: JavaScript в БЭМ-терминах. Елена Глухова, Варвара Степанова, Яндексi-bem.js: JavaScript в БЭМ-терминах. Елена Глухова, Варвара Степанова, Яндекс
i-bem.js: JavaScript в БЭМ-терминах. Елена Глухова, Варвара Степанова, Яндекс
 
Дом из готовых кирпичей. Библиотека блоков, тюнинг, инструменты. Елена Глухов...
Дом из готовых кирпичей. Библиотека блоков, тюнинг, инструменты. Елена Глухов...Дом из готовых кирпичей. Библиотека блоков, тюнинг, инструменты. Елена Глухов...
Дом из готовых кирпичей. Библиотека блоков, тюнинг, инструменты. Елена Глухов...
 
Модели в профессиональной инженерии и тестировании программ. Александр Петрен...
Модели в профессиональной инженерии и тестировании программ. Александр Петрен...Модели в профессиональной инженерии и тестировании программ. Александр Петрен...
Модели в профессиональной инженерии и тестировании программ. Александр Петрен...
 
Администрирование небольших сервисов или один за всех и 100 на одного. Роман ...
Администрирование небольших сервисов или один за всех и 100 на одного. Роман ...Администрирование небольших сервисов или один за всех и 100 на одного. Роман ...
Администрирование небольших сервисов или один за всех и 100 на одного. Роман ...
 
Мониторинг со всех сторон. Алексей Симаков, Яндекс
Мониторинг со всех сторон. Алексей Симаков, ЯндексМониторинг со всех сторон. Алексей Симаков, Яндекс
Мониторинг со всех сторон. Алексей Симаков, Яндекс
 
Истории про разработку сайтов. Сергей Бережной, Яндекс
Истории про разработку сайтов. Сергей Бережной, ЯндексИстории про разработку сайтов. Сергей Бережной, Яндекс
Истории про разработку сайтов. Сергей Бережной, Яндекс
 
Разработка приложений для Android на С++. Юрий Береза, Shturmann
Разработка приложений для Android на С++. Юрий Береза, ShturmannРазработка приложений для Android на С++. Юрий Береза, Shturmann
Разработка приложений для Android на С++. Юрий Береза, Shturmann
 
Кросс-платформенная разработка под мобильные устройства. Дмитрий Жестилевский...
Кросс-платформенная разработка под мобильные устройства. Дмитрий Жестилевский...Кросс-платформенная разработка под мобильные устройства. Дмитрий Жестилевский...
Кросс-платформенная разработка под мобильные устройства. Дмитрий Жестилевский...
 
Julia Stoyanovich - Making interval-based clustering rank-aware
Julia Stoyanovich - Making interval-based clustering rank-awareJulia Stoyanovich - Making interval-based clustering rank-aware
Julia Stoyanovich - Making interval-based clustering rank-aware
 
Mike Thelwall - Sentiment strength detection for the social web: From YouTube...
Mike Thelwall - Sentiment strength detection for the social web: From YouTube...Mike Thelwall - Sentiment strength detection for the social web: From YouTube...
Mike Thelwall - Sentiment strength detection for the social web: From YouTube...
 
Evangelos Kanoulas — Advances in Information Retrieval Evaluation
Evangelos Kanoulas — Advances in Information Retrieval EvaluationEvangelos Kanoulas — Advances in Information Retrieval Evaluation
Evangelos Kanoulas — Advances in Information Retrieval Evaluation
 
Ben Carterett — Advances in Information Retrieval Evaluation
Ben Carterett — Advances in Information Retrieval EvaluationBen Carterett — Advances in Information Retrieval Evaluation
Ben Carterett — Advances in Information Retrieval Evaluation
 
Raffaele Perego "Efficient Query Suggestions in the Long Tail"
Raffaele Perego "Efficient Query Suggestions in the Long Tail"Raffaele Perego "Efficient Query Suggestions in the Long Tail"
Raffaele Perego "Efficient Query Suggestions in the Long Tail"
 
"Efficient Diversification of Web Search Results"
"Efficient Diversification of Web Search Results""Efficient Diversification of Web Search Results"
"Efficient Diversification of Web Search Results"
 
Salvatore_Orlando
Salvatore_OrlandoSalvatore_Orlando
Salvatore_Orlando
 
Fast dynamic analysis, Kostya Serebryany
Fast dynamic analysis, Kostya SerebryanyFast dynamic analysis, Kostya Serebryany
Fast dynamic analysis, Kostya Serebryany
 
Adapting Rankers Online, Maarten de Rijke
Adapting Rankers Online, Maarten de RijkeAdapting Rankers Online, Maarten de Rijke
Adapting Rankers Online, Maarten de Rijke
 

Recently uploaded

When stars align: studies in data quality, knowledge graphs, and machine lear...
When stars align: studies in data quality, knowledge graphs, and machine lear...When stars align: studies in data quality, knowledge graphs, and machine lear...
When stars align: studies in data quality, knowledge graphs, and machine lear...
Elena Simperl
 
FIDO Alliance Osaka Seminar: Passkeys and the Road Ahead.pdf
FIDO Alliance Osaka Seminar: Passkeys and the Road Ahead.pdfFIDO Alliance Osaka Seminar: Passkeys and the Road Ahead.pdf
FIDO Alliance Osaka Seminar: Passkeys and the Road Ahead.pdf
FIDO Alliance
 
Empowering NextGen Mobility via Large Action Model Infrastructure (LAMI): pav...
Empowering NextGen Mobility via Large Action Model Infrastructure (LAMI): pav...Empowering NextGen Mobility via Large Action Model Infrastructure (LAMI): pav...
Empowering NextGen Mobility via Large Action Model Infrastructure (LAMI): pav...
Thierry Lestable
 
FIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdfFIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance
 
The Future of Platform Engineering
The Future of Platform EngineeringThe Future of Platform Engineering
The Future of Platform Engineering
Jemma Hussein Allen
 
GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...
GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...
GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...
Sri Ambati
 
IOS-PENTESTING-BEGINNERS-PRACTICAL-GUIDE-.pptx
IOS-PENTESTING-BEGINNERS-PRACTICAL-GUIDE-.pptxIOS-PENTESTING-BEGINNERS-PRACTICAL-GUIDE-.pptx
IOS-PENTESTING-BEGINNERS-PRACTICAL-GUIDE-.pptx
Abida Shariff
 
JMeter webinar - integration with InfluxDB and Grafana
JMeter webinar - integration with InfluxDB and GrafanaJMeter webinar - integration with InfluxDB and Grafana
JMeter webinar - integration with InfluxDB and Grafana
RTTS
 
GraphRAG is All You need? LLM & Knowledge Graph
GraphRAG is All You need? LLM & Knowledge GraphGraphRAG is All You need? LLM & Knowledge Graph
GraphRAG is All You need? LLM & Knowledge Graph
Guy Korland
 
From Siloed Products to Connected Ecosystem: Building a Sustainable and Scala...
From Siloed Products to Connected Ecosystem: Building a Sustainable and Scala...From Siloed Products to Connected Ecosystem: Building a Sustainable and Scala...
From Siloed Products to Connected Ecosystem: Building a Sustainable and Scala...
Product School
 
Smart TV Buyer Insights Survey 2024 by 91mobiles.pdf
Smart TV Buyer Insights Survey 2024 by 91mobiles.pdfSmart TV Buyer Insights Survey 2024 by 91mobiles.pdf
Smart TV Buyer Insights Survey 2024 by 91mobiles.pdf
91mobiles
 
Knowledge engineering: from people to machines and back
Knowledge engineering: from people to machines and backKnowledge engineering: from people to machines and back
Knowledge engineering: from people to machines and back
Elena Simperl
 
Mission to Decommission: Importance of Decommissioning Products to Increase E...
Mission to Decommission: Importance of Decommissioning Products to Increase E...Mission to Decommission: Importance of Decommissioning Products to Increase E...
Mission to Decommission: Importance of Decommissioning Products to Increase E...
Product School
 
Unsubscribed: Combat Subscription Fatigue With a Membership Mentality by Head...
Unsubscribed: Combat Subscription Fatigue With a Membership Mentality by Head...Unsubscribed: Combat Subscription Fatigue With a Membership Mentality by Head...
Unsubscribed: Combat Subscription Fatigue With a Membership Mentality by Head...
Product School
 
From Daily Decisions to Bottom Line: Connecting Product Work to Revenue by VP...
From Daily Decisions to Bottom Line: Connecting Product Work to Revenue by VP...From Daily Decisions to Bottom Line: Connecting Product Work to Revenue by VP...
From Daily Decisions to Bottom Line: Connecting Product Work to Revenue by VP...
Product School
 
ODC, Data Fabric and Architecture User Group
ODC, Data Fabric and Architecture User GroupODC, Data Fabric and Architecture User Group
ODC, Data Fabric and Architecture User Group
CatarinaPereira64715
 
"Impact of front-end architecture on development cost", Viktor Turskyi
"Impact of front-end architecture on development cost", Viktor Turskyi"Impact of front-end architecture on development cost", Viktor Turskyi
"Impact of front-end architecture on development cost", Viktor Turskyi
Fwdays
 
Designing Great Products: The Power of Design and Leadership by Chief Designe...
Designing Great Products: The Power of Design and Leadership by Chief Designe...Designing Great Products: The Power of Design and Leadership by Chief Designe...
Designing Great Products: The Power of Design and Leadership by Chief Designe...
Product School
 
Search and Society: Reimagining Information Access for Radical Futures
Search and Society: Reimagining Information Access for Radical FuturesSearch and Society: Reimagining Information Access for Radical Futures
Search and Society: Reimagining Information Access for Radical Futures
Bhaskar Mitra
 
Slack (or Teams) Automation for Bonterra Impact Management (fka Social Soluti...
Slack (or Teams) Automation for Bonterra Impact Management (fka Social Soluti...Slack (or Teams) Automation for Bonterra Impact Management (fka Social Soluti...
Slack (or Teams) Automation for Bonterra Impact Management (fka Social Soluti...
Jeffrey Haguewood
 

Recently uploaded (20)

When stars align: studies in data quality, knowledge graphs, and machine lear...
When stars align: studies in data quality, knowledge graphs, and machine lear...When stars align: studies in data quality, knowledge graphs, and machine lear...
When stars align: studies in data quality, knowledge graphs, and machine lear...
 
FIDO Alliance Osaka Seminar: Passkeys and the Road Ahead.pdf
FIDO Alliance Osaka Seminar: Passkeys and the Road Ahead.pdfFIDO Alliance Osaka Seminar: Passkeys and the Road Ahead.pdf
FIDO Alliance Osaka Seminar: Passkeys and the Road Ahead.pdf
 
Empowering NextGen Mobility via Large Action Model Infrastructure (LAMI): pav...
Empowering NextGen Mobility via Large Action Model Infrastructure (LAMI): pav...Empowering NextGen Mobility via Large Action Model Infrastructure (LAMI): pav...
Empowering NextGen Mobility via Large Action Model Infrastructure (LAMI): pav...
 
FIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdfFIDO Alliance Osaka Seminar: Overview.pdf
FIDO Alliance Osaka Seminar: Overview.pdf
 
The Future of Platform Engineering
The Future of Platform EngineeringThe Future of Platform Engineering
The Future of Platform Engineering
 
GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...
GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...
GenAISummit 2024 May 28 Sri Ambati Keynote: AGI Belongs to The Community in O...
 
IOS-PENTESTING-BEGINNERS-PRACTICAL-GUIDE-.pptx
IOS-PENTESTING-BEGINNERS-PRACTICAL-GUIDE-.pptxIOS-PENTESTING-BEGINNERS-PRACTICAL-GUIDE-.pptx
IOS-PENTESTING-BEGINNERS-PRACTICAL-GUIDE-.pptx
 
JMeter webinar - integration with InfluxDB and Grafana
JMeter webinar - integration with InfluxDB and GrafanaJMeter webinar - integration with InfluxDB and Grafana
JMeter webinar - integration with InfluxDB and Grafana
 
GraphRAG is All You need? LLM & Knowledge Graph
GraphRAG is All You need? LLM & Knowledge GraphGraphRAG is All You need? LLM & Knowledge Graph
GraphRAG is All You need? LLM & Knowledge Graph
 
From Siloed Products to Connected Ecosystem: Building a Sustainable and Scala...
From Siloed Products to Connected Ecosystem: Building a Sustainable and Scala...From Siloed Products to Connected Ecosystem: Building a Sustainable and Scala...
From Siloed Products to Connected Ecosystem: Building a Sustainable and Scala...
 
Smart TV Buyer Insights Survey 2024 by 91mobiles.pdf
Smart TV Buyer Insights Survey 2024 by 91mobiles.pdfSmart TV Buyer Insights Survey 2024 by 91mobiles.pdf
Smart TV Buyer Insights Survey 2024 by 91mobiles.pdf
 
Knowledge engineering: from people to machines and back
Knowledge engineering: from people to machines and backKnowledge engineering: from people to machines and back
Knowledge engineering: from people to machines and back
 
Mission to Decommission: Importance of Decommissioning Products to Increase E...
Mission to Decommission: Importance of Decommissioning Products to Increase E...Mission to Decommission: Importance of Decommissioning Products to Increase E...
Mission to Decommission: Importance of Decommissioning Products to Increase E...
 
Unsubscribed: Combat Subscription Fatigue With a Membership Mentality by Head...
Unsubscribed: Combat Subscription Fatigue With a Membership Mentality by Head...Unsubscribed: Combat Subscription Fatigue With a Membership Mentality by Head...
Unsubscribed: Combat Subscription Fatigue With a Membership Mentality by Head...
 
From Daily Decisions to Bottom Line: Connecting Product Work to Revenue by VP...
From Daily Decisions to Bottom Line: Connecting Product Work to Revenue by VP...From Daily Decisions to Bottom Line: Connecting Product Work to Revenue by VP...
From Daily Decisions to Bottom Line: Connecting Product Work to Revenue by VP...
 
ODC, Data Fabric and Architecture User Group
ODC, Data Fabric and Architecture User GroupODC, Data Fabric and Architecture User Group
ODC, Data Fabric and Architecture User Group
 
"Impact of front-end architecture on development cost", Viktor Turskyi
"Impact of front-end architecture on development cost", Viktor Turskyi"Impact of front-end architecture on development cost", Viktor Turskyi
"Impact of front-end architecture on development cost", Viktor Turskyi
 
Designing Great Products: The Power of Design and Leadership by Chief Designe...
Designing Great Products: The Power of Design and Leadership by Chief Designe...Designing Great Products: The Power of Design and Leadership by Chief Designe...
Designing Great Products: The Power of Design and Leadership by Chief Designe...
 
Search and Society: Reimagining Information Access for Radical Futures
Search and Society: Reimagining Information Access for Radical FuturesSearch and Society: Reimagining Information Access for Radical Futures
Search and Society: Reimagining Information Access for Radical Futures
 
Slack (or Teams) Automation for Bonterra Impact Management (fka Social Soluti...
Slack (or Teams) Automation for Bonterra Impact Management (fka Social Soluti...Slack (or Teams) Automation for Bonterra Impact Management (fka Social Soluti...
Slack (or Teams) Automation for Bonterra Impact Management (fka Social Soluti...
 

Using classifiers to compute similarities between face images. Prof. Lior Wolf, Tel-Aviv University

  • 1. Learning visual similarity using classifiers Lior Wolf, The Blavatnik School of Computer Science, Tel-Aviv University Collaborators: Students: YanivTaigman Tal Hassner Orit Klipper-Gross Itay Maoz face.com Open U Weizmann inst. Tel-Aviv U 1
  • 2. The Blavatnik School of Computer ScienceTel-Aviv University An example of higher education in Israel 2
  • 3. A school in the Faculty of Exact Sciences that also includes: Mathematics, Physics, Chemistry, Geophysics and Planetary Sciences Originated in the 1970’s as part of the School of Math, since 2000 a separate School 39 Faculty Members ~1000 undergrads ~200 MSc students ~70 PhD students Post Docs and other research personnel 3
  • 4. School Ranking in the world TAU/CS Ranked #29 in number of citations - Thompson Scientific, (for the years 2000-2010).[Technion #33 , Weizmann #72, HebrewU #105] TAU/CS Ranked #28 by the Shanghai Academic Ranking of World Universities in Computer Science – 2011[Weizmann #12, Technion #15, HebrewU #21] TAU/CS Ranked #14 in the world in CS impact – Scientometrics, Vol. 76, No. 2, 2008. 12 TAU/CS faculty in positions 1-100 in “list of most central computer Scientists in Theory of Computer Science” (Kuhn – Wattenhofer, Sigact news, Dec ’07) 4
  • 5. Computer vision in search Query Raw data: images,video, audio Information:objects,tags, IDs,context Searchresults Preprocessing 5
  • 6. The pain: too many images On : Over 1,000,000,000 photos uploaded each month shared by 200,000,000+ users 10’s of billions served/week No tags  No Photos… “can I see all my photos?” “tagging takes hours, can you do that for me?” 6
  • 7. The evolution of perceptual search Reranking bysimilarity Text-basedimage search Specializationin face identification Low-level vision Mid-level vision No vision With basicproperties Gist-based Image similarity Catalog basedsearch High-level vision: scene understanding 7
  • 8. Photo Finder for facebook 8
  • 9. THE 1st MOBILE APP TO FIND 3D ITEMS 9
  • 10. WHAT MAKES IT SO HARD? 10
  • 11. High-level vision: what is where? A happy couple walks in a fieldWhat kind of field? Where? Which season? How old are they? Gender? How attractive? What are they wearing? High-level vision: scene understanding 11
  • 12. YaC, Moscow, September 19, 2011 Learning visual similarity using classifiers Lior Wolf, The Blavatnik School of Computer Science, Tel-Aviv University Collaborators: Students: YanivTaigman Tal Hassner Orit Klipper-Gross Itay Maoz face.com Open U Weizmann inst. Tel-Aviv U 12
  • 14. Modeling never before seen objects Natural setup for image retrieval with no categories Training: The Pair-Matching Problem 14
  • 15. Instances Document Analysis Video Action Recogniti0n Face Recogniti0n Video Face Recogniti0n 15
  • 17. Labeled Faces in the Wild (LFW) 13,000 labeled images of faces collected from the web 5,749 individuals 1-150 images per individual Training: 17
  • 18. Restricted Protocol 10-fold cross validation tests on randomly generated splits, each with: 300 same pairs 300 not same pairs 18
  • 19. Pipeline (take 1)* Training. Note: no use of labels! =1 Sim ( , ) same =2 Sim( , ) Classifier (e.g.SVM) =i Sim ( , ) not same =i+1 Sim( , ) Threshold * “Descriptor Based Methods in the Wild,” ECCVw’08 19
  • 20. Pipeline (take 1)* Training – multiple descriptors similarities (1,1,1,2,…,1,n) same (2,1,2,2,…,2,n) Classifier (e.g.SVM) (i,1,i,2,…,i,n) not same (i+1,1,i+1,2,…,i+1,n) * “Descriptor Based Methods in the Wild,” ECCVw’08 20
  • 21. Some Questions How to represent the images? Which similarity to use? Grayscales, Edge responces [Brunelli & Poggio’93], C1-Gabor [e.g., Riesenhuber & Poggio’99], SIFT [Lowe’04], LBP [e.g., Ojala & Pietikainen & Harwood’96],… Later on: How can subject IDs help improve pair-matching performance? L2, Correlation, Learned metrics [e.g., Bilenko etal.’04, Cristianini etal.’02, Hertz etal. 04, …], “hand-crafted” metrics [e.g., Belongie etal.’01] 21
  • 22. One-Shot Similarity (OSS) Score* What: A measure of the similarity between two vectors Input: The two vectors A set of “Background samples” How: Use “One-Shot Learning” (classification with one positive example) * “Descriptor Based Methods in the Wild,” ECCVw’08 “The One-Shot Similarity Kernel”, ICCV’09 22
  • 23. Computing the “One-Shot” Similarity Step a: Model1 = train (p, A) Step b: Score1 = classify(q, Model1) Step c: Model2 = train (q, A) Set “A” of background examples Step d: Score2 = classify(p, Model1) q One-Shot-Sim = (score1 + score2) /2 Similarity  p 23
  • 24. Euclidean vs. One-Shot Visualized One-Shot Euclidean 24
  • 25. Euclidean vs. One-Shot Visualized One-Shot Euclidean 25
  • 26. Computing the “One-Shot” Similarity Using LDA as the underlying classifier : Where is the mean of set A, and is the pseudo-inverse of the intra-class cov. matrix. * “The One-Shot Similarity Kernel”, ICCV’09 26
  • 27. Computing the “One-Shot” Similarity Using Free-Scale LDA as the underlying classifier : Where is the mean of set A, and is the pseudo-inverse of the intra-class cov. matrix. * “The One-Shot Similarity Kernel”, ICCV’09 27
  • 28. Some Properties of the OSS* Uses unlabeled training data OSS based on Free-Scale LDA is a CPD Kernel May be efficiently computed Complexity: is independent of the two vectors compared, and so computed only once. Also, repeated comparisons of a vector xi to different xj may be performed in O(n) * “The One-Shot Similarity Kernel”, ICCV’09 28
  • 29. Some Properties of the OSS* * “The One-Shot Similarity Kernel”, ICCV’09 29
  • 30. Some Properties of the OSS* OSS based on Free-Scale LDA is a CPD Kernel * “The One-Shot Similarity Kernel”, ICCV’09 30
  • 31. Metric learning for OSS* Instead of examples xi use Txi for some “optimal” T The transformation T is obtained by a gradient decent procedure that optimizes the score: *“One Shot Similarity Metric Learning for Action Recognition”, In submission. 31
  • 32. The Unrestricted Protocol 10-fold cross validation tests on randomly generated splits, each with: 300 same pairs 300 not same pairs Training now includes subject labels 32
  • 33. Multiple One-Shots* We now have IDs. How do we use them? Compute multiple OSS, each time using examples from a single class * “Multiple One-Shots for Utilizing Class Label Information,” BMVC’09 33
  • 35. Multiple One-Shots We now have IDs. How do we use them? Compute multiple OSS, each time using examples from a single class Discrimination based on different sources of variation: Subject ID, Pose, etc. 35
  • 36. The Pose Issue Most confident wrong results* * “Descriptor Based Methods in the Wild,” ECCVw’08 36
  • 37. Getting Poses To compute Pose based OSS, you need sets of images in the same pose… 7 fiducial points (eyes, mouth, nose) 14 x,y coordinates 14D vector of alignment errors (similarity trnsf.) Project to first Principal Component Bin to 10 classes 37
  • 39. Multiple One-Shots - Examples 5 Id-based OSS and 5 Pose-based OSS scores Identity Pose 39
  • 40. Multiple One-Shots - Examples Identity Pose 40
  • 41. Multiple One-Shots - Examples Identity Pose 41
  • 42. Pipeline* Input image pair Image alignment Commercial alignment software by * “Multiple One-Shots for Utilizing Class Label Information,” BMVC’09 42
  • 43.
  • 45. TPLBP, FPLBP [Wolf etal.’08]* “Multiple One-Shots for Utilizing Class Label Information,” BMVC’09 43
  • 46. Pipeline* Input image pair Image alignment Feature vectors PCA+ITML Information Theoretic Metric Learning [Davis etal.’07] * “Multiple One-Shots for Utilizing Class Label Information,” BMVC’09 44
  • 47. Pipeline* Input image pair Image alignment Feature vectors Multiple OSS scores PCA+ITML 20 Subjects 10 Poses * “Multiple One-Shots for Utilizing Class Label Information,” BMVC’09 45
  • 48. Pipeline* Input image pair Image alignment Feature vectors Multiple OSS scores SVM classifier Output PCA+ITML Same Not-same * “Multiple One-Shots for Utilizing Class Label Information,” BMVC’09 46
  • 49. Pipeline – Multiple Descriptors* Feature vectors SIFT Multiple OSS scores PCA+ITML Image alignment SVM classifier Output Same Not-same Feature vectors LBP Multiple OSS scores PCA+ITML * “Multiple One-Shots for Utilizing Class Label Information,” BMVC’09 47
  • 50. 0.7847 ± 0.0051 [WHT’08] 0.8398 ± 0.0035 [WHT’08 + alignment] Results 0.8517 ± 0.0061 [this work, only LBP] 0.8950 ± 0.0051 [this work, multi-desc.] 0.9753 [Kumar etal. 09 - HUMAN] 48
  • 51. Pair-Matching of Sets * Face Recognition in Unconstrained Videos with Matched B/G Similarity. CVPR 2011. 49
  • 52. Pair-Matching of Sets Training: 50
  • 53.
  • 54. pose based methods, comparing the two most frontal faces in each video or the two faces with the most similar pose.
  • 55. algebraic methods set-to-set methods, such as max correlation, projection and Procrustes.
  • 56. non algebraic methods such as PMK and LLC. 51
  • 57.
  • 58. B: background set of faces.Similarity = MBGS(X1, X2, B) B1 = Find_Nearest_Neighbors(X1,B) Model1 = train(X1, B1) Confidences1 = classify(X2, Model1) Sim1 = mean(Confidences1) X2 Similarity  X1 52
  • 59.
  • 60. B: background set of faces.Similarity = MBGS(X1, X2, B) B1 = Find_Nearest_Neighbors(X1,B) Model1 = train(X1, B1) Confidences1 = classify(X2, Model1) Sim1 = mean(Confidences1) B2 = Find_Nearest_Neighbors(X2, B) Model2 = train(X2, B2) Confidences2 = classify(X1, Model2) Sim2 = mean(Confidences2) Similarity = (Sim1+Sim2)/2 53
  • 61. Thank you! Software available: http://www.cs.tau.ac.il/~wolf 54
  • 62.
  • 63. Easy to benchmark
  • 64. We show that whatever works for pair matching works for identification
  • 65. L. Wolf, T. Hassner, and Y. Taigman Effective Face Recognition by Combining Multiple Descriptors and Learned Background Statistic. IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 2011.55
  • 66. Papers L. Wolf, T. Hassner, and Y. Taigman Effective Face Recognition by Combining Multiple Descriptors and Learned Background Statistic. IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 2011. L. Wolf, T. Hassner, and Y. Taigman The One-Shot Similarity Kernel IEEE International Conference on Computer Vision (ICCV), 2009. L. Wolf, Y. Taigman and T. Hassner Similarity Scores based on Background Samples Asian Conference on Computer Vision (ACCV), 2009. Y. Taigman, L. Wolf, and T. Hassner Multiple One-Shots for Utilizing Class Label Information The British Machine Vision Conference (BMVC) , 2009.  L. Wolf, T. Hassner, and Y. Taigman Descriptor Based Methods in the Wild Post ECCV workshop on Faces in Real-Life Images: Detection, Alignment, and Recognition , 2008 56
  • 67. Papers L. Wolf, T. Hassner, and I. Maoz. Face Recognition in Unconstrained Videos with Matched Background Similarity. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR),2011. L. Wolf, R. Littman, N. Mayer, T. German, N. Dershowitz, R. Shweka, and Y. Choueka. Identifying Join Candidates in the Cairo Genizah. International Journal of Computer Vision (IJCV), 2011. O. Klipper-Gross, T. Hassner, and L. Wolf. One Shot Similarity Metric Learning for Action Recognition. Submitted, 2011. 57