INTERACTIVE SEARCH FOR IMAGE CATEGORIES BY MENTAL MATCHING
|
|
- Roy Ellis
- 6 years ago
- Views:
Transcription
1 INTERACTIVE SEARCH FOR IMAGE CATEGORIES BY MENTAL MATCHING Donald GEMAN Dept. of Applied Mathematics and Statistics Center for Imaging Science Johns Hopkins University and INRIA, France
2 Collaborators Yuchun FANG (former postdoc, INRIA) Marin FERECATU (postdoc, INRIA) 11/14/2007 2
3 Outline Standard Image Retrieval Mental Matching A Mathematical Framework Modeling Human Behavior Experiments 11/14/2007 3
4 Query-by-Example (QBE) Start from an query image in a database. Find other images which are close or closest in overall color or texture or shape, or in some spatial region, or in a semantic sense, or Matching is performed by the system. Good results in limited domains, e.g., comparing paintings, plants and landscapes. 11/14/2007 4
5 Random I IKONA Search Engine (INRIA) 11/14/2007 5
6 RetrieveI.1 QBE Alinari database images 11/14/2007 6
7 Retrieve I.2 QBE Alinari database images 11/14/2007 7
8 Relevance Feedback (RF) Find the images in a database which satisfy a particular theme. An iterative learning process. An active user, providing positive and negative examples. Requires a starting point (some examples) Better adapted to semantics than QBE. 11/14/2007 8
9 Web Database, images Target theme: interior car design QBE: semantic gap Relevance Feedback 11/14/2007 9
10 Outline Standard Image Retrieval Mental Matching A Mathematical Framework Modeling Human Behavior Experiments 11/14/
11 Page Zero Problem QBE and RF require a starting point: A query image for QBE Positive and negative examples for RF Dilemma: Random sampling a large database is too slow in practice. 11/14/
12 External Images Mental Image: The user has a picture in mind, e.g., a face or painting or house. Viewed Image: The user is looking at a picture, e.g., in a magazine or on the web. Physical Object: The user is holding an object. 11/14/
13 Who is that person? Beckham?? Steve McQueen?? zizou?? 11/14/
14 Mental Category Search Assume this external query is represented in our database, either by a version of the same image (e.g., same person), or variations on a theme, i.e., a category of images (e.g., similar houses). Objective: Find an efficient way to display this version or representatives of this category. Solution: Small database: direct inspection Large database: interactive search 11/14/
15 Small database: direct search Large database:?? 11/14/
16 Potential Applications Image retrieval ( page zero ) Web browsing Security Art management E-Commerce Multimedia content providers Blah blah blah 11/14/
17 Simplifications Single target search is a special case of category search with singleton categories. Hence, Target: the object of the search, whether variations on a single image or on a theme. Assume the user always recognizes an instance of his target. 11/14/
18 Interactive Search At each iteration, some images are displayed, typically two to sixteen. The user responds by signaling his target if present; otherwise by selecting some as relevant and not relevant, or choosing the one deemed closest, or 11/14/
19 Interactive Search (cont) Based on this feedback, the system chooses another set of images to display. Goal: Minimize the number of iterations until an exemplar of the target is displayed. Then display other examples ( page zero ) for specialization and refinement. 11/14/
20 Scenarios The user may be naïve or primed (about the nature of the image representations). The target may or may not be constantly displayed. The database may or may not be structured into categories. 11/14/
21 Mental Face Retrieval Interface 11/14/
22 Category Search Interface 11/14/
23 Complications Mental matching involves human memory, perception and opinions. People are semantically oriented. However, images are not indexed by semantic content, but rather by low-level features ( semantic gap ). Large databases, order 10,000 to 1,000, /14/
24 Outline Standard Image Retrieval Mental Matching A Mathematical Framework Modeling Human Behavior Experiments 11/14/
25 Structured Database Ω= N k I, I,..., I : Database of images { } 1 2 Ω=Ω Ω 1 { } N K : Division into K categories : The number of images in category k Y 1,2,...,K : The user's category 11/14/
26 Features and Metrics f ( I ), f ( I ),..., f ( I ) :"features" in R { } 1 2 [ ] M M d : R R 0,1 : Metric on features f d ( i, j) : "Metric" on categories: N 1 1 d ( i, j) = d f ( f ( I), f ( I ')) N N I Ω I Ω i j i ' j M 11/14/
27 Metrics vs. Semantics First two principal components (PCA): Red: Meadow category, 100 images Green: 100 random images Blue: Whole database Distance histogram: Red: Meadow category, 100 images Blue: 100 random images 11/14/
28 Metrics vs. Semantics (cont) First two principal components (PCA): Red: Waterfall category, 100 images Green: 100 random images Blue: Whole database Distance histogram: Red: Waterfall category, 100 images Blue: 100 random images 11/14/
29 Display D 1, 2,..., K : A set of L distinct categories. { } (The actual display is Limages, one per category.) Dt ( ) : The categories displayed at time t= 1,2,... X D : The response of the user to D. For Y D, X = i means i is "closest" to D Y, in the opinion of the user and for the category Y in the mind of the user. 11/14/
30 Bayesian Framework Prior Distribution: A probability on categories (on individual images for single target search). Answer Model: The distribution of the user s response given the target. Display Model: An algorithm for choosing the display. Posterior Distribution (at time t): The likelihood after iteration t of each category being the target. 11/14/
31 Bayesian Framework (cont) Prior model: p ( k) = P( Y = k), k = 1,..., K 0 Answer ("data") model: P( X = i Y = k), k D D History ("evidence") after t steps: {,..., } H = X = a X = a t D (1) 1 D ( t) t Posterior distribution at step t : p ( k) = P( Y = k H ) t t 11/14/
32 The Posterior Distribution Updating p for t 0 requires the joint t distribution of Y and { X,..., X } D(1) D( t+ 1) Basic assumption: Given Y, the answers X for different Ds ' are independent random variables Consequently, p ( k) = P( Y = k H, X = a ) t+ 1 t D( t+ 1) t+ 1 PX ( = a Y= kp ) ( k) Dt ( + 1) t+ 1 t D 11/14/
33 Display Criterion D( t + 1) = arg max I( X ; Y H ) = D D t arg min H ( Y H, X ) D t D Due to conditional independence, H ( Y H, X ) is determined by p and X Y t t D D 11/14/
34 Recall: I( X ; Y ) = H ( Y ) H ( Y X ) = H ( X ) H ( X Y ) where and H ( Y ) = P( Y = y) log P( Y = y) y H ( Y X ) = P( X = x) x P ( Y = y X = x) log P( Y = y X = x) y 11/14/
35 Outline Standard Image Retrieval Mental Matching A Mathematical Framework Modeling Human Behavior Experiments 11/14/
36 Taking Stock In our framework, category search by mental matching reduces to two difficult tasks: An optimization problem: Discover approximations to the optimal display. A modeling problem: Discover answer models which match human behavior. 11/14/
37 Ideal User For k D : P( X = i Y = k ) = 1 D if d ( i, k ) < d ( j, k ) for all i, j D, j i. Since Y determines X : D ( t + 1) arg max I ( X ; Y H ) = arg m ax H ( X H ), w hich m otivates the follow ing: D D D t D D t 11/14/
38 Display Model: Heuristics Select the category k with the maximum posterior mass as the first element of D(t) Based on distances to k, collect categories until the total mass is close to 1/L. Remove this cluster from consideration. Repeat until L categories are selected. 11/14/
39 Optimal Display: The Voronoi Cells Have Equal Mass /14/
40 Real Answers φ( dik (, )) P( X D = i Y = k) = φ( d( j, k)) ( φ decreasing on [0,1]) Ex: Random response: φ 1 Ex: Optimistic choices: j D φ( d) = 1 d, φ( d ) = 1 / d Ex: M ore realistic choice (where θ1 and θ 2 are estimated from data): 0 0 θ /14/ θ 2 φ
41 Parameters Interpretation: θ 2 controls coherence with the distance, e.g.,the mass on a near-perfect match; θ 1 represents a no preference threshold (between two categories this far away from Y) Estimated from real data by maximum likelihood assuming independent decisions. 11/14/
42 Outline Standard Image Retrieval Mental Matching A Mathematical Framework Modeling Human Behavior Experiments 11/14/
43 Measures of Performance T: number of iterations until Y is displayed. P(T<t): The probability distribution some population of users. E(T): The mean of this population. Coherence: The probability that the user selects the i th closest category to his target, i=1,,l. 11/14/
44 Baseline Means: Random Search Parameters: N: # images in database K: # of classes L: # displayed per iteration Structured database, no category re-displayed: E(T) K/(2L) Unstructured database: E(T) N/(L(N*+1)), where N* = # images satisfying the user s theme. 11/14/
45 Experiment I: Mental Face Retrieval Feret Database 1199 images of distinct faces User memorizes one face Image Descriptors Preprocessing: lighting adjustment, alignment Region-based subspace methods, e.g., PCA, LDA, ICA, KPCA, KLDA. 11/14/
46 Experimental Conditions Web-based interface FERET balanced database (512 subjects, 1 image/subject) N=8 images per display θ 1 ~.1, θ 2 ~ complete searches (seven researchers at INRIA) representing 665 total decisions. 11/14/
47 Search Time Distribution: One Face 1 Probability distribution of T Number of iterations φ (d)=1-d, E(T)=10.5 S im ulatio n,, E (T )=9.7 R e al T e s t, E (T )=1 3.9 Random Response, E(T)=32 11/14/
48 Experiment II: Category Search Test Database 246 categories 9 images per category High intra-cluster semantic coherence Image Descriptors Global color, texture and shape, equally weighted 120 dimensions 11/14/
49 Category to Category Diversity 11/14/
50 Four Themes 11/14/
51 Experimental Conditions Web-based interface Target displayed alongside n=8 images per display θ 1 ~.27, θ 2 ~ complete searches (10 researchers at INRIA) representing 874 total decisions 11/14/
52 Coherence with System Metric 11/14/
53 Search Time Distribution: Categories 11/14/
54 Experiment III: Unstructured Database Ω= I, I,..., I : Database of images { } 1 2 S Ω : The category in the user's mind, a random set. Define Y = 1 if k S and k Y = 0 if k S, k=1,...,n k N Maintain a separate Bayesian system for each k. 11/14/
55 11/14/ Answer Models + + = = = D x j i k i D j k x d k x d Y x X P )), ( ( )), ( ( 1) ( φ φ = = = D x j i k i D j k x d k x d Y x X P )), ( ( )), ( ( 0) ( φ φ Positive model Negative model
56 Parameter Estimation 652 data items collected from 12 users:( Si, Di, xi) L + ( θ 1, θ 2 ) = Pi ( Si, Di, xi ) P ( S i i, D i, x i i ) = φ ( d( x + x j D i + i, S φ ( d( x j i )), S i )) φ φ + φ θ1 θ /14/
57 Semantic Ground Truth Sample from three semantic categories: - Monument Valley (left) - Pedigree dogs (middle) - Waterfalls (right) Other classes: African antelope, Butterfly, Doors of Paris, Fireworks, Deep forest, Molecule, Owl. 11/14/
58 Coherence with System Metric The probability that the user selects the m-th closest image to the target. 11/14/
59 The User Interface 11/14/
60 Alinari Database Madonna and Child (top rows) and Horse and Rider (bottom rows). 11/14/
61 Search Time Distribution: N=20,000 Cumulative distribution of the search time for real, ideal and random users. 11/14/
62 Conclusions Rich possibilities for mathematical modeling in building efficient man-machine interfaces. Mixes geometry, probability, optimization and information theory. Solving the vision problem is probably not around the corner. Hence extending to databases of order 1,000,000 remains a challenge. 11/14/
Bayesian Approaches to Content-based Image Retrieval
Bayesian Approaches to Content-based Image Retrieval Simon Wilson Georgios Stefanou Department of Statistics Trinity College Dublin Background Content-based Image Retrieval Problem: searching for images
More informationCS 223B Computer Vision Problem Set 3
CS 223B Computer Vision Problem Set 3 Due: Feb. 22 nd, 2011 1 Probabilistic Recursion for Tracking In this problem you will derive a method for tracking a point of interest through a sequence of images.
More informationBayesian analysis of genetic population structure using BAPS: Exercises
Bayesian analysis of genetic population structure using BAPS: Exercises p S u k S u p u,s S, Jukka Corander Department of Mathematics, Åbo Akademi University, Finland Exercise 1: Clustering of groups of
More informationCS 231A Computer Vision (Fall 2012) Problem Set 3
CS 231A Computer Vision (Fall 2012) Problem Set 3 Due: Nov. 13 th, 2012 (2:15pm) 1 Probabilistic Recursion for Tracking (20 points) In this problem you will derive a method for tracking a point of interest
More informationComputer Vision. Exercise Session 10 Image Categorization
Computer Vision Exercise Session 10 Image Categorization Object Categorization Task Description Given a small number of training images of a category, recognize a-priori unknown instances of that category
More information10701 Machine Learning. Clustering
171 Machine Learning Clustering What is Clustering? Organizing data into clusters such that there is high intra-cluster similarity low inter-cluster similarity Informally, finding natural groupings among
More informationIMAGE SEGMENTATION. Václav Hlaváč
IMAGE SEGMENTATION Václav Hlaváč Czech Technical University in Prague Faculty of Electrical Engineering, Department of Cybernetics Center for Machine Perception http://cmp.felk.cvut.cz/ hlavac, hlavac@fel.cvut.cz
More informationSensor Tasking and Control
Sensor Tasking and Control Outline Task-Driven Sensing Roles of Sensor Nodes and Utilities Information-Based Sensor Tasking Joint Routing and Information Aggregation Summary Introduction To efficiently
More informationGene Clustering & Classification
BINF, Introduction to Computational Biology Gene Clustering & Classification Young-Rae Cho Associate Professor Department of Computer Science Baylor University Overview Introduction to Gene Clustering
More informationObject Recognition. Lecture 11, April 21 st, Lexing Xie. EE4830 Digital Image Processing
Object Recognition Lecture 11, April 21 st, 2008 Lexing Xie EE4830 Digital Image Processing http://www.ee.columbia.edu/~xlx/ee4830/ 1 Announcements 2 HW#5 due today HW#6 last HW of the semester Due May
More informationObject Recognition Using Pictorial Structures. Daniel Huttenlocher Computer Science Department. In This Talk. Object recognition in computer vision
Object Recognition Using Pictorial Structures Daniel Huttenlocher Computer Science Department Joint work with Pedro Felzenszwalb, MIT AI Lab In This Talk Object recognition in computer vision Brief definition
More informationCOSC160: Detection and Classification. Jeremy Bolton, PhD Assistant Teaching Professor
COSC160: Detection and Classification Jeremy Bolton, PhD Assistant Teaching Professor Outline I. Problem I. Strategies II. Features for training III. Using spatial information? IV. Reducing dimensionality
More informationAutomatic Categorization of Image Regions using Dominant Color based Vector Quantization
Automatic Categorization of Image Regions using Dominant Color based Vector Quantization Md Monirul Islam, Dengsheng Zhang, Guojun Lu Gippsland School of Information Technology, Monash University Churchill
More informationLearning Spatial Context: Using Stuff to Find Things
Learning Spatial Context: Using Stuff to Find Things Wei-Cheng Su Motivation 2 Leverage contextual information to enhance detection Some context objects are non-rigid and are more naturally classified
More informationECLT 5810 Clustering
ECLT 5810 Clustering What is Cluster Analysis? Cluster: a collection of data objects Similar to one another within the same cluster Dissimilar to the objects in other clusters Cluster analysis Grouping
More informationECLT 5810 Clustering
ECLT 5810 Clustering What is Cluster Analysis? Cluster: a collection of data objects Similar to one another within the same cluster Dissimilar to the objects in other clusters Cluster analysis Grouping
More informationLearning to Recognize Faces in Realistic Conditions
000 001 002 003 004 005 006 007 008 009 010 011 012 013 014 015 016 017 018 019 020 021 022 023 024 025 026 027 028 029 030 031 032 033 034 035 036 037 038 039 040 041 042 043 044 045 046 047 048 049 050
More informationA Content Based Image Retrieval System Based on Color Features
A Content Based Image Retrieval System Based on Features Irena Valova, University of Rousse Angel Kanchev, Department of Computer Systems and Technologies, Rousse, Bulgaria, Irena@ecs.ru.acad.bg Boris
More informationMultimedia Data Management M
ALMA MATER STUDIORUM - UNIVERSITÀ DI BOLOGNA Multimedia Data Management M Second cycle degree programme (LM) in Computer Engineering University of Bologna Semantic Multimedia Data Annotation Home page:
More informationSupervised Learning for Image Segmentation
Supervised Learning for Image Segmentation Raphael Meier 06.10.2016 Raphael Meier MIA 2016 06.10.2016 1 / 52 References A. Ng, Machine Learning lecture, Stanford University. A. Criminisi, J. Shotton, E.
More informationExtracting Layers and Recognizing Features for Automatic Map Understanding. Yao-Yi Chiang
Extracting Layers and Recognizing Features for Automatic Map Understanding Yao-Yi Chiang 0 Outline Introduction/ Problem Motivation Map Processing Overview Map Decomposition Feature Recognition Discussion
More informationClustering and The Expectation-Maximization Algorithm
Clustering and The Expectation-Maximization Algorithm Unsupervised Learning Marek Petrik 3/7 Some of the figures in this presentation are taken from An Introduction to Statistical Learning, with applications
More informationK Means Clustering Using Localized Histogram Analysis and Multiple Assignment. Michael Bryson 4/18/2007
1 K Means Clustering Using Localized Histogram Analysis and Multiple Assignment Michael Bryson 4/18/2007 2 Outline Introduction Redefining Distance Preliminary Results Multiple Assignment Discussion 3
More informationMesh segmentation. Florent Lafarge Inria Sophia Antipolis - Mediterranee
Mesh segmentation Florent Lafarge Inria Sophia Antipolis - Mediterranee Outline What is mesh segmentation? M = {V,E,F} is a mesh S is either V, E or F (usually F) A Segmentation is a set of sub-meshes
More informationOptimizing feature representation for speaker diarization using PCA and LDA
Optimizing feature representation for speaker diarization using PCA and LDA itsikv@netvision.net.il Jean-Francois Bonastre jean-francois.bonastre@univ-avignon.fr Outline Speaker Diarization what is it?
More informationPSU Student Research Symposium 2017 Bayesian Optimization for Refining Object Proposals, with an Application to Pedestrian Detection Anthony D.
PSU Student Research Symposium 2017 Bayesian Optimization for Refining Object Proposals, with an Application to Pedestrian Detection Anthony D. Rhodes 5/10/17 What is Machine Learning? Machine learning
More informationClustering. Mihaela van der Schaar. January 27, Department of Engineering Science University of Oxford
Department of Engineering Science University of Oxford January 27, 2017 Many datasets consist of multiple heterogeneous subsets. Cluster analysis: Given an unlabelled data, want algorithms that automatically
More informationComputer vision: models, learning and inference. Chapter 13 Image preprocessing and feature extraction
Computer vision: models, learning and inference Chapter 13 Image preprocessing and feature extraction Preprocessing The goal of pre-processing is to try to reduce unwanted variation in image due to lighting,
More informationClassification. Vladimir Curic. Centre for Image Analysis Swedish University of Agricultural Sciences Uppsala University
Classification Vladimir Curic Centre for Image Analysis Swedish University of Agricultural Sciences Uppsala University Outline An overview on classification Basics of classification How to choose appropriate
More informationQuery Decomposition: A Multiple Neighborhood Approach to Relevance Feedback Processing in Content-based Image Retrieval
Query Decomposition: A Multiple Neighborhood Approach to Relevance Feedback Processing in Content-based Image Retrieval Kien A. Hua, Ning Yu, Danzhou Liu School of Electrical Engineering and Computer Science
More informationImage Processing (IP)
Image Processing Pattern Recognition Computer Vision Xiaojun Qi Utah State University Image Processing (IP) Manipulate and analyze digital images (pictorial information) by computer. Applications: The
More informationFUSION OF MULTITEMPORAL AND MULTIRESOLUTION REMOTE SENSING DATA AND APPLICATION TO NATURAL DISASTERS
FUSION OF MULTITEMPORAL AND MULTIRESOLUTION REMOTE SENSING DATA AND APPLICATION TO NATURAL DISASTERS Ihsen HEDHLI, Josiane ZERUBIA INRIA Sophia Antipolis Méditerranée (France), Ayin team, in collaboration
More informationEnergy Minimization for Segmentation in Computer Vision
S * = arg S min E(S) Energy Minimization for Segmentation in Computer Vision Meng Tang, Dmitrii Marin, Ismail Ben Ayed, Yuri Boykov Outline Clustering/segmentation methods K-means, GrabCut, Normalized
More informationVideo Google faces. Josef Sivic, Mark Everingham, Andrew Zisserman. Visual Geometry Group University of Oxford
Video Google faces Josef Sivic, Mark Everingham, Andrew Zisserman Visual Geometry Group University of Oxford The objective Retrieve all shots in a video, e.g. a feature length film, containing a particular
More informationCollection Guiding: Multimedia Collection Browsing and Visualization. Outline. Context. Searching for data
Collection Guiding: Multimedia Collection Browsing and Visualization Stéphane Marchand-Maillet Viper CVML University of Geneva marchand@cui.unige.ch http://viper.unige.ch Outline Multimedia data context
More informationObject perception by primates
Object perception by primates How does the visual system put together the fragments to form meaningful objects? The Gestalt approach The whole differs from the sum of its parts and is a result of perceptual
More informationImage Features: Local Descriptors. Sanja Fidler CSC420: Intro to Image Understanding 1/ 58
Image Features: Local Descriptors Sanja Fidler CSC420: Intro to Image Understanding 1/ 58 [Source: K. Grauman] Sanja Fidler CSC420: Intro to Image Understanding 2/ 58 Local Features Detection: Identify
More information8/3/2017. Contour Assessment for Quality Assurance and Data Mining. Objective. Outline. Tom Purdie, PhD, MCCPM
Contour Assessment for Quality Assurance and Data Mining Tom Purdie, PhD, MCCPM Objective Understand the state-of-the-art in contour assessment for quality assurance including data mining-based techniques
More informationLecture on Modeling Tools for Clustering & Regression
Lecture on Modeling Tools for Clustering & Regression CS 590.21 Analysis and Modeling of Brain Networks Department of Computer Science University of Crete Data Clustering Overview Organizing data into
More informationExpectation Maximization!
Expectation Maximization! adapted from: Doug Downey and Bryan Pardo, Northwestern University and http://www.stanford.edu/class/cs276/handouts/lecture17-clustering.ppt Steps in Clustering Select Features
More informationUsing the Deformable Part Model with Autoencoded Feature Descriptors for Object Detection
Using the Deformable Part Model with Autoencoded Feature Descriptors for Object Detection Hyunghoon Cho and David Wu December 10, 2010 1 Introduction Given its performance in recent years' PASCAL Visual
More informationLast week. Multi-Frame Structure from Motion: Multi-View Stereo. Unknown camera viewpoints
Last week Multi-Frame Structure from Motion: Multi-View Stereo Unknown camera viewpoints Last week PCA Today Recognition Today Recognition Recognition problems What is it? Object detection Who is it? Recognizing
More informationCHAPTER 6 PROPOSED HYBRID MEDICAL IMAGE RETRIEVAL SYSTEM USING SEMANTIC AND VISUAL FEATURES
188 CHAPTER 6 PROPOSED HYBRID MEDICAL IMAGE RETRIEVAL SYSTEM USING SEMANTIC AND VISUAL FEATURES 6.1 INTRODUCTION Image representation schemes designed for image retrieval systems are categorized into two
More informationSYDE Winter 2011 Introduction to Pattern Recognition. Clustering
SYDE 372 - Winter 2011 Introduction to Pattern Recognition Clustering Alexander Wong Department of Systems Design Engineering University of Waterloo Outline 1 2 3 4 5 All the approaches we have learned
More informationDisguised Face Identification (DFI) with Facial KeyPoints using Spatial Fusion Convolutional Network. Nathan Sun CIS601
Disguised Face Identification (DFI) with Facial KeyPoints using Spatial Fusion Convolutional Network Nathan Sun CIS601 Introduction Face ID is complicated by alterations to an individual s appearance Beard,
More informationA NOVEL FEATURE EXTRACTION METHOD BASED ON SEGMENTATION OVER EDGE FIELD FOR MULTIMEDIA INDEXING AND RETRIEVAL
A NOVEL FEATURE EXTRACTION METHOD BASED ON SEGMENTATION OVER EDGE FIELD FOR MULTIMEDIA INDEXING AND RETRIEVAL Serkan Kiranyaz, Miguel Ferreira and Moncef Gabbouj Institute of Signal Processing, Tampere
More informationApplications. Foreground / background segmentation Finding skin-colored regions. Finding the moving objects. Intelligent scissors
Segmentation I Goal Separate image into coherent regions Berkeley segmentation database: http://www.eecs.berkeley.edu/research/projects/cs/vision/grouping/segbench/ Slide by L. Lazebnik Applications Intelligent
More informationAnalysis: TextonBoost and Semantic Texton Forests. Daniel Munoz Februrary 9, 2009
Analysis: TextonBoost and Semantic Texton Forests Daniel Munoz 16-721 Februrary 9, 2009 Papers [shotton-eccv-06] J. Shotton, J. Winn, C. Rother, A. Criminisi, TextonBoost: Joint Appearance, Shape and Context
More informationAn Overview of Mathematics 6
An Overview of Mathematics 6 Number (N) read, write, represent, and describe numbers greater than one million and less than one-thousandth using symbols, expressions, expanded notation, decimal notation,
More informationClassifying Images with Visual/Textual Cues. By Steven Kappes and Yan Cao
Classifying Images with Visual/Textual Cues By Steven Kappes and Yan Cao Motivation Image search Building large sets of classified images Robotics Background Object recognition is unsolved Deformable shaped
More informationProbabilistic Facial Feature Extraction Using Joint Distribution of Location and Texture Information
Probabilistic Facial Feature Extraction Using Joint Distribution of Location and Texture Information Mustafa Berkay Yilmaz, Hakan Erdogan, Mustafa Unel Sabanci University, Faculty of Engineering and Natural
More informationConsistent Line Clusters for Building Recognition in CBIR
Consistent Line Clusters for Building Recognition in CBIR Yi Li and Linda G. Shapiro Department of Computer Science and Engineering University of Washington Seattle, WA 98195-250 shapiro,yi @cs.washington.edu
More informationTexton Clustering for Local Classification using Scene-Context Scale
Texton Clustering for Local Classification using Scene-Context Scale Yousun Kang Tokyo Polytechnic University Atsugi, Kanakawa, Japan 243-0297 Email: yskang@cs.t-kougei.ac.jp Sugimoto Akihiro National
More informationRough Feature Selection for CBIR. Outline
Rough Feature Selection for CBIR Instructor:Dr. Wojciech Ziarko presenter :Aifen Ye 19th Nov., 2008 Outline Motivation Rough Feature Selection Image Retrieval Image Retrieval with Rough Feature Selection
More informationContents III. 1 Introduction 1
III Contents 1 Introduction 1 2 The Parametric Distributional Clustering Model 5 2.1 The Data Acquisition Process.................... 5 2.2 The Generative Model........................ 8 2.3 The Likelihood
More informationEfficient Content Based Image Retrieval System with Metadata Processing
IJIRST International Journal for Innovative Research in Science & Technology Volume 1 Issue 10 March 2015 ISSN (online): 2349-6010 Efficient Content Based Image Retrieval System with Metadata Processing
More informationComputer vision: models, learning and inference. Chapter 10 Graphical Models
Computer vision: models, learning and inference Chapter 10 Graphical Models Independence Two variables x 1 and x 2 are independent if their joint probability distribution factorizes as Pr(x 1, x 2 )=Pr(x
More informationSegmentation: Clustering, Graph Cut and EM
Segmentation: Clustering, Graph Cut and EM Ying Wu Electrical Engineering and Computer Science Northwestern University, Evanston, IL 60208 yingwu@northwestern.edu http://www.eecs.northwestern.edu/~yingwu
More informationLecture 7: Decision Trees
Lecture 7: Decision Trees Instructor: Outline 1 Geometric Perspective of Classification 2 Decision Trees Geometric Perspective of Classification Perspective of Classification Algorithmic Geometric Probabilistic...
More informationAutomatically Improving 3D Neuron Segmentations for Expansion Microscopy Connectomics. by Albert Gerovitch
Automatically Improving 3D Neuron Segmentations for Expansion Microscopy Connectomics by Albert Gerovitch 1 Abstract Understanding the geometry of neurons and their connections is key to comprehending
More informationSnakes, level sets and graphcuts. (Deformable models)
INSTITUTE OF INFORMATION AND COMMUNICATION TECHNOLOGIES BULGARIAN ACADEMY OF SCIENCE Snakes, level sets and graphcuts (Deformable models) Centro de Visión por Computador, Departament de Matemàtica Aplicada
More information8. Automatic Content Analysis
8. Automatic Content Analysis 8.1 Statistics for Multimedia Content Analysis 8.2 Basic Parameters for Video Analysis 8.3 Deriving Video Semantics 8.4 Basic Parameters for Audio Analysis 8.5 Deriving Audio
More informationELL 788 Computational Perception & Cognition July November 2015
ELL 788 Computational Perception & Cognition July November 2015 Module 6 Role of context in object detection Objects and cognition Ambiguous objects Unfavorable viewing condition Context helps in object
More informationCS Introduction to Data Mining Instructor: Abdullah Mueen
CS 591.03 Introduction to Data Mining Instructor: Abdullah Mueen LECTURE 8: ADVANCED CLUSTERING (FUZZY AND CO -CLUSTERING) Review: Basic Cluster Analysis Methods (Chap. 10) Cluster Analysis: Basic Concepts
More informationClustering CE-324: Modern Information Retrieval Sharif University of Technology
Clustering CE-324: Modern Information Retrieval Sharif University of Technology M. Soleymani Fall 2014 Most slides have been adapted from: Profs. Manning, Nayak & Raghavan (CS-276, Stanford) Ch. 16 What
More informationComputer Vision Patch-based Object Recognition (2)
Computer Vision Patch-based Object Recognition (2) Contents Papers on patch-based object recognition Previous class: basic idea Bayes Theorem: probability background Papers in this class Hierarchy recognition
More informationEE 701 ROBOT VISION. Segmentation
EE 701 ROBOT VISION Regions and Image Segmentation Histogram-based Segmentation Automatic Thresholding K-means Clustering Spatial Coherence Merging and Splitting Graph Theoretic Segmentation Region Growing
More informationGenerative and discriminative classification techniques
Generative and discriminative classification techniques Machine Learning and Category Representation 2014-2015 Jakob Verbeek, November 28, 2014 Course website: http://lear.inrialpes.fr/~verbeek/mlcr.14.15
More informationGrade 6 Middle School Math Solution Alignment to Oklahoma Academic Standards
6.N.1 Read, write, and represent integers and rational numbers expressed as fractions, decimals, percents, and ratios; write positive integers as products of factors; use these representations in real-world
More informationFeature LDA: a Supervised Topic Model for Automatic Detection of Web API Documentations from the Web
Feature LDA: a Supervised Topic Model for Automatic Detection of Web API Documentations from the Web Chenghua Lin, Yulan He, Carlos Pedrinaci, and John Domingue Knowledge Media Institute, The Open University
More informationMatrix Co-factorization for Recommendation with Rich Side Information HetRec 2011 and Implicit 1 / Feedb 23
Matrix Co-factorization for Recommendation with Rich Side Information and Implicit Feedback Yi Fang and Luo Si Department of Computer Science Purdue University West Lafayette, IN 47906, USA fangy@cs.purdue.edu
More informationImage segmentation. Václav Hlaváč. Czech Technical University in Prague
Image segmentation Václav Hlaváč Czech Technical University in Prague Center for Machine Perception (bridging groups of the) Czech Institute of Informatics, Robotics and Cybernetics and Faculty of Electrical
More informationDirect Matrix Factorization and Alignment Refinement: Application to Defect Detection
Direct Matrix Factorization and Alignment Refinement: Application to Defect Detection Zhen Qin (University of California, Riverside) Peter van Beek & Xu Chen (SHARP Labs of America, Camas, WA) 2015/8/30
More informationImage Classification. RS Image Classification. Present by: Dr.Weerakaset Suanpaga
Image Classification Present by: Dr.Weerakaset Suanpaga D.Eng(RS&GIS) 6.1 Concept of Classification Objectives of Classification Advantages of Multi-Spectral data for Classification Variation of Multi-Spectra
More informationLecture 6: Multimedia Information Retrieval Dr. Jian Zhang
Lecture 6: Multimedia Information Retrieval Dr. Jian Zhang NICTA & CSE UNSW COMP9314 Advanced Database S1 2007 jzhang@cse.unsw.edu.au Reference Papers and Resources Papers: Colour spaces-perceptual, historical
More informationIMPROVING THE PERFORMANCE OF CONTENT-BASED IMAGE RETRIEVAL SYSTEMS WITH COLOR IMAGE PROCESSING TOOLS
IMPROVING THE PERFORMANCE OF CONTENT-BASED IMAGE RETRIEVAL SYSTEMS WITH COLOR IMAGE PROCESSING TOOLS Fabio Costa Advanced Technology & Strategy (CGISS) Motorola 8000 West Sunrise Blvd. Plantation, FL 33322
More informationData Clustering Hierarchical Clustering, Density based clustering Grid based clustering
Data Clustering Hierarchical Clustering, Density based clustering Grid based clustering Team 2 Prof. Anita Wasilewska CSE 634 Data Mining All Sources Used for the Presentation Olson CF. Parallel algorithms
More informationLinear combinations of simple classifiers for the PASCAL challenge
Linear combinations of simple classifiers for the PASCAL challenge Nik A. Melchior and David Lee 16 721 Advanced Perception The Robotics Institute Carnegie Mellon University Email: melchior@cmu.edu, dlee1@andrew.cmu.edu
More informationAS STORAGE and bandwidth capacities increase, digital
974 IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 13, NO. 7, JULY 2004 Concept-Oriented Indexing of Video Databases: Toward Semantic Sensitive Retrieval and Browsing Jianping Fan, Hangzai Luo, and Ahmed
More informationIntroduction to digital image classification
Introduction to digital image classification Dr. Norman Kerle, Wan Bakx MSc a.o. INTERNATIONAL INSTITUTE FOR GEO-INFORMATION SCIENCE AND EARTH OBSERVATION Purpose of lecture Main lecture topics Review
More informationUnsupervised Learning
Unsupervised Learning Pierre Gaillard ENS Paris September 28, 2018 1 Supervised vs unsupervised learning Two main categories of machine learning algorithms: - Supervised learning: predict output Y from
More informationCHAPTER 6 QUANTITATIVE PERFORMANCE ANALYSIS OF THE PROPOSED COLOR TEXTURE SEGMENTATION ALGORITHMS
145 CHAPTER 6 QUANTITATIVE PERFORMANCE ANALYSIS OF THE PROPOSED COLOR TEXTURE SEGMENTATION ALGORITHMS 6.1 INTRODUCTION This chapter analyzes the performance of the three proposed colortexture segmentation
More informationMarkov Networks in Computer Vision
Markov Networks in Computer Vision Sargur Srihari srihari@cedar.buffalo.edu 1 Markov Networks for Computer Vision Some applications: 1. Image segmentation 2. Removal of blur/noise 3. Stereo reconstruction
More informationIJREAT International Journal of Research in Engineering & Advanced Technology, Volume 1, Issue 5, Oct-Nov, 2013 ISSN:
Semi Automatic Annotation Exploitation Similarity of Pics in i Personal Photo Albums P. Subashree Kasi Thangam 1 and R. Rosy Angel 2 1 Assistant Professor, Department of Computer Science Engineering College,
More informationRobust Shape Retrieval Using Maximum Likelihood Theory
Robust Shape Retrieval Using Maximum Likelihood Theory Naif Alajlan 1, Paul Fieguth 2, and Mohamed Kamel 1 1 PAMI Lab, E & CE Dept., UW, Waterloo, ON, N2L 3G1, Canada. naif, mkamel@pami.uwaterloo.ca 2
More informationFlat Clustering. Slides are mostly from Hinrich Schütze. March 27, 2017
Flat Clustering Slides are mostly from Hinrich Schütze March 7, 07 / 79 Overview Recap Clustering: Introduction 3 Clustering in IR 4 K-means 5 Evaluation 6 How many clusters? / 79 Outline Recap Clustering:
More informationTopological Mapping. Discrete Bayes Filter
Topological Mapping Discrete Bayes Filter Vision Based Localization Given a image(s) acquired by moving camera determine the robot s location and pose? Towards localization without odometry What can be
More informationComputer Vision I - Filtering and Feature detection
Computer Vision I - Filtering and Feature detection Carsten Rother 30/10/2015 Computer Vision I: Basics of Image Processing Roadmap: Basics of Digital Image Processing Computer Vision I: Basics of Image
More informationAn Enhanced Image Retrieval Using K-Mean Clustering Algorithm in Integrating Text and Visual Features
An Enhanced Image Retrieval Using K-Mean Clustering Algorithm in Integrating Text and Visual Features S.Najimun Nisha 1, Mrs.K.A.Mehar Ban 2, 1 PG Student, SVCET, Puliangudi. najimunnisha@yahoo.com 2 AP/CSE,
More informationClassification. Vladimir Curic. Centre for Image Analysis Swedish University of Agricultural Sciences Uppsala University
Classification Vladimir Curic Centre for Image Analysis Swedish University of Agricultural Sciences Uppsala University Outline An overview on classification Basics of classification How to choose appropriate
More informationINF 4300 Classification III Anne Solberg The agenda today:
INF 4300 Classification III Anne Solberg 28.10.15 The agenda today: More on estimating classifier accuracy Curse of dimensionality and simple feature selection knn-classification K-means clustering 28.10.15
More informationCS443: Digital Imaging and Multimedia Binary Image Analysis. Spring 2008 Ahmed Elgammal Dept. of Computer Science Rutgers University
CS443: Digital Imaging and Multimedia Binary Image Analysis Spring 2008 Ahmed Elgammal Dept. of Computer Science Rutgers University Outlines A Simple Machine Vision System Image segmentation by thresholding
More informationLec 08 Feature Aggregation II: Fisher Vector, Super Vector and AKULA
Image Analysis & Retrieval CS/EE 5590 Special Topics (Class Ids: 44873, 44874) Fall 2016, M/W 4-5:15pm@Bloch 0012 Lec 08 Feature Aggregation II: Fisher Vector, Super Vector and AKULA Zhu Li Dept of CSEE,
More informationMarkov Networks in Computer Vision. Sargur Srihari
Markov Networks in Computer Vision Sargur srihari@cedar.buffalo.edu 1 Markov Networks for Computer Vision Important application area for MNs 1. Image segmentation 2. Removal of blur/noise 3. Stereo reconstruction
More informationContent-Based Image Retrieval Readings: Chapter 8:
Content-Based Image Retrieval Readings: Chapter 8: 8.1-8.4 Queries Commercial Systems Retrieval Features Indexing in the FIDS System Lead-in to Object Recognition 1 Content-based Image Retrieval (CBIR)
More informationCluster Evaluation and Expectation Maximization! adapted from: Doug Downey and Bryan Pardo, Northwestern University
Cluster Evaluation and Expectation Maximization! adapted from: Doug Downey and Bryan Pardo, Northwestern University Kinds of Clustering Sequential Fast Cost Optimization Fixed number of clusters Hierarchical
More informationEfficient Acquisition of Human Existence Priors from Motion Trajectories
Efficient Acquisition of Human Existence Priors from Motion Trajectories Hitoshi Habe Hidehito Nakagawa Masatsugu Kidode Graduate School of Information Science, Nara Institute of Science and Technology
More informationClustering Relational Data using the Infinite Relational Model
Clustering Relational Data using the Infinite Relational Model Ana Daglis Supervised by: Matthew Ludkin September 4, 2015 Ana Daglis Clustering Data using the Infinite Relational Model September 4, 2015
More informationVideo Summarization Using MPEG-7 Motion Activity and Audio Descriptors
Video Summarization Using MPEG-7 Motion Activity and Audio Descriptors Ajay Divakaran, Kadir A. Peker, Regunathan Radhakrishnan, Ziyou Xiong and Romain Cabasson Presented by Giulia Fanti 1 Overview Motivation
More informationQuantitative Assessment of Composition in Art
NICOGRAPH International 202, pp. 80-85 Quantitative Assessment of Composition in Art Sachi URANO Junichi HOSHINO University of Tsukuba s0853 (at) u.tsukuba.ac.jp Abstract We present a new method to evaluate
More information