86
Stochastic Tracking of Stochastic Tracking of Humans Humans Michael J. Black http://www.cs.brown.edu/~black Department of Computer Science Brown University

Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Embed Size (px)

Citation preview

Page 1: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Stochastic Tracking of HumansStochastic Tracking of Humans

Michael J. Black

http://www.cs.brown.edu/~black

Department of Computer ScienceBrown University

Page 2: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

CollaboratorsCollaboratorsHedvig Sidenbladh

Royal Institute of Technology (KTH), Sweden

Dirk Ormoneit and Trevor Hastie Dept. of Statistics, Stanford University

David Fleet Xerox PARC and Queen’s University

Allan JepsonUniversity of Toronto

Page 3: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Goal: 3D Human MotionGoal: 3D Human Motion

* 3D articulated model* Perspective projection* Monocular sequence* Unknown, cluttered, environment

* Infer 3D human motion from 2D image motion.

Page 4: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

OverviewOverview* Why is 3D human motion important?* Why is recovering it hard?* A Bayesian approach

* generative model * robust likelihood function* temporal prior model (learning)* stochastic search (particle filtering)

* Where are we going?* Recent advances & state of the art.* What remains to be done?

Page 5: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Why is it Important?Why is it Important?ApplicationsApplications

• Human-Computer Interaction• Surveillance • Motion capture (games and animation)• Video search/annotation• Work practice analysis.

Social display of puzzlement

* detect moving regions* estimate motion* model articulated objects* model temporal patterns of activity * interpret the motion

Page 6: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Why is it Hard?Why is it Hard?

The appearance of peoplecan vary dramatically.

Bones and jointsare unobservable(muscle, skin, clothing hide the underlying structure).

(inference)

Page 7: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Why is it hard?Why is it hard?

People can appear in arbitrary poses.

They can deform in complexways.

Occlusion results inambiguities and multiple interpretations.

Page 8: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Why is it hard?Why is it hard?

Geometrically under-constrained.

Page 9: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Other ProblemsOther Problems

* non-linear dynamics of limbs* similarity of appearance of different limbs (matching ambiguities)* image noise* outliers

Our models are approximations.Image changes that are not modeled(e.g. clothing deformation) will be outliers.

Page 10: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Common AssumptionsCommon Assumptions

* Multiple Cameras(additional constraints, occlusion)

* Color Images(locate face and hands)

* Known Background(background subtraction to locate person)

* Batch process an entire sequence.* Known Initialization

(to be avoided)

Page 11: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

RequirementsRequirements

1. Represent uncertainty and multiple hypotheses.

2. Model non-linear dynamics of the body.

3. Exploit image cues in a robust fashion.

4. Integrate information over time.

5. Combine multiple image cues.

Page 12: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Bayesian InferenceBayesian InferenceBuild models of human form and motion. Learnpriors over model parameters:

p(model)

Exploit cues in the images. Model robust likelihoods:p(image cue | model)

Represent the posterior distributionp(model | cue) p(cue | model) p(model)

Page 13: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

ProblemsProblemsA simple articulated human model may have 30+parameters (e.g. joint angles. 60+ w/ velocities).

Models of human action are non-linear and likelihood models will be multi-modal.

Key challenges Key challenges (common to other domains)• representation,• learning, and• search

in high dimensional spaces.

Page 14: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Bayesian FormulationBayesian Formulation

Represent a distributionover 3D poses.

* define generative model of image appearance* multi-modal posterior over model parameters - sampled representation - particle filtering approach.* focus on image motion as a cue (adding edges,…)

Page 15: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Generative Model: ShapeGenerative Model: Shape

* 3D Articulated Body Model* pinhole camera* parameter vector =

Page 16: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Generative Model: MotionGenerative Model: Motion

t-1

t

),( 11t ttM IR

Projection ofimage textureonto the 3D model

),(1ttt M RI

Projection of modelappearance intoimage coordinates

Page 17: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Appearance ModelAppearance ModelCould be many things

* template (Cham & Rehg ‘99)* eigen-model (Sidenbladh, et al ‘00)* texture model* filter responses (edges, ridges, …)* learned over time

Simple probabilistic model:

)),((),|(),,|( 1111111 tttttttttt Mpp IRIRRIR

Markov assumption

Page 18: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Noise ModelNoise Model

),,()(I 1jttjt M xRx

Generative model:

cGp t )()1();(Mixture of Gaussian and uniform outlier distribution:

Function ofsurface orientation

256/110

Page 19: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Generative Model: TemporalGenerative Model: Temporal

),|(),|( 1111 tttttt pp VV

)|(),|( 111 ttttt pp VVVV

* general smooth motion or,* action-specific motion (walking)

First order Markov assumption on angles, , and angular velocity, V:

Explore two models of human motion

Page 20: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Bayesian FormulationBayesian Formulation

1111111

11111

))|,,(

),|()|(),|((),(

)|,,(

ttttttt

ttttttttttt

tttt

dddp

pppp

p

RVIRV

IRVVVRI

IRV

Posterior over shape, velocity, and appearance given an image sequence.

Likelihood ofobserving the imagegiven the shape andappearance parameters

Temporal model

Posterior fromprevious time instant

Page 21: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Robust LikelihoodRobust Likelihood

For n random pixels from limb j compute: ij ,x

n

i j

ijtijtimagej

jp

1)(22

2)),(),(ˆ(

)(21

256, )exp(

xIxI

);,(M)(I ,1

, ijttijt xRx where

imagejj

ttt pp ,),|( RI

Page 22: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Temporal Model: Smooth MotionTemporal Model: Smooth Motion

otherwise0

],[if)),((

),|(

max,min,,1,1,,

1,1,,

iitiitititi

tititi

G

p

V

V

),()|( V1,,1,, ititititi Gp VVVV

* individual angles and velocities assumed independent

Page 23: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

What does the posterior look like?What does the posterior look like?

x y z

Shoulder: 3dofElbow: 1dof

Elbow bends

Page 24: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Particle FilteringParticle Filtering* large literature (Gordon et al ‘93, Isard & Blake ‘96,…)

* non-Gaussian posterior approximated by N discrete samples

* explicitly represent the ambiguities

* exploit stochastic sampling for tracking

],,[ )()()( nt

nt

ntt RVs )10( 4NNn ,...,1

Page 25: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Particle FilterParticle Filter

samplesample

samplesample

normalizenormalize

Posterior

)I|V,( 111 tttp

Temporal dynamics )V,|V,( 11 ttttp

Likelihood)|I( ttp )I|V,( tttp

Posterio

r

Page 26: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Arm Tracking: Smooth motion priorArm Tracking: Smooth motion prior

Particle filter * represents ambiguity * propagates information over time

x y z

Display: expectedvalue of joint angles.

Page 27: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Full-Body TrackingFull-Body Tracking * parameter space too large * constrain posterior to valid 3D human motions. * learn generative models automatically from training data.

time

joint angles3D motion-capturedata: * segment into “movemes” (Bregler) * train probabilistic model.

(from M. Gleicher)

Page 28: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Learning Temporal ModelsLearning Temporal Models

* Motion capture data is noisy, data is missing, activities are performed differently.

* For cyclic motion (important but special class):1. Detect cycles and segment2. Account for missing data3. Preserve continuity of cycles4. Statistical model of variation

* Approaches should generalize to non-cyclic motion.

(Dirk Ormoneit & Trevor Hastie)

Page 29: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Detecting CyclesDetecting Cycles

Automatically detect length of cycles,Automatically segment and align cycles.

Page 30: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Modeling Cyclic MotionModeling Cyclic Motion

Automaticallyalign 3D data with a reference curverepresented usingperiodicallyconstrainedregression splines.

Page 31: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Modeling Cyclic MotionModeling Cyclic Motion

* Iterative SVD method (from gene expression work)* computes SVD in Fourier domain* construct a rank-q approximation and take inverse Fourier transform* impute missing data from the approximation* repeat until convergence.

* Segment into cycles, compute mean curve and represent variation by performing PCA on data.

* SVD must enforce periodicity and cope with missing data.

Page 32: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Action-Specific ModelAction-Specific Model

q

ktkkttt vc

1, )()(~

The joint angles at time t are a linear combinationof the basis motions evaluated at phase

Mean curve Basis curves

Page 33: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Temporal Model: WalkingTemporal Model: Walking

],,,,[ gt

gttttt c

sParameters of the generative model are now

),()|(

),]100[]1,([),,|(

),()|(

)),(()|(

),,()|(

11

1111111

111

111

,1,,1,

gggg

Ttt

Tgtt

gtt

gt

tttt

ttttt

kcc

ktktktkt

ttttGp

Gp

Gp

Gp

ccGccp

TT

Probabilistic model for )|( 1ttp ss

Page 34: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Learned Walking ModelLearned Walking Model

* mean walker

Page 35: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Learned Walking ModelLearned Walking Model

* sample with small

Page 36: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Learned Walking ModelLearned Walking Model

* sample with moderate

Page 37: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Learned Walking ModelLearned Walking Model

* sample with very large

Page 38: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Stochastic 3D TrackingStochastic 3D Tracking

Stochastic 3D tracking (manual initialization)

Use motion informationto update and trackdistribution over time

Page 39: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Stochastic 3D TrackingStochastic 3D Tracking

* significant changes in view and depth.* template-based methods will fail.

Page 40: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

No likelihoodNo likelihood

* how strong is the walking prior? (or is our likelihood doing anything?)

Page 41: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

IssuesIssues* Large parameter space

* approx. 10000 samples * sparsely represented* not real time

* Flow-based models can drift

* Requires initialization

Page 42: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Lessons LearnedLessons Learned* Probabilistic (Bayesian) framework allows - integration of information over time

- modeling of priors- explicit generative image model

* Particle filtering allows- multi-modal distributions- tracking with ambiguities and non-linear models

* Weak image cues necessitate strong priors and many samples.

Page 43: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Work to be doneWork to be done

* better appearance model - other cues (Color, edges, appearance,…)

* automatic initialization using 2D models* learn more general models of motion* better occlusion model (new)* model of the background motion (new)

* better representations of the posterior (Fleet&Chou)* better sampling methods (Fleet&Ormoneit)* adapt shape of limbs

Page 44: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Very preliminary work…

Page 45: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

The Statistics of People in The Statistics of People in Images and VideoImages and Video

How do people appearin natural scenes?

Want a general model.

EdgeFilters

RidgeFilters

Page 46: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Statistics of ImagesStatistics of Images

Ruderman. Lee, Mumford, Huang. Portilla and Simoncelli. Olshausen & Field. Xu, Wu, & Mumford. …

Learning Pon and Poff for edge detection and road following:

Geman and JednyakKonishi, Yuille, and Coughlan

Page 47: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Example Training ImagesExample Training Images

Page 48: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Distribution of Filter ResponsesDistribution of Filter Responses

Page 49: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Ratios for different limbsRatios for different limbs

Page 50: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Local Contrast NormalizationLocal Contrast Normalization

Page 51: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

LikelihoodLikelihood

bf x

bb

x

ffbf xIpxIpIp )|)(()|)((),|(

f

f

x

bfx

ff

x

b

xIp

xIpxIp

)|)((

)|)(()|)((

f

f

x

bfx

ff

xIp

xIpC

)|)((

)|)((

Foreground pixels

Background pixels

Page 52: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University
Page 53: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

BenefitsBenefits• Generic model of appearance.

• Principled way to chose filters.

• Model of foreground and background is incorporated into the tracking framework.

• exploits the ratio between foreground and background likelihoods.• improves tracking.

• Done the same for ridges and motion.

Page 54: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

OutlookOutlook5 years:

- Relatively reliable people tracking in monocular video.- Path is pretty clear.

… solve the vision problem.

Next step: Beyond person-centric- people interacting with object/world

Beyond that: Recognizing action- goals, intentions, ...

… solve the AI problem.

Page 55: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Some Related WorkSome Related Work* Bregler & Malik: image motion, single hypothesis,

full-body required multiple cameras, scaled ortho.* Ju, Black, Yacoob: cardboard person model,

image motion, 2D* Deutscher et al: Condensation, edge cues,

background subtraction.* Cham& Rehg: known templates, 2D (SPM), particle

filter.* Wachter & Nagel: nicely combines motion and edges,

single hypothesis (Kalman filter).* Leventon & Freeman: assumes 2D tracking,

probabilistic formulation, learned temporal model

(full body, monocular, articulated)

Page 56: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

ConclusionsConclusionsBayesian formulation for tracking 3D human figuresusing monocular image information.

* Generative model of image appearance.* Non-linear model represents ambiguities, singularities occlusion, etc - sampled representation of posterior.* Particle filtering for incremental estimation.* Automatic learning of cyclic motion prior.

Rich framework for modeling the complexity ofhuman motion.

Page 57: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Initialization Using 2D ModelInitialization Using 2D Model

* Full-body walking model.

* Constructed from 3D mocap data.

* 2D, view-based (every 30 degrees)

* 4 subjects, 14 cycles

Page 58: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

2D, View-Based Walker2D, View-Based Walker* Construct linear optical flow basis

* Use similar Bayesian framework for tracking (Black CVPR’99)

* Coarse estimate of 3D parameters

* Automatic initialization

Example Bases:

...

...

0 degrees

90 degrees

Page 59: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Recent ResultsRecent Results

* Box indicates mean position and scale.* Recovers distribution over phase and 3D scale.

Page 60: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Contrast NormalizationContrast Normalization

Locally weight image derivatives by

contrast

OcontrastSw

*2

)*tanh(1

Global contrast normalization (Lee, Mumford & Huang)

log(I

IInorm

Page 61: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Optimizing the FiltersOptimizing the Filters• Chose contrast normalization to maximize detection accuracy

• ROC curve

• Battacharyya

• Kullback-Leibler

dyypyppp offonoffonB )()(log),(

dyyp

ypyppp

off

ononoffonKL )(

)(log)(),(

Page 62: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Local Contrast NormalizationLocal Contrast Normalization

Page 63: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Representing the PosteriorRepresenting the Posterior

N

i

itt

ntt

p

pnt

1

)(

)(

)|(

)|()(

sI

sI

),( )()( nt

nt s

)|,,( ttttp IRV

represented by discrete set of N samples

Normalized likelihood:

Page 64: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

CondensationCondensation1. Selection Sample from posterior at t-1

Most probable states selected most often.2. Prediction.3. Updating

Page 65: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

1. Selection2. Prediction/Diffusion (sample from )

ie from the temporal prior:

1. Compute

2. Sample from

3. Sample from

3. Updating

CondensationCondensation

)|( 1ttp ss

),|()|(),|( 11111 tttttttt ppp IRVVV

),|( 11 tttp IR

)|( 1ttp VV

),|( 11 tttp V

tR

Page 66: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

states

p

ts

1ts

CondensationCondensation1. Selection2. Prediction/Diffusion (sample from )

Models the dynamics:

3. Updating

)|( 1ttp ss

Page 67: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

CondensationCondensation1. Selection2. Prediction3. Updating (the distribution)

Evaluate new likelihood.

Repeat until N new samples have beengenerated.

Compute normalized probability distribution.

),|()|( ttttt pp RIsI

Page 68: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Visualizing ResultsVisualizing Results

)(

1

)( )(|)( nt

N

n

nttt ffE

sIs

Expected value of state parameter )( tf s

Page 69: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

LikelihoodLikelihood

* To cope with occluded limbs or those viewed at narrow angles, we introduce a probability of occlusion.* likelihood of observing limb j is then

occludedimagej pqpqp )1(

* likelihood of the model is product of limb likelihoods

jj

ttt pp ),|( RI

Page 70: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Indexing/SearchIndexing/SearchThe crux of the problem.

• The parameter space is huge.• Brute force search is infeasible (ditto discretely sampling the space).• Need to index into correct part of the space.• Use a hierarchy of models of increasing complexity

Images Generic Models(expansion, rotation,…)

Coarse ObjectModels

(EigenPeople)

Detailed Models(shape & activity)

Compute likelihood

Index Indexw/ Jepson & Fleet.

Page 71: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

InitializationInitialization

* new spatially constrained mixture model * find appropriate mid-level representations* initialize high level models using mid-level cues

Page 72: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Digital Video AnalysisDigital Video Analysis

Social display of puzzlement

To automatically analyze such a sequence we must * detect moving regions * estimate and interpret the motion * model complex articulated objects such as humans * model temporal patterns of activity

Page 73: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Tracking Moving StructureTracking Moving Structure

Next steps * split/merge/kill/initialize/grow/shrink operations * probabilistic search for best interpretation of the scene * detect more complex structures (articulation)

Page 74: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Generative ModelGenerative Model

Page 75: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Mouth Training DataMouth Training Data

* 3000 image training sequence* motion estimated between pairs of frames* utterances: “center”, “print”, “track”, “release”

Page 76: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Learned Spatial ModelLearned Spatial Model

* 3 basis flow fields account for 85% of variance.* fewer needed for recognition than for accurate estimation.

Page 77: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Mouth Temporal ModelsMouth Temporal Models

Page 78: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Mouth ResultsMouth Results

Page 79: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Results

Page 80: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Mouth ResultsMouth Results

Page 81: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Let be the image measurements at time t.

Let be a sequence of measurements from 0 to t .

Bayesian FormulationBayesian FormulationLet ),,,,,( ps t be a state.

)|()()|( 1 tttttt pzkpp zsszs

We want

* not Gaussian.

tz

Measurement likelihood.Can’t represented in closedform.

Temporal prior.Can be sampled

tz

Page 82: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Generative ModelGenerative Model(Brightness Constancy)(Brightness Constancy)

),0()1),;((I )),((I tt ttt sxu)x(ssx

ppxsx )()( t

Optical flow

))(()();( ,1

psxbsxu

ti

n

iit a

Page 83: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Representing the PosteriorRepresenting the Posterior

S

i

itt

ntt

zp

zpnt

1

)(

)(

)|(

)|()(

s

s

),( )()( nt

nt s

)|( ttp zs represented by discrete set of S samplesSnn

tn

t ,...,1),,( )()( s

Page 84: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Stochastic SearchStochastic Search

* Particle filtering (Condensation):1. Sample from posterior at time t-1.2. Predict using temporal prior.3. Evaluate likelihood.

* Predict non-Gaussian distribution over time.* Update posterior with new measurements.* Allocate computational resources to effectively explore the space.

Page 85: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Generative Model: MotionGenerative Model: Motion

t-1

t

),P( 11-t tyx

),P(t tyx

),P(),P( 1t tt yyu

Page 86: Stochastic Tracking of Humans Michael J. Black black Department of Computer Science Brown University

Learned Walking ModelLearned Walking Model

* sample with large