Prediction of Non-Linear Aging Trajectories of Faces

Slides:



Advertisements
Similar presentations
Pattern Recognition and Machine Learning
Advertisements

Active Appearance Models
Machine Learning and Data Mining Linear regression
Principal Component Analysis Based on L1-Norm Maximization Nojun Kwak IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008.
SVM - Support Vector Machines A new classification method for both linear and nonlinear data It uses a nonlinear mapping to transform the original training.
SVM—Support Vector Machines
Pattern Recognition and Machine Learning: Kernel Methods.
Computer vision: models, learning and inference Chapter 8 Regression.
CSCI 347 / CS 4206: Data Mining Module 07: Implementations Topic 03: Linear Models.
Data Modeling and Parameter Estimation Nov 9, 2005 PSCI 702.
Sami Romdhani Volker Blanz Thomas Vetter University of Freiburg
Face Alignment with Part-Based Modeling
« هو اللطیف » By : Atefe Malek. khatabi Spring 90.
3D Face Modeling Michaël De Smet.
4/15/2017 Using Gaussian Process Regression for Efficient Motion Planning in Environments with Deformable Objects Barbara Frank, Cyrill Stachniss, Nichola.
Basis Expansion and Regularization Presenter: Hongliang Fei Brian Quanz Brian Quanz Date: July 03, 2008.
Uncertainty Representation. Gaussian Distribution variance Standard deviation.
Gaussian process emulation of multiple outputs Tony O’Hagan, MUCM, Sheffield.
Content Based Image Clustering and Image Retrieval Using Multiple Instance Learning Using Multiple Instance Learning Xin Chen Advisor: Chengcui Zhang Department.
Kernel methods - overview
Exchanging Faces in Images SIGGRAPH ’04 Blanz V., Scherbaum K., Vetter T., Seidel HP. Speaker: Alvin Date: 21 July 2004.
x – independent variable (input)
Face Recognition Based on 3D Shape Estimation
Development of Empirical Models From Process Data
ECIV 301 Programming & Graphics Numerical Methods for Engineers REVIEW III.
3D Geometry for Computer Graphics
Support Vector Regression (Linear Case:)  Given the training set:  Find a linear function, where is determined by solving a minimization problem that.
Laurent Itti: CS599 – Computational Architectures in Biological Vision, USC Lecture 7: Coding and Representation 1 Computational Architectures in.
Learning the Appearance of Faces: A Unifying Approach for the Analysis and Synthesis of Images. Thomas Vetter Germany University of Freiburg
Classification and Prediction: Regression Analysis
Face Recognition Using Neural Networks Presented By: Hadis Mohseni Leila Taghavi Atefeh Mirsafian.
Gender and 3D Facial Symmetry: What’s the Relationship ? Xia BAIQIANG (University Lille1/LIFL) Boulbaba Ben Amor (TELECOM Lille1/LIFL) Hassen Drira (TELECOM.
Empirical Modeling Dongsup Kim Department of Biosystems, KAIST Fall, 2004.
Cao et al. ICML 2010 Presented by Danushka Bollegala.
PATTERN RECOGNITION AND MACHINE LEARNING
Machine Learning CUNY Graduate Center Lecture 3: Linear Regression.
1 Hybrid methods for solving large-scale parameter estimation problems Carlos A. Quintero 1 Miguel Argáez 1 Hector Klie 2 Leticia Velázquez 1 Mary Wheeler.
Efficient Irradiance Normal Mapping Ralf Habel, Michael Wimmer Institute of Computer Graphics and Algorithms Vienna University of Technology.
Graphite 2004 Statistical Synthesis of Facial Expressions for the Portrayal of Emotion Lisa Gralewski Bristol University United Kingdom
Machine Learning Seminar: Support Vector Regression Presented by: Heng Ji 10/08/03.
1 Recognition by Appearance Appearance-based recognition is a competing paradigm to features and alignment. No features are extracted! Images are represented.
Jeff Howbert Introduction to Machine Learning Winter Regression Linear Regression.
Model Construction: interpolation techniques 1392.
Kernel Methods A B M Shawkat Ali 1 2 Data Mining ¤ DM or KDD (Knowledge Discovery in Databases) Extracting previously unknown, valid, and actionable.
PATTERN RECOGNITION AND MACHINE LEARNING CHAPTER 3: LINEAR MODELS FOR REGRESSION.
A hybrid SOFM-SVR with a filter-based feature selection for stock market forecasting Huang, C. L. & Tsai, C. Y. Expert Systems with Applications 2008.
Classifiers Given a feature representation for images, how do we learn a model for distinguishing features from different classes? Zebra Non-zebra Decision.
Sparse Bayesian Learning for Efficient Visual Tracking O. Williams, A. Blake & R. Cipolloa PAMI, Aug Presented by Yuting Qi Machine Learning Reading.
Ohad Hageby IDC Support Vector Machines & Kernel Machines IP Seminar 2008 IDC Herzliya.
Machine Learning ICS 178 Instructor: Max Welling Supervised Learning.
Point Distribution Models Active Appearance Models Compilation based on: Dhruv Batra ECE CMU Tim Cootes Machester.
Learning Photographic Global Tonal Adjustment with a Database of Input / Output Image Pairs.
Instructor: Mircea Nicolescu Lecture 7
Neural Network Approximation of High- dimensional Functions Peter Andras School of Computing and Mathematics Keele University
Support Vector Regression in Marketing Georgi Nalbantov.
Basis Expansions and Generalized Additive Models Basis expansion Piecewise polynomials Splines Generalized Additive Model MARS.
RiskTeam/ Zürich, 6 July 1998 Andreas S. Weigend, Data Mining Group, Information Systems Department, Stern School of Business, NYU 2: 1 Nonlinear Models.
Machine Learning Supervised Learning Classification and Regression K-Nearest Neighbor Classification Fisher’s Criteria & Linear Discriminant Analysis Perceptron:
1 C.A.L. Bailer-Jones. Machine Learning. Data exploration and dimensionality reduction Machine learning, pattern recognition and statistical data modelling.
Chapter 7. Classification and Prediction
DEEP LEARNING BOOK CHAPTER to CHAPTER 6
CSE 4705 Artificial Intelligence
Project 4: Facial Image Analysis with Support Vector Machines
Recognition: Face Recognition
Machine Learning Basics
Final Year Project Presentation --- Magic Paint Face
Identification of Wiener models using support vector regression
Machine learning overview
Introduction to Sensor Interpretation
Introduction to Sensor Interpretation
Presentation transcript:

Prediction of Non-Linear Aging Trajectories of Faces K. Scherbaum, M. Sunkel, V. Blanz and H.-P. Seidel [ 2007/5/9, Eurographics 2007, Prague ]

Motivation / Goal automated growth-prediction system applications photofit-pictures of missing children automated animation, art Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Age Progression – Optimal Case 9 years 10 years 11 years child 1 child 2 challenges: depends on individual face depends on age (curved trajectory) no longitudinal study child 3 face space Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Real Case – Support Vector Regression only 1 sample per person no longitudinal study find isosurfaces and gradients 9 years 10 years 11 years 11 years 9 years 10 years challenges: depends on individual face depends on age (curved trajectory) no longitudinal study Runge-Kutta Integration face space Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Main Assumption - Curved Trajectories growing faces transform along curved trajectories use machine learning non-linear Support Vector Regression integration of local age-gradient challenges: depends on individual face depends on age (curved trajectory) no longitudinal study Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Challenges learn change over time of individual faces non-linear dependency on time, curved trajectory learn how the change depends on individual face non-linear dependency in face space sparse dataset, no longitudinal study challenges: depends on individual face depends on age (curved trajectory) no longitudinal study Kristina Scherbaum scherbaum@mpi-inf.mpg.de

3D Morphable Facemodel System is based on a Morphable 3D Facemodel [Blanz,Vetter‘99] Built from 200 3D-face-scans of adults Kristina Scherbaum scherbaum@mpi-inf.mpg.de

3D Morphable Facemodel vector space of faces vectors with point-to-point correspondence Shape linear combinations of faces Texture Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Representation of Faces - Face Spaces PCA to reduce dimensionality (yields coefficients) 3 different representations: vector representation linear combinations pca space (reduced complexity), variation In the morphable model, faces are represented as vectors for the shape “S” and for the texture “T” each linear combination of different faces is a new realistic face in order to reduce complexity a PCA was applied on the shape and texture vectors this defines an orthogonal set of basis vectors si and ti with the average shape and texture we can define shape and texture coefficients c_is and c_it Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Extended Morphable Model Extension by … plus ~238 facemodels of teenagers 3 simultaneous laser scans per face Correspondence by … top-down approach fitting Morphable Model to new 3D faces merging original data and best fit Evtl erweitern: 2-3 folien Evtl. auch Simultanfit erklaeren Beispiele der Teenagerdatenbank zeigen (Diplomarbeit S 58 und S 59) Zoom, zeigen dass Korrepondenz vorhanden ist (evtl morph) und beleuchtung normiert wurde evtl noch die Texturreko zeigen. Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Fitting the Morphable Model to 3D Scans no optical flow because scans are often incomplete best fit of the morphable model merged result 3D laser scans Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Texture Extraction from Pictures 3 pictures per face under inconsistent lighting conditions view dependent mapping result with normalised lighting 3D reco extracted textures pictures Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Age Progression Algorithm 1 learn function that maps any face x to a scalar age y to learn this function we use … non-linear Support-Vector-Regression on training sets of l pairs f: Face  Age R^n  R maps any face x to a scalar age value let a face x consist either of shape coefficients or texture coefficients to reduce the complexity we do not use all principal components (k=20,40,80) yi denotes the age of eac h example face i Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Fitting a Regression Curve 2 for a given set of samples find f(x) such that all samples are within an e-tube preselect e and tradeoff between smoothness and errors of outliers y e e x f: Face  Age gross darstellen evtl bilder dazu R^n  R discuss linear methods at the end x Linear: f(x) = wx + b Non-linear: f is sum of Gaussian RBF kernels K(x-xi) Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Non-Linear SVM Regression 2 Gaussian RBF (Radial Basis Function) as kernel we applied grid search using cross validation to optimize parameters such as g (Kernelwidth) i and b are determined by SVM training using LIBSVM for e-Support Vector Regression a_i and be are real numbered values determined by the SVM training set of 20, 40 and 80 PCs are used in cross validation we split the dataset in 11 different random ways into 90% training and 10% test faces Kristina Scherbaum scherbaum@mpi-inf.mpg.de

3 Local Aging Isosurfaces are defined in PCA space Gradient gives shortest path to next isosurface Along the gradient … many facial changes due to aging almost no other changes (known technique, Blanz et al. 99) Thus: Compute growth along the gradient! Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Gradient Example - Facial Attributes 3 gender manipulation male female original Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Growth Simulation: New Approach 3 growth curve with given face x0 at time t currently we compute the local gradient and walk along this gradient instead we should compute the curved trajectory Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Runge Kutta Integration 4 Solve differential equation … to compute curved trajectories integrate the differential equation using Runge-Kutta algorithm perform small steps for all x and t the gradient of f describes the direction of minimal change in x to achieve a given change in t so that the characteristic features of the face are retained in the best way possible Runge Kutta (4th order): performs small steps along the gradient of f the gradient of f can be computed from the RBF regression function Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Visualized Aging Trajectories 4 Kristina Scherbaum scherbaum@mpi-inf.mpg.de

4 Reducing Complexity growth leads to overall change of facial size we did not train on all principle components speedup of SVM training we experimented with 20, 40 or 80 PCs Justification … growth leads to overall change of facial size significant changes are represented by the first PCs [ large variance ] facial growth should happen in the first PCs Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Growth Example growth simulation for both, shape and texture 12 14 16 18 20 years 22 24 26 28 30 years Kristina Scherbaum scherbaum@mpi-inf.mpg.de

More Examples 10 years 3D laser scans, original age 12 13 12 10 14 look at the bottom line – different individuals not always the same result / type characteristic features remain 20 years 30 years Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Rendering the Result into Images [EG’04] Background, Haircut Pose, Light Face Composed Result 3D reconstruction and aging Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Photofit Picture Example Input at the age of 11 Possible appearances at the age of 17 Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Aging in Images - Example Picture (1999) Different prediction renderings 3D reconstruction and aging Ground truth pictures (2005) Kristina Scherbaum scherbaum@mpi-inf.mpg.de

3D reconstruction aging (extrapolated) Extrapolated Example GROUND TRUTH picture 2 years old prediction, 14 years 3D reconstruction aging (extrapolated) extreme example database did not contain babies or toddlers Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Linear vs. Non-Linear Linear age progression Disadvantages … perform linear regression (yields a function) [ straight-forward least squares fit ] transform faces also along the gradient Disadvantages … the gradient is constant [ linear function ] each face moves along the same straight trajectory pro und contra linear zum vergleich nochmal linear am ende – SVR schon erklaert. neue fit-funktion zum vergleich Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Linear vs. Non-Linear comparison of age estimation error (in months) mean squared training and generalization errors non-linear (RBF) 32.68 18.12 linear 66.14 60.05 non-linear (RBF) 38.90 29.35 linear 67.66 62.87 (here for 20 PCs) non-linear SVM regression behaves superior! generalization indicates: no overfitting Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Remember the Challenges Are growth trajectories curved? Mean angle between start- and target-tangent 10.3º 30.0º  the trajectories are curved, not linear Have different faces distinct trajectories? Mean angle of trajectories of different faces 15.7º 33.5º  the trajectories are different challenges: depends on individual face depends on age (curved trajectory) no longitudinal study Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Conclusions Results … aging involves non-linear components trajectories are distinct for different individuals linear systems are a reasonable approximation technique works without longitudinal data But … more data would be helpful longitudinal data would allow for exact evaluation Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Thank you for your attention! MOVIE Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Transforming Faces Along Trajectories 5 given face x0 with estimated start age test = f(x0) simulate trajectory z along a time period t = ttarget - test compute a face vector for the target age ttarget z fuer from und textur getrennt! if t0 and test differ compensate for error (shift face) ensures that face does not look too young or too old Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Observations mean angles as non-linearity measures between … trajectories of different faces linear and RBF gradients start- and target-tanget of a face aging trajectories are curved! they tend to be linear for more PCs Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Evaluation Prediction is not truly evaluated … Possible improvements … no 3D ground-truth data available exact evaluation not possible we reconstruct a 3D model from pictures as ground truth but the reconstruction is always the best guess only Possible improvements … extend the database of teenagers, re-scan faces today user studies (prediction from pictures, comparison) considering side effects (parents look, nutrition, smoking …) Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Piecewise Linear Approach given input face and age averages find start and target age range compute start and target average face subtract and add scaled personality Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Simple Solution - Piecewise linear simple and fast solution easy to calculate sufficient results but … no smooth transitions sometimes individuals are recognizable more data needed SVR is a more fundamental machine learning approach Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Representation of Faces - Face Spaces arbitrary faces by linear combinations of examples PCA to reduce dimensionality (yields coefficients) 3 different representations: vector representation linear combinations pca space (reduced complexity), variation In the morphable model, faces are represented as vectors for the shape “S” and for the texture “T” each linear combination of different faces is a new realistic face in order to reduce complexity a PCA was applied on the shape and texture vectors this defines an orthogonal set of basis vectors si and ti with the average shape and texture we can define shape and texture coefficients c_is and c_it Kristina Scherbaum scherbaum@mpi-inf.mpg.de

4 Aging Trajectories Main Idea … compute aging trajectories z(t) locally along gradient of the aging function f(x) and going through a start vector or face x0: for all x and t the gradient of f describes the direction of minimal change in x to achieve a given change in t so that the characteristic features of the face are retained in the best way possible Runge Kutta (4th order): performs small steps along the gradient of f the gradient of f can be computed from the RBF regression function Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Aging Information extracted from the database of 200 adult face scans and new database of 238 face scans of teenagers teenager overview Kristina Scherbaum scherbaum@mpi-inf.mpg.de

Kristina Scherbaum scherbaum@mpi-inf.mpg.de