Download presentation
Presentation is loading. Please wait.
Published byVictor Davidson Modified over 9 years ago
1
Multimedia Retrieval
2
Outline Overview Indexing Multimedia Generative Models & MMIR –Probabilistic Retrieval –Language models, GMMs Experiments –Corel experiments –TREC Video benchmark
3
Indexing Multimedia
4
A Wealth of Information Speech Audio Images Temporal composition Database
5
Associated Information id gender name country history picture Player Profile Biography
6
User Interaction Database Views results Gives examples video segments Poses a query query text feedback Evaluates
7
Indexing Multimedia Manually added descriptions –‘Metadata’ Analysis of associated data –Speech, captions, OCR, … Content-based retrieval –Approximate retrieval –Domain-specific techniques
8
Limitations of Metadata Vocabulary problem –Dark vs. somber Different people describe different aspects –Dark vs. evening
9
Limitations of Metadata Encoding Specificity Problem –A single person describes different aspects in different situations Many aspects of multimedia simply cannot be expressed unambiguously –Processes in left (analytic, verbal) vs. right brain (aesthetics, synthetic, nonverbal)
10
Approximate Retrieval Based on similarity –Find all objects that are similar to this one –Distance function –Representations capture some (syntactic) meaning of the object ‘Query by Example’ paradigm
11
N-dimensional space Feature extraction Ranking Display
12
Low-level Features
14
Query image
15
So, … Retrieval?!
16
IR is about satisfying vague information needs provided by users, (imprecisely specified in ambiguous natural language) by satisfying them approximately against information provided by authors (specified in the same ambiguous natural language) Smeaton
17
No ‘Exact’ Science! Evaluation is not done analytically, but experimentally –real users (specifying requests) –test collections (real document collections) –benchmarks (TREC: text retrieval conference) –Precision –Recall –...
18
Known Item
19
Query
20
Results …
21
Query
22
Results …
23
Semantic gap… raw multimedia data features concepts ?
24
Observation Automatic approaches are successful under two conditions: –the query example is derived from the same source as the target objects –a domain-specific detector is at hand
25
1. Generic Detectors
26
Retrieval Process Database Query Parsing Detector / Feature selection Filtering Ranking Query type Nouns Adjectives Camera operations People, Names Natural/physical objects Invariant color spaces........
27
Parameterized detectors People detector Example Query text Topic 41 Results
28
Query Parsing Other examples of overhead zooming in views of canyons in the Western United States Nouns Adjectives The query type Find + names
29
Detectors Camera operations (pan, zoom, tilt, …) People (face based) Names (VideoOCR) Natural objects (color space selection) Physical objects (color space selection) Monologues (specifically designed) Press conferences (specifically designed) Interviews (specifically designed) Domain specific detectors FOCUSFOCUS The universe and everything
30
2. Domain knowledge
31
Player Segmentation Original image Initial segmentation Final segmentation
32
Show clips from tennis matches, starring Sampras, playing close to the net; Advanced Queries
33
3. Get to know your users
34
Mirror Approach Gather User’s Knowledge –Introduce semi-automatic processes for selection and combination of feature models Local Information –Relevance feedback from a user Global Information –Thesauri constructed from all users
35
N-dimensional space Feature extraction ClusteringRanking Concepts Display Thesauri
36
Low-level Features
37
Identify Groups
38
Representation Groups of feature vectors are conceptually equivalent to words in text retrieval So, techniques from text retrieval can now be applied to multimedia data as if these were text!
39
Query Formulation Clusters are internal representations, not suited for user interaction Use automatic query formulation based on global information (thesaurus) and local information (user feedback)
40
Interactive Query Process Select relevant clusters from thesaurus Search collection Improve results by adapting query –Remove clusters occuring in irrelevant images –Add clusters occuring in relevant images
41
Assign Semantics
42
Visual Thesaurus Correct cluster representing ‘Tree’, ‘Forest’ Glcm_47 ‘Incoherent’ cluster Fractal_23 Mis-labeled cluster Gabor_20
43
Learning Short-term: Adapt query to better reflect this user’s information need Long-term: Adapt thesaurus and clustering to improve system for all users
44
Thesaurus Only After Feedback
45
4. Nobody is unique!
46
Collaborative Filtering Also: social information filtering –Compare user judgments –Recommend differences between similar users People’s tastes are not randomly distributed You are what you buy (Amazon)
47
Collaborative Filtering Benefits over content-based approach –Overcomes problems with finding suitable features to represent e.g. art, music –Serendipity –Implicit mechanism for qualitative aspects like style Problems: large groups, broad domains
48
5. Ask for help
49
Query Articulation N-dimensional space Feature extraction How to articulate the query?
50
What is the query semantics ?
51
Details matter
52
Problem Statement Feature vectors capture ‘global’ aspects of the whole image Overall image characteristics dominate the feature-vectors Hypothesis: users are interested in details
53
Irrelevant Background QueryResult
54
Image Spots Image-spots articulate desired image details –Foreground/background colors –Colors forming ‘shapes’ –Enclosure of shapes by background colors Multi-spot queries define the spatial relations between a number of spots
55
Hist16HistSpotSpot+Hist 5968656319214 6274706222 6098710744 5953688833 6612703411 Results Query Images
56
A: Simple Spot Query `Black sky’
57
B: Articulated Multi-Spot Query `Black sky’ above `Monochrome ground’
58
C: Histogram Search in `Black Sky’ images 2-4: 14:
59
Complicating Factors What are Good Feature Models? What are Good Ranking Functions? Queries are Subjective!
60
Probabilistic Approaches
61
Generative Models… A statistical model for generating data –Probability distribution over samples in a given ‘language’ M P ( | M )= P ( | M ) P ( | M, ) © Victor Lavrenko, Aug. 2002 aka ‘Language Modelling’
62
Basic question: –What is the likelihood that this document is relevant to this query? P(rel|I,Q) = P(I,Q|rel)P(rel) / P(I,Q) … in Information Retrieval P(I,Q|rel) = P(Q|I,rel)P(I|rel)
63
‘Language Modelling’ Not just ‘English’ But also, the language of –author –newspaper –text document –image Hiemstra or Robertson? ‘Parsimonious language models explicitly address the relation between levels of language models that are typically used for smoothing.’
64
‘Language Modelling’ Guardian or Times?Not just ‘English’ But also, the language of –author –newspaper –text document –image
65
‘Language Modelling’ or ?Not just English! But also, the language of –author –newspaper –text document –image
66
Unigram and higher-order models Unigram Models N-gram Models Other Models –Grammar-based models, etc. –Mixture models = P ( ) P ( | ) P ( ) P ( ) P ( ) P ( ) P ( | ) P ( | ) P ( | ) © Victor Lavrenko, Aug. 2002
67
The fundamental problem Usually we don’t know the model M –But have a sample representative of that model First estimate a model from a sample Then compute the observation probability P ( | M ( ) ) M © Victor Lavrenko, Aug. 2002
68
Indexing: determine models Indexing –Estimate Gaussian Mixture Models from images using EM –Based on feature vector with colour, texture and position information from pixel blocks –Fixed number of components DocsModels
69
Retrieval: use query likelihood Query: Which of the models is most likely to generate these 24 samples?
70
Probabilistic Image Retrieval ?
71
Query Rank by P(Q|M) P(Q|M 1 ) P(Q|M 4 ) P(Q|M 3 ) P(Q|M 2 )
72
Topic Models Models P(Q|M 1 ) P(Q|M 4 ) P(Q|M 3 ) P(Q|M 2 ) Query P(D 1 |QM) P(D 2 |QM) P(D 3 |QM) P(D 4 |QM) Documents Query Model
73
Probabilistic Retrieval Model Text –Rank using probability of drawing query terms from document models Images –Rank using probability of drawing query blocks from document models Multi-modal –Rank using joint probability of drawing query samples from document models
74
Unigram Language Models (LM) –Urn metaphor Text Models P( ) ~ P ( ) P ( ) P ( ) P ( ) = 4/9 * 2/9 * 4/9 * 3/9 © Victor Lavrenko, Aug. 2002
75
Generative Models and IR Rank models (documents) by probability of generating the query Q: P( | ) = 4/9 * 2/9 * 4/9 * 3/9 = 96/9 P( | ) = 3/9 * 3/9 * 3/9 * 3/9 = 81/9 P( | ) = 2/9 * 3/9 * 2/9 * 4/9 = 48/9 P( | ) = 2/9 * 5/9 * 2/9 * 2/9 = 40/9
76
The Zero-frequency Problem Suppose some event not in our example –Model will assign zero probability to that event –And to any set of events involving the unseen event Happens frequently with language It is incorrect to infer zero probabilities –Especially when dealing with incomplete samples ?
77
Smoothing Idea: shift part of probability mass to unseen events Interpolation with background (General English) –Reflects expected frequency of events –Plays role of IDF – +(1- )
78
Hierarchical Language Model MNM Smoothed over multiple levels Alpha * P(T|Shot) + Beta * P(T|‘Scene’) + Gamma * P(T|Video) + (1–Alpha–Beta–Gamma) * P(T|Collection) Also common in XML retrieval –Element score smoothed with containing article
79
Image Models Urn metaphor not useful –Drawing pixels useless Pixels carry no semantics –Drawing pixel blocks not effective chances of drawing exact query blocks from document slim Use Gaussian Mixture Models (GMM) –Fixed number of Gaussian components/clusters/concepts
80
Key-frame representation Query model split colour channels Take samples Cr Cb Y DCT coefficients position EM algorithm 675912111941517-9-30001850154014-211 6617135-511315362-4011084454-201 12 668-7133-3015340-5000083733-30-2113 6651011245215340-500008290300014 669-5187-31-515340-50000833-5403 15
81
? Image Models Expectation-Maximisation (EM) algorithm –iteratively estimate component assignments re-estimate component parameters
82
Component 1Component 2Component 3 Expectation Maximization E M
83
animation Component 1Component 2Component 3 E M
84
VisualSEEk Querying by image regions and spatial laylout –Joint content-based/spatial querying –Automated region extraction –Direct indexing of color feature
85
Image Query Process
86
Color Similarity Color space similarity –measure the closeness in the HSV color space Color histograms distance –Minkowski metric between h q and h t Histogram quadratic distance –used in QBIC project –measure the weighted similarity between histograms –compute the cross similarity between colors It is computational expensive
87
Color sets –A compact alternative to color histograms Color set examples –Transform RGB to HSV Quantize the HSV color space to 2 hues, 2 saturations, and 2 values Assign a unique index m to each quantized HSV color => right dimensional binary space color set: a selection from the eight colors
88
Color sets and Back-Projection Process stage –Color selection 1, 2 colors, …, until extract salient regions –Back-projection onto the image map I[x,y] to the most similar color in the color set –Thresholding –Labeling
90
Color Set Query Strategy Given the query color set –perform several range queries on the query color set’s colors –take the intersection of these lists –minimize the sum of attributes in the intersection list
91
Single Region Query Region absolute location –fixed query location The Euclidean distance of centroids –bounded query location fall within a designed area: d q,t =0 otherwise: the Euclidean distance of the centroids
92
Index Structure Centroid location spatial access –Spatial quad-tree Rectangle (MBR) location spatial access –R-trees
93
Size Comparison Area –The absolute distance between two regions Spatial Extent –measure the distance among the width and the height of the MBRs –is much simpler than shape information
94
Single Region Query Strategy
95
Multiple Regions Query
96
Region Relative Location Use 2D-string
97
Spatial Invariance Provide scaling/rotation –approximate rotation invariance by providing different projections
98
Multiple Regions Query Strategy Relatives locations –perform query on all attributes except location –find the intersection of the region lists –for each candidate image the 2D-string is generated compared to the 2D-string of the query image
99
Query Formulation User tools –sketches regions –position them on the query grid –assign colors, size, and absolute location –may assign boundaries
100
Query Examples(1)
101
Query Examples(2)
102
Query Examples(3)
103
Bolbworld Image is treated as a few “blobs” –Image regions are roughly homogeneous with respect color and texture
104
Grouping Pixels into Regions For each pixel –assign a vector consisting of color, texture, and position features (8-D space) Model the distribution of pixels –Use EM algorithm to fit the mixture of Gaussians model to the data Perform spatial grouping –Connected pixels belonging to the same color/texture cluster
105
Describe the Regions For each region –store its color histogram Match the color of two regions –Use the quadratic distance between their histograms x and y –d 2 hist (x,y) = (x-y) T A(x-y) –Matrix A a symmetric matrix of weights between 0 and 1 represent the similarity between bin i and bin j
106
Querying in Blobworld The user composes a query –By submitting an image see its Blobworld representation select the relevant blobs to match specify the relative importance of the blob features –Atomic query Specify a particular blob with feature vector v i to match For each blob v j in the database image Find the distance between v i and v j –Measure the similarity between b i and b j –Take the maximum value
107
Querying in Blobworld Compound query –calculated using fuzzy-logic operator –user may specify a weight for each atomic query –rank the images according to the overall score indicate the blobs provided the highest score helps the user refine the query –Change the weighting of blob features –Specify new blobs to match
109
Indexing of Blobworld Indexing the color feature vectors –to speed up atomic queries –use R*-trees –higher dimensional data require larger data entries => lower fanout –Need a low dimensional approximation to the full color feature vectors use SVM to find A k : the best rank k approximation to the weight matrix A project the feature space into the subspace spanned by the rows of A k index the low-dimensional vector
110
Multimedia Cross-model Correlation Case study: automatic image captioning –Each image has a set of extracted regions –Some of the regions have a caption
111
Feature Extraction Discover image regions –Extracted by a standard segmentation algorithm –Each region is mapped into a 30-dim feature vector Mean, standard deviation of RGB values, average responses to various texture filters, position in the entire image layout, shape description How to capture cross-media correlations?
112
Mixed Media Graph (MMG) Graph construction –V(O): the vertex of object O –V(a i ): the vertex of the attribute value A= a i –Add an edge if and only if the two token values are close enough for each feature-vector, choose its k nearest neighbors => add the corresponding edges –the nearest neighbor relationship is not symmetric NN-links: the nearest neighbor links OAV-links: the object-attribute-value-links
113
Example of MMG
114
Correlation Discovery Correlation discovery by random walk –random walk with restart –To compute the affinity of node “B” for node “A” a random walker starts from node “A” choose randomly among the available edges every time –with probability c to go back to node “A” (restart) the steady-state probability that the random walker will find at node “B) –the affinity of “B” with respect to “A”
115
RWR Algorithm Given a query object O q –do an RWR from node q=V(O q ) –compute the steady state probability vector u q =(u q (1), …, u q (N)). Example –Given an image I 3 –Estimate u i3 for all nodes in the G MMG –Report the top few caption words for the image
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.