Information-theoretic Computer Vision for Autonomous Robots

download Information-theoretic Computer Vision for Autonomous Robots

of 94

Transcript of Information-theoretic Computer Vision for Autonomous Robots

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    1/94

    Information-theoretic Computer Visionfor Autonomous Robots

    Boyan Bonev

    Robot Vision GroupUniversity of Alicante

    November 26th, 2010

    Talk at the Max Planck Institute for Biological Cybernetics,Tubingen, Germany

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 1 / 94

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    2/94

    Outline

    1 IntroductionThe Robot Vision GroupResearchInformation theory

    2 TheoriesThe Method of Types

    3 MeasuresMutual information for alignmentEntropy in visual navigationJensen-Renyi divergence

    4 PrinciplesMinimum description lengthMinimax entropy

    5 Entropy EstimationFeature Selection

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 2 / 94

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    3/94

    Outline

    1 IntroductionThe Robot Vision GroupResearchInformation theory

    2 TheoriesThe Method of Types

    3 MeasuresMutual information for alignmentEntropy in visual navigationJensen-Renyi divergence

    4 PrinciplesMinimum description lengthMinimax entropy

    5 Entropy EstimationFeature Selection

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 3 / 94

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    4/94

    University of Alicante

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 4 / 94

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    5/94

    The Robot Vision Group - People

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 5 / 94

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    6/94

    The Robot Vision Group - Research

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 6 / 94

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    7/94

    The Robot Vision Group - Platforms

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 7 / 94

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    8/94

    The Robot Vision Group - Mobile

    Bench Project: Collaboration with James Coughlan, Smith Kettlewel Eye Research Institute (California)

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 8 / 94

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    9/94

    Outline

    1 IntroductionThe Robot Vision Group

    ResearchInformation theory

    2 TheoriesThe Method of Types

    3 MeasuresMutual information for alignmentEntropy in visual navigationJensen-Renyi divergence

    4 PrinciplesMinimum description lengthMinimax entropy

    5 Entropy EstimationFeature Selection

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 9 / 94

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    10/94

    Former research (1)

    Quadruped walk calibration

    Video

    Bonev, Cazorla, Martnez (2005)

    Walk calibration in a four-legged robot. Climbing and Walking Robots, London, U.K.

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 10 / 94

    http://video/robocup.wmvhttp://video/robocup.wmvhttp://video/robocup.wmv
  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    11/94

    Former research (2)

    Localization

    0

    500

    1000

    1500

    2000

    2500

    -1500 -1000 -500 0 500 1000 1500

    Y(mm)

    X (mm)

    Path followed

    Ground truthEstimated

    0

    500

    1000

    1500

    2000

    2500

    -1500 -1000 -500 0 500 1000 1500

    Y(mm

    )

    X (mm)

    Path followed

    Ground truthEstimated

    speedmeans+errors(odometry)

    0

    500

    1000

    1500

    2000

    2500

    -1500 -1000 -500 0 500 1000 1500

    Y(mm)

    X (mm)

    Path followed

    Ground truthEstimated

    0

    500

    1000

    1500

    2000

    2500

    -1500 -1000 -500 0 500 1000 1500

    Y(mm

    )

    X (mm)

    Path followed

    Ground truthEstimated

    Bonev, Cazorla, Martn, Matellan (2010)

    Portable autonomous walk calibration for 4-legged robots. Applied Intelligence

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 11 / 94

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    12/94

    Former research (3)

    Architecture and robotic tasks

    Commander

    Perceptual

    Anchoring

    Module

    Hierarchical

    Behaviour

    Module

    Global

    Map

    Hierarchical

    FiniteState Machine

    Team

    Communication

    Module

    Lower Layer

    Middle Layer

    Higher Layer

    Communication

    Layer

    Sensor

    Data

    Motor

    Commands

    Local State

    Local State

    Global State

    Behaviours

    Messages

    Other

    Robot

    Other

    Robot

    Probability maps

    Martnez, Matellan, Cazorla, Saffiotti, Herrero, Martn, Bonev, LeBlanc (2005)

    Team Chaos description paper. RoboCup (Competition), Osaka, Japan

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 12 / 94

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    13/94

    Former research (4)

    Teamwork

    University of Murcia, University Rey Juan Carlos (Madrid), University of Alicante, Orebro University (Sweden)

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 13 / 94

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    14/94

    Motivation

    From controlled, constrained, laboratory environments

    To different indoor/outdoor environments

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 14 / 94

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    15/94

    Ph.D. Thesis

    B. Bonev (2010)Feature Selection based onInformation Theory

    Supervised by M. Cazorlaand F. Escolano

    Estimation of

    mutual information To optimize (for classification)

    a high-dimensional set offeatures:

    Image filters Spectral graph features Genes

    0 2 4 6 8 10 1 2 1 4 16 1 8 200

    2

    4

    6

    8

    10

    12

    14

    16

    18

    20

    Fluorescent

    labeling

    Sample RNA

    Referece ADNc

    Combination

    Hybridization

    Fluorescent

    labeling

    MDF

    eature

    Sele

    ction

    Num

    ber

    of

    Sele

    ctedGen

    e

    Class (disease)MELANOMA

    MELANOMA

    MELANOMA

    MELANOMA

    MELANOMA

    MELANOMA

    BREAST

    BREAST

    MELANOMA

    NSCLC

    NSCLC

    NSCLC

    BREAST

    MCF7Drepro

    BREAST

    MCF7Arepro

    COLON

    COLON

    COLON

    COLON

    COLON

    COLON

    COLON

    LEUKEMIA

    LEUKEMIA

    LEUKEMIA

    LEUKEMIA

    LEUKEMIA

    K562Arepro

    K562Brepro

    LEUKEMIA

    NSCLC

    NSCLC

    NSCLC

    PROSTATE

    OVARIAN

    OVARIAN

    OVARIAN

    OVARIAN

    OVARIAN

    PROSTATE

    MELANOMA

    OVARIAN

    UNKNOWN

    RENAL

    NSCLC

    BREAST

    RENAL

    RENAL

    RENAL

    RENAL

    RENAL

    RENAL

    RENAL

    NSCLC

    NSCLC

    BREAST

    CNS

    CNS

    BREAST

    RENAL

    CNS

    CNS

    CNS

    19135

    246663766982

    117714701671

    2080

    32273400396440574063411042894357444146634813522654815494549555085790589260136019603260456087

    6145

    61846643

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 15 / 94

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    16/94

    Outline

    1 IntroductionThe Robot Vision Group

    ResearchInformation theory

    2 TheoriesThe Method of Types

    3 MeasuresMutual information for alignmentEntropy in visual navigationJensen-Renyi divergence

    4 PrinciplesMinimum description lengthMinimax entropy

    5 Entropy EstimationFeature Selection

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 16 / 94

    I d i (1)

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    17/94

    Introduction (1)

    Information Theory

    Specifies how to encode data which obey a probabilitydistribution so that they can be transmitted and thendecoded.

    Cover and Thomas (1991)

    Elements of Information Theory. Wiley-Interscience

    Information Theory in Computer Vision

    Encoding is performed by light rays reflectedoff the objects in the scene.Depends on the reflectance properties, spatiallocations, light sources: encoding is out of ourcontrol.We can look for common structures or models.

    Yuille (2010) An information theory perspective on computational vision.

    Front. Electr. Electron. Eng. China

    M.C. Escher, Three worlds

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 17 / 94

    I d i (2)

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    18/94

    Introduction (2)

    Escolano, Suau, Bonev (2009) Information Theory in Computer Vision and Pattern Recognition. SpringerBoyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 18 / 94

    I d i (2)

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    19/94

    Introduction (2)

    Escolano, Suau, Bonev (2009) Information Theory in Computer Vision and Pattern Recognition. SpringerBoyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 19 / 94

    I t d ti (2)

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    20/94

    Introduction (2)

    Escolano, Suau, Bonev (2009) Information Theory in Computer Vision and Pattern Recognition. SpringerBoyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 20 / 94

    I t d ti (2)

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    21/94

    Introduction (2)

    Escolano, Suau, Bonev (2009) Information Theory in Computer Vision and Pattern Recognition. SpringerBoyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 21 / 94

    Introduction (2)

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    22/94

    Introduction (2)

    Escolano, Suau, Bonev (2009) Information Theory in Computer Vision and Pattern Recognition. SpringerBoyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 22 / 94

    Introduction (3)

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    23/94

    Introduction (3)

    A glance at IT in several Computer Vision and Autonomous Robotic

    tasks. Classification of the topics in 4 dimensions:

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 23 / 94

    Outline

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    24/94

    Outline

    1 IntroductionThe Robot Vision Group

    ResearchInformation theory

    2 TheoriesThe Method of Types

    3 MeasuresMutual information for alignmentEntropy in visual navigationJensen-Renyi divergence

    4

    PrinciplesMinimum description lengthMinimax entropy

    5 Entropy EstimationFeature Selection

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 24 / 94

    Outline

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    25/94

    Outline

    1 IntroductionThe Robot Vision Group

    ResearchInformation theory

    2 TheoriesThe Method of Types

    3 MeasuresMutual information for alignmentEntropy in visual navigationJensen-Renyi divergence

    4

    PrinciplesMinimum description lengthMinimax entropy

    5 Entropy EstimationFeature Selection

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 25 / 94

    The Method of Types

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    26/94

    The Method of Types

    The method of types (Csiszar and Korner) Partition the n-sized samples into classes according to their type

    (empirical distribution). There are only a polynomial number of types (wrt n). There are an exponential number of samples of each type.

    The sequence {2, 2, 6} has the type

    P(2) =23 , P(6) =

    13 , P(1) = P(3) = P(4) = P(5) = 0.

    The class type of P is the set of all sequences of length 3with two 2s and one 6:T(P) = {226, 262, 622}

    For samples drawn i.i.d. according to a distribution Q The probability of each type class depends exponentially on the

    relative entropy distance between the type P and the distribution Q Thus, type classes that are far from the true distribution have

    exponentially smaller probability.

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 26 / 94

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    27/94

    Filtering

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    28/94

    Filtering

    Finding a threshold to discard all points x whose relative entropy atSmax is (x) < .

    Pixel filtering with increasing values.

    Finding is an image-dependent task. Exploit the method of types to ensure the best filtering.

    Bonev, Escolano, Lozano, Suau, Cazorla, Aguilar (2007)Constellations and the Unsupervised Learning of Graphs .

    6th IAPR -TC-15 Workshop on Graph-based Representations in Pattern Recognition. Alicante, SpainBoyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 28 / 94

    Optimal filtering

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    29/94

    Optimal filtering

    Pon() is the pdf of the probability to have a relative entropy given

    that a point is part of the salient regions. Poff() is the pdf of the probability to have a relative entropy given

    that a point is part of the discarded regions.

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 29 / 94

    Chernoff information

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    30/94

    Chernoff information

    The Chernoff information

    C(Pon, Poff) = min01

    log J

    j=1

    Pon(xj)P1off (xj)

    ,where xj is the histogram bin j,measures how discriminable are Pon and Poff.

    The expected error rate of the likelihood test logPon()

    Poff()< T

    decreases exponentially wrt C(Pon(), Poff()). T is bound by the Kullback-Leibler divergence:D(P

    off()||P

    on()) < T < D(P

    on()||P

    off())

    A low C(Pon, Poff) means that the images in the set are tooheterogeneous less points will be discarded.

    Video

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 30 / 94

    Different thresholds

    http://video/trayectoria.avihttp://video/trayectoria.avihttp://video/trayectoria.avi
  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    31/94

    Different thresholds

    Environment C(Pon, Poff ) %Pointsoffice 0.2434 33.63%

    corridor#1 0.4491 38.58%corridor#2 0.4223 36.69%

    hall 0.2732 34.46%entrance 0.1405 29.17%

    trees-avenue 0.2279 43.00%

    Lozano, Escolano, Bonev, Suau, Aguilar, Saez, Cazorla (2008)

    Region and constellations based categorization of images with unsupervised graph learning. Image and Vision ComputingBoyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 31 / 94

    Outline

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    32/94

    1 IntroductionThe Robot Vision Group

    ResearchInformation theory

    2 TheoriesThe Method of Types

    3 MeasuresMutual information for alignmentEntropy in visual navigationJensen-Renyi divergence

    4 PrinciplesMinimum description lengthMinimax entropy

    5 Entropy EstimationFeature Selection

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 32 / 94

    Outline

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    33/94

    1 IntroductionThe Robot Vision Group

    ResearchInformation theory

    2 TheoriesThe Method of Types

    3 MeasuresMutual information for alignmentEntropy in visual navigationJensen-Renyi divergence

    4 PrinciplesMinimum description lengthMinimax entropy

    5 Entropy EstimationFeature Selection

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 33 / 94

    Image alignment

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    34/94

    g g

    Quadcopter video Vertical camera video

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 34 / 94

    Local and global approaches

    http://video/videofer.mp4http://video/videofer.mp4http://video/videowarping.mp4http://video/videowarping.mp4http://video/videowarping.mp4http://video/videofer.mp4
  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    35/94

    g pp

    Based on local features: SURF, SIFT, saliency, . . . (a problem if thereare no features or there is noise)

    Based on the global appearance: correlation, mutual information,entropy, . . . (time-consuming)

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 35 / 94

    Conditional entropy and mutual information

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    36/94

    Among the space of transformations , find a transformation T whichmaximizes some measure of the alignment between T(I2) and I1.

    Conditional entropy: arg minT

    H(T(I2)|I1) self-predictability problem

    Mutual information: arg maxT

    I (T(I2)|I1)

    = arg maxT

    {H(T(I2)) + H(I1) H(T(I2)|I1)}

    I(X,Y,Z)

    H(Y|X,Z)H(X|Y,Z)

    H(Z|X,Y)

    H(X) H(Y)

    H(Z) H(X,Y,Z)

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 36 / 94

    The histogram-binning problem

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    37/94

    Joint histogram

    0 50 100 150 200

    0

    50

    100

    150

    200

    0

    0

    50

    50

    200 200200 150

    X Y

    x

    y

    2

    1 1

    A high number of bins: sparse histogram

    sensitive to noise

    Without noise:

    10, 50 and 255 binned histograms

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 37 / 94

    The histogram-binning problem

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    38/94

    Joint histogram

    0 50 100 150 200

    0

    50

    100

    150

    200

    0

    0

    50

    50

    200 200200 150

    X Y

    x

    y

    2

    1 1

    A high number of bins: sparse histogram

    sensitive to noise

    With noise:

    10, 50 and 255 binned histograms

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 38 / 94

    The isocontours method (1)

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    39/94

    50 100 150 200 250 300

    20

    40

    60

    80

    100

    120

    140

    160

    180

    200

    220

    Image considered as a continuous surface; divided in Q iso-intensity lines.

    1

    2

    3

    4

    1

    2

    42

    1

    4

    (a) (b)

    (c)

    1

    2

    42

    1

    4

    (d)

    (e)

    a) Subpixel interpolationb) Iso-intensities intersect inside(vote)c) Iso-intensities intersect

    outsided) Iso-intensities are parallele) Intersection area ofiso-surfaces

    Rajwade, Banerjee, Rangarajan (2009)Probability density estimation using isocontours and isosurfaces: Application to information theoretic image registration.

    IEEE Transactions on Pattern Analysis and Machine IntelligenceBoyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 39 / 94

    The isocontours method (2)

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    40/94

    Classical, point-counting and area-based.

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 40 / 94

    The isocontours method (3)

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    41/94

    0

    10

    20

    30

    40

    50 0

    10

    20

    30

    40

    50

    0

    0.1

    0.2

    0.3

    0.4

    0.5

    0.6

    0.7

    0.8

    0.9

    POINT COUNTING

    0

    10

    20

    30

    40

    50 0

    10

    20

    30

    40

    50

    0.5

    1

    1.5

    2

    2.5

    AREA BETWEEN ISOCONTOURS

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 41 / 94

    Outline

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    42/94

    1 IntroductionThe Robot Vision Group

    ResearchInformation theory

    2 TheoriesThe Method of Types

    3 MeasuresMutual information for alignmentEntropy in visual navigationJensen-Renyi divergence

    4 PrinciplesMinimum description lengthMinimax entropy

    5 Entropy EstimationFeature Selection

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 42 / 94

    Omnidirectional camera

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    43/94

    Basic skills for topological navigation in a structured world:finding the direction and avoiding obstacles

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 43 / 94

    Entropy for finding the direction (1)

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    44/94

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 44 / 94

    Entropy for finding the direction (2)

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    45/94

    0 180 3600.2

    0.3

    0.4

    0.5

    0.6

    0.7

    0.8

    0.9

    1

    Angle

    Entropy

    Entropy approximation

    Entropy

    2ndorder Fourier Series Approximation

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 45 / 94

    Entropy for finding the direction (3)

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    46/94

    Indoor and outdoor results.

    0 20 40 60 80 100 120

    80

    60

    40

    20

    0

    20

    40

    60

    80

    Entropybased direction estimation

    # frames

    Angle(degrees)

    Estimated directionDesired direction

    0 50 100 150

    80

    60

    40

    20

    0

    20

    40

    60

    80

    Entropybased direction estimation

    # frames

    Angle(degrees)

    Estimated directionDesired direction

    Bonev, Cazorla, Escolano (2007)Robot Navigation Behaviors based on Omnidirectional Vision and Information Theory.

    Journal of Physical Agents

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 46 / 94

    Obstacle avoidance

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    47/94

    Visual sonars based on gradient

    maxD

    vk

    fk

    f*

    Video

    Bonev, Cazorla, Escolano (2007)Robot Navigation Behaviors based on Omnidirectional Vision and Information Theory.

    Journal of Physical Agents

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 47 / 94

    Outline

    http://video/navigation.avihttp://video/navigation.avihttp://video/navigation.avi
  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    48/94

    1 IntroductionThe Robot Vision Group

    ResearchInformation theory

    2 TheoriesThe Method of Types

    3 MeasuresMutual information for alignmentEntropy in visual navigationJensen-Renyi divergence

    4 Principles

    Minimum description lengthMinimax entropy

    5 Entropy EstimationFeature Selection

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 48 / 94

    The Jensen-Renyi divergence

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    49/94

    Jensen-Renyi divergence

    JR (p1, , pn) = H

    n

    i=1

    ipi

    ni=1

    iH(pi), p1, p2, , pn are n probability distributions

    H(p) is the Renyi entropy of order

    = (1, 2, , 3) is a weight vector satisfyingn

    i=1 i = 1 with i 0

    Symmetric

    n weighted distributions

    Robust to noise

    Renyi entropy

    H(X) =1

    1 log

    ni=1

    xi

    0 0.2 0.4 0.6 0.8 10

    0.2

    0.4

    0.6

    0.8

    1

    p

    H

    (p)

    =0

    =0.2

    =0.5

    Shannon

    =2

    =10

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 49 / 94

    Trajectory segmentation (1)

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    50/94

    Segment a sequence of images based on their distributions of low-levelfilters. Useful for topological localization and navigation.

    Region A Region B

    W1 W2

    W1 W2

    W1 W2

    W1 W2

    W1 W2

    W1 W2

    W1 W2

    Data

    J-R divergence

    JR = 0.4

    JR = 0.5

    JR = 0.6

    JR = 0.7

    JR = 0.6

    JR = 0.5

    JR = 0.4

    What window size?Bonev, Cazorla (2010)

    Large scale environment partitioning in mobile robotics recognition tasks.

    Journal of Physical Agents

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 50 / 94

    Trajectory segmentation (2)

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    51/94

    220 230 240 250 260 2702800

    10

    20

    30

    JRdivergence at various window sizes

    # image

    window size

    JRdivergence

    #241 #254 #273

    discriminative segmentation

    Bonev, Cazorla (2010)Large scale environment partitioning in mobile robotics recognition tasks.

    Journal of Physical Agents

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 51 / 94

    Localization

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    52/94

    0 50 100 150 200 2500

    100

    200

    300

    400

    Single similiarity function response

    # test image

    #referenceimag

    e

    H

    0 50 100 150 200 2500

    100

    200

    300

    400

    # test image

    #referenceimag

    e

    0 50 100 150 200 2500

    100

    200

    300

    400

    Similiarity functions responses

    # test image

    #referenceimage

    H1

    H2

    H3

    H4

    H5

    H6

    H7

    H8

    H9

    H10

    H11H12

    H13

    H14

    H15

    H16

    H17

    H18

    H19

    H20

    0 100 200 300 400

    0

    0.5

    1Particle filter, iterations 1,5,8,11

    likelihood

    0 100 200 300 400

    0

    0.5

    1

    likelihood

    0 100 200 300 400

    0

    0.5

    1

    likeliho

    od

    0 100 200 300 400

    0

    0.5

    1

    particle position (# reference image)

    likelihood

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 52 / 94

    Other IT measures

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    53/94

    Henze-Penrose divergence. Based onthe Friedman-Rafsky test (using

    spanning trees). Symmetrized Kullback-Leibler

    divergence (using k-NN).

    Jensen-Tsallis -divergence (using

    k-NN). Symmetrized and normalized entropy

    square variation (using k-NN).

    Total variation divergence (usingkd-partitions).

    Escolano, Lozano, Bonev, Suau (2010)Bypass information-theoretic shape similarity from non-rigid points-based alignment.

    Workshop on Non-Rigid Shape Analysis and Deformable Image Alignment (NORDIA), in conjunction with CVPR.

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 53 / 94

    Outline

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    54/94

    1 IntroductionThe Robot Vision Group

    ResearchInformation theory

    2 TheoriesThe Method of Types

    3

    MeasuresMutual information for alignmentEntropy in visual navigationJensen-Renyi divergence

    4 Principles

    Minimum description lengthMinimax entropy

    5 Entropy EstimationFeature Selection

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 54 / 94

    Outline

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    55/94

    1 IntroductionThe Robot Vision Group

    ResearchInformation theory

    2 TheoriesThe Method of Types

    3

    MeasuresMutual information for alignmentEntropy in visual navigationJensen-Renyi divergence

    4 Principles

    Minimum description lengthMinimax entropy

    5 Entropy EstimationFeature Selection

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 55 / 94

    Minimum description length

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    56/94

    Minimum description lenght (MDL) principle Formalization of Occams Razor.

    The best hypothesis is the one that leads to the best compression ofthe data.

    A tradeoff between the complexity of the hypothesis and thecomplexity of the data given the hypothesis (avoids overfitting).

    The MDL principle

    For any probability distribution P, it is possible to construct a code C such

    that C(x) is log2 P(x) bits long.

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 56 / 94

    Example: the EBEM algorithm

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    57/94

    Expectation-Maximization (EM) schemes have the model order selectionproblem.

    Example: the Entropy-based EM for Gaussian mixtures (generativeapproach)

    No initialization problem: starts with one Gaussian kernel, parameters

    given by the sample. Divides the kernel whose data is not Gaussian enough.

    Need for a stopping criterion: otherwise the maximum likelihoodhappens when each data point is described by one kernel (overfitting).

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 57 / 94

    EBEM iterations and MDL

    M d l d l ti H k l l ki f ?

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    58/94

    Model order selection: How many kernels are we looking for?We cannot establish a threshold without knowledge about the data.

    Escolano, Penalver, Bonev (2010)Entropy-based Variational Scheme for Fast Bayesian Learning of Gaussian Mixtures.

    Statistical, Structural and Syntactic Pattern Recognition. Cezme, Turkey

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 58 / 94

    MDL for model order selection

    f

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    59/94

    MDL for model order selection

    The optimal instance of a model of any order M is the one that minimizes

    L(D|M) + L(M).

    The problem usually is how to estimate the model and code lenghts.In the EBEM case: L(D|M) is given by the likelihood of the data D given the model M

    L(M) depends on the number of parameters of the mixtures

    Video

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 59 / 94

    Other approaches

    http://video/ebem1.wmvhttp://video/ebem1.wmvhttp://video/ebem1.wmv
  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    60/94

    Other approches related to MDL:

    AIC, An Information Criterion ofAkaike.

    BIC, Bayesian InformationCriterion of Schwarz.

    MML, Minimum MessageLength of Wallace.

    Alternatives to MDL:

    Variational EM algorithms whichdo not need a stopping criterion.

    Video

    Example of EBEM segmentation of the colour space.

    Penalver, Escolano, Saez (2010)Learning Gaussian Mixture Models with Entropy Based Criteria.

    IEEE Transactions on neural networks

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 60 / 94

    Outline

    1 I t d ti

    http://video/ebem3.wmvhttp://video/ebem3.wmvhttp://video/ebem3.wmv
  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    61/94

    1 IntroductionThe Robot Vision Group

    ResearchInformation theory

    2 TheoriesThe Method of Types

    3

    MeasuresMutual information for alignmentEntropy in visual navigationJensen-Renyi divergence

    4 Principles

    Minimum description lengthMinimax entropy

    5 Entropy EstimationFeature Selection

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 61 / 94

    The Maximum Entropy principle

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    62/94

    Maximum Entropy principle

    When learning a probability distribution from data, the most unbiased(neutral) hypothesis is the distribution with maximum entropy whichsatisfies the expectation constraints on the datas statistics.

    p() = arg maxp()

    p()log p()ds.t.

    p()Gi()d = E(Gi()) = i, i = 1, . . . , m

    p()d = 1 A way to find the hypothesis with less assumptions (maximize

    generalization).

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 62 / 94

    FRAME

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    63/94

    Filters, Random Fields and Maximum Entropy (Zhu, Wu, Mumford)

    A statistical theory for texture modeling. Textures are modeled by a general filter bank fi(), i = 1, . . . , m.

    Generative approach: a pdf of filters is learnt textures can besynthesized by sampling the pdf.

    The maximum entropy principle is used to learn the pdf: Estimates of the marginal distributions of f(I) by applying the filters tothe texture.

    Derive a maximum entropy distribution p(I) s.t. have the samemarginal distributions.

    Select a set Sm of filters by filter pursuit through minimax entropy.

    Zhu, Wu, Mumford (1997)Minimax entropy principle and its applications to texture modeling, Neural Computation.

    Neural Computation

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 63 / 94

    The Mini-Max Entropy principle

    Filter pursuit (incremental feature

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    64/94

    selection)

    Select the filter which changes morethe distribution (the less redundantwith the already selected filters).

    Minimax principle:

    The optimal set of filters should be

    chosen to minimize theKullback-Leibler divegence betweenthe filters marginals of the originaltexture and the synthesized texture.

    As f(I) is fixed, then Sm is chosensuch that p(I; m, Sm) has theminimum entropy. Thus,

    Sm = arg minSmSmaxm H(p(I))

    Zhu, Wu, Mumford c1997 MIT Press

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 64 / 94

    Outline

    1 Introduction

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    65/94

    1 IntroductionThe Robot Vision Group

    ResearchInformation theory

    2 TheoriesThe Method of Types

    3 MeasuresMutual information for alignmentEntropy in visual navigationJensen-Renyi divergence

    4 Principles

    Minimum description lengthMinimax entropy

    5 Entropy EstimationFeature Selection

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 65 / 94

    Entropy in Information Theory

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    66/94

    Entropy is a measure of the disorder or amount of information. Entropy is related to predictability with several interpretations:

    Measure of the amount of information that an event provides Measure of the uncertainty in the outcome of an event Measure of the dispersion in the probability distribution

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 66 / 94

    The problem

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    67/94

    Multivariate entropy is difficult to estimate with a high number of

    dimensions (more than 2).

    Points (n=9) following a Gaussian distribution in 1D, 2D and 3D.

    Data points get exponentially sparser

    Computational issues of the estimation algorithms

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 67 / 94

    Entropy estimation approaches

    E t ti ti

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    68/94

    Entropy estimation plug-in: Estimate the density p(x) underlying the samples X and

    plug it into the formula: H(X) =

    x p(x)log p(x) Parzens Window with variable kernel width [Viola, 1997], Wavelet

    density estimation [Peter and Rangarajan, 2008] Estimation degrades exponentially wrt # dimensions

    bypass: Bypass the density estimation and estimate entropy directly

    from the data, based on: Entropic spanning graphs [Hero and Michel, 2002], Nearest neighbors

    [Leonenko et al., 2008], k-d partitioning [Stowell and Plumbley, 2009]

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 68 / 94

    Bypass entropy estimation

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    69/94

    Renyi entropy estimator [Heroand Michel, 2002]

    Estimates H(X) from thelength of the minimal

    spanning tree of the samplesRenyi entropy has a discontinuityat = 1

    In the limit it is the Shannonentropy:

    lim1 H(X) = H(X)

    0 0.2 0.4 0.6 0.8 10

    0.2

    0.4

    0.6

    0.8

    1

    p

    H

    (p)

    =0

    =0.2

    =0.5

    Shannon

    =2

    =10

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 69 / 94

    Bypass entropy estimation

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    70/94

    Shannon entropy from the Renyi entropy estimates [Penalver et al., 2009]

    Find an

    value close to 1 to approximate the value of the Shannonentropy

    depends on the number of dimensions, number of samples andsome parameters which depend on the nature of the data and have tobe set experimentally.

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 70 / 94

    Bypass entropy estimation

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    71/94

    Shannon entropy from k-d partitioning [Stowell and Plumbley, 2009]

    Estimation depends on the number of samples in each partition andits volume

    Upper and lower limits have to be set

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 71 / 94

    Bypass entropy estimation

    Shannon entropy from the distances of the k-nearest neighbors graph

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    72/94

    Shannon entropy from the distances of the k nearest neighbors graph[Leonenko et al.2008]

    Estimation depends on N, d, and the distances between each sampleand its k-NN.

    3-NN

    3-NN

    k-NN of two different datasets.The result is the same.

    MST

    MST

    MST of two different datasets.The result is different.

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 72 / 94

    Comparison on Gaussian distributions

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    73/94

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 73 / 94

    Comparison on Gaussian distributions

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    74/94

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 74 / 94

    Comparison

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    75/94

    The k-NN-based estimator is a good alternative to the MST-basedestimation (does not need calibration of parameters).

    The k-NN-based estimator may fail with very separated modes.

    The k-NN-based estimator performs well for distributions in Rd.

    The k-d partitioning estimator performs better for distributions with afinite support.

    The k-d partitioning estimator tends to underestimate entropy whilethe k-NN-based tends to overestimate

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 75 / 94

    Outline

    1 Introduction

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    76/94

    The Robot Vision GroupResearchInformation theory

    2 TheoriesThe Method of Types

    3 MeasuresMutual information for alignmentEntropy in visual navigationJensen-Renyi divergence

    4 Principles

    Minimum description lengthMinimax entropy

    5 Entropy EstimationFeature Selection

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 76 / 94

    Feature selection in supervised classification

    Feature selection

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    77/94

    Discarding from all samples those features or variables which are less

    useful for some purpose, e.g. classification.Example: wrapper feature selection for supervised classification.

    M

    Images

    M

    NF

    Vectors

    M

    NS

    Vectors

    All Features Selected F.

    10-Fold

    CV

    Train

    Test

    Error

    Best F. Set

    ?

    Motivation: Datasets with thousands of features Dependencies among features

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 77 / 94

    Feature dependencies

    Uni ariate MI does not capt re the interactions among feat res

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    78/94

    Univariate MI does not capture the interactions among features.

    Figure by Guyon and Elisseeff, 2003

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 78 / 94

    MI-based criterion

    Filter feature selection criterion:

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    79/94

    Select the feature which, in

    combination with the rest,provides more information aboutthe class. (No independenceassumptions).

    Perform well with thousands ofdimensions.

    I(S|C) = H(S) H(S|C) k-NN-based estimator:

    small sample performance distributions in Rd no additional parameters

    (Discriminative approach).

    I(X;C)

    Cx1

    x2 x3x4

    0 20 40 60 80 100 120 140 16010

    15

    20

    25

    30

    35

    40

    45

    # features

    %e

    rrorandMI

    MD Feature Selection on Microarray

    LOOCV error (%)

    I(S;C)

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 79 / 94

    Application to visual localization

    Approach

    S l f f l

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    80/94

    Select from a set of general

    purpose filters Environment independence

    Bank of general, low-level filters

    Nitzberg

    Canny Horizontal Gradient

    Vertical Gradient

    Gradient Magnitude

    12 Color FiltersHi, 1 i 12

    (Feature independence cannot beassumed)

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 80 / 94

    Feature extraction

    Rings: distance dependent, orientation independent

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    81/94

    1

    2

    .

    .

    .

    K

    Img

    i C = 4

    K = 17

    Filters

    Bank

    .

    .

    .

    C x K

    1

    2

    3

    4

    1

    2

    68

    Rings Histograms

    .

    .

    .

    4 bins

    Feature Vector

    N = C x K x (B-1) = 204 features

    ... ...

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 81 / 94

    Nearest neighbors (1)

    800Confusion Trajectory for Fine Localization (P=10NN)

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    82/94

    0 100 200 300 400 5000

    100

    200

    300

    400

    500

    600

    700

    800

    Test Image #

    NN#

    Escolano, Bonev, Suau, Aguilar, Frauel, Saez, Cazorla (2007)Contextual visual localization: cascaded submap classification, optimized saliency detection, and fast view matching.

    IEEE International Conference on Intelligent Robots and Systems. San Diego, California, USA

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th , 2010 82 / 94

    Nearest neighbors (2)Test Image 1st NN 2nd NN 3rd NN 4th NN 5th NN

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    83/94

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th 2010 83 / 94

    Application to microarray analysis

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    84/94

    Fluorescent

    labeling

    Sample RNA

    Referece ADNc

    Combination

    Hybridization

    Fluorescent

    labeling

    To predict the tumor class, based on the microarray analysis of apatient.

    To identify a reduced set of genes which are related to the diseases.

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th 2010 84 / 94

    Gene selection on the NCI dataset

    MD Feature Selection

    RENALCNSCNSCNS

    mRMR Feature Selection

    RENALCNSCNSCNS 1

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    85/94

    Number of Selected Gene

    Class(disease)

    MELANOMAMELANOMAMELANOMAMELANOMAMELANOMAMELANOMA

    BREASTBREAST

    MELANOMANSCLC

    NSCLCNSCLC

    BREASTMCF7Drepro

    BREASTMCF7Arepro

    COLONCOLONCOLONCOLONCOLONCOLONCOLON

    LEUKEMIALEUKEMIALEUKEMIALEUKEMIALEUKEMIA

    K562AreproK562BreproLEUKEMIANSCLC

    NSCLCNSCLC

    PROSTATEOVARIANOVARIANOVARIANOVARIANOVARIAN

    PROSTATEMELANOMA

    OVARIANUNKNOWN

    RENALNSCLC

    BREASTRENALRENALRENALRENALRENALRENALRENAL

    NSCLCNSCLC

    BREASTCNSCNS

    BREAST

    19

    135

    246

    663

    766

    982

    1177

    1470

    1671

    2080

    3227

    3400

    3964

    4057

    4063

    4110

    4289

    4357

    4441

    4663

    4813

    5226

    5481

    5494

    5495

    5508

    5790

    5892

    6013

    6019

    6032

    6045

    6087

    6145

    6184

    6643

    Number of Selected Gene

    MELANOMAMELANOMAMELANOMAMELANOMAMELANOMAMELANOMA

    BREASTBREAST

    MELANOMANSCLC

    NSCLCNSCLC

    BREASTMCF7Drepro

    BREASTMCF7Arepro

    COLONCOLONCOLONCOLONCOLONCOLONCOLON

    LEUKEMIALEUKEMIALEUKEMIALEUKEMIALEUKEMIA

    K562AreproK562BreproLEUKEMIANSCLC

    NSCLCNSCLC

    PROSTATEOVARIANOVARIANOVARIANOVARIANOVARIAN

    PROSTATEMELANOMA

    OVARIANUNKNOWN

    RENALNSCLC

    BREASTRENALRENALRENALRENALRENALRENALRENAL

    NSCLCNSCLC

    BREASTCNSCNS

    BREAST

    133

    134

    135

    233

    259

    381

    561

    1378

    1382

    1409

    1841

    2080

    2081

    2083

    2086

    3253

    3371

    3372

    4383

    4459

    4527

    5435

    5504

    5538

    5696

    5812

    5887

    5934

    6072

    6115

    6145

    6305

    6399

    6429

    6430

    6566

    0

    0.1

    0.2

    0.3

    0.4

    0.5

    0.6

    0.7

    0.8

    0.9

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th

    2010 85 / 94

    Datasets and results

    NCI: 60 samples, 14 classes, 6380 features 10.94% LOOCV (39selected)

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    86/94

    selected)

    Leukemia: 38 + 34 samples, 2 classes, 6817 features 2.94% test (7selected) Colon: 62 samples, 2 classes, 2000 features 0% LOOCV (15

    selected) CNS Embryonal: 60 samples, 2 classes, 7129 features

    1.67% LOOCV (9 selected) Prostate: 102 + 43 samples, 2 classes, 12600 features 5.88% test

    (5 selected)

    Successful feature selection (outperforming state-of-the-art results) on

    data sets with A low number of samples and a high number of features High-order dependencies among informative features

    Bonev, Escolano, Cazorla (2008)Feature selection, mutual information, and the classification of high-dimensional patterns.

    Pattern Analysis and Applications

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th

    2010 86 / 94

    3D Object classification

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    87/94

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th

    2010 87 / 94

    Graph extraction

    Graphs from 3D shapes (SHREC database)

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    88/94

    Extended Reeb graphs: global topological structure; local featuresidentified by a function:

    a) geodesic distanceb) mass centerc) center of the circumscribing sphere

    Bonev, Escolano, Giorgi, Biasotti (2010)High-dimensional Spectral Feature Selection for 3D Object Recognition based on Reeb Graphs.

    Statistical, Structural and Syntactic Pattern Recognition. Cezme, Turkey

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th

    2010 88 / 94

    Unattributed graphs classificationSpectral features

    Complexity Flow

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    89/94

    Sphere

    Baricenter

    Geodesic

    Fiedler vector

    Adjacency spectrum

    Degrees

    Perron-Frobenius

    Norm. Lapl. Spectrum

    Node centrality

    Commute times

    3D ShapeSample 1Class

    2 4 6 8

    ...Feature vector - Sample 1

    .

    .

    .

    Feature vector - Sample n

    Feature vector -Sample n-1

    Feature selection

    C1

    C1

    C1

    C2

    C15

    C15

    .

    .

    .

    0 100 200 300 400 50020

    25

    30

    35

    40

    45

    50

    X: 222

    Y: 23.33

    # features

    %er

    ror

    Mutual Information Feature Selection

    10fold CV error

    Mutual Information

    2 bins4 bins6 bins8 bins

    Commute Times 1

    Commute Times 2

    Node Centrality

    N.Laplacian Spectrum

    PerronFrobeniusDegrees

    Adjacency Spectrum

    Fiedler Vector

    Complexity Flow

    Geodesic Graph

    Baricenter Graph

    Sphere Graph

    Statistics for the first 222 selected features

    Boyan Bonev (University of Alicante) Information-theoretic Computer Vision November 26th

    2010 89 / 94

    Feature analysis

    Proportion of features during selection

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    90/94

    0 100 200 300 400 500

    Commute Times 1

    Commute Times 2

    Node Centrality

    N.Laplacian Spectrum

    PerronFrobenius

    DegreesAdjacency Spectrum

    Fiedler Vector

    Complexity Flow

    # features

    Geodesic Graph

    Baricenter Graph

    Sphere Graph

    Proportion of features during selection

    Boyan Bonev (University of Alicante) Information theoretic Computer Vision November 26th

    2010 90 / 94

    Summary

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    91/94

    Information-theory as a theoretical framework and a set of tools whichhelp many decoding tasks in computer vision. Some of them:

    Method of Types error bound analysis

    Measures (entropy, mutual information, divergences) alignment,regions of interest, segmentation, etc

    Minimum description length model order selection, avoid overfitting

    Maximum entropy the most unbiased hypothesis

    Mutual information feature selection

    IT in both generative and discriminative approaches

    Boyan Bonev (University of Alicante) Information theoretic Computer Vision November 26th

    2010 91 / 94

    Conclusions

    Entropy estimation

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    92/94

    the cornerstone of many information-theoretical implementations

    some complexity issues depending on the estimation method

    bypass estimation advances have motivated the use of IT in CVPR

    Information theory in CV for robotic tasks

    towards environment-independent applications treat images as general information provided to solve a task deal with noise and unuseful information minimize the number of assumptions

    challenges related to computational cost (even with low complexity in

    some cases)

    Everything should be made as simple as possible, but no simpler.

    Boyan Bonev (University of Alicante) Information theoretic Computer Vision November 26th

    2010 92 / 94

    References

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    93/94

    A. Yuille (2010)

    An information theory perspective on computational vision.Front. Electr. Electron. Eng. China

    F. Escolano, P. Suau, B. Bonev (2009)Information Theory in Computer Vision and Pattern Recognition.Springer

    Penalver, Escolano, Saez (2009)Learning Gaussian Mixture Models with Entropy Based Criteria.IEEE Transactions on Neural Networks

    B. Bonev, F. Escolano, M. Cazorla (2008)

    Feature selection, mutual information, and the classification ofhigh-dimensional patterns.Pattern Analysis and Applications

    Boyan Bonev (University of Alicante) Information theoretic Computer Vision November 26th

    2010 93 / 94

    Information-theoretic Computer Visionfor Autonomous Robots

  • 8/3/2019 Information-theoretic Computer Vision for Autonomous Robots

    94/94

    Boyan Bonev

    Robot Vision GroupUniversity of Alicante

    November 26th, 2010

    Talk at the Max Planck Institute for Biological Cybernetics,Tubingen, Germany

    Bo an Bone (Uni ersit of Alicante) Information theoretic Comp ter Vision No ember 26th

    2010 94 / 94