Presentation is loading. Please wait.

Presentation is loading. Please wait.

Learning the Appearance of Faces: A Unifying Approach for the Analysis and Synthesis of Images. Thomas Vetter Germany University of Freiburg

Similar presentations


Presentation on theme: "Learning the Appearance of Faces: A Unifying Approach for the Analysis and Synthesis of Images. Thomas Vetter Germany University of Freiburg"— Presentation transcript:

1 Learning the Appearance of Faces: A Unifying Approach for the Analysis and Synthesis of Images. Thomas Vetter Germany University of Freiburg http://graphics.informatik.uni-freiburg.de

2 Computer Vision & Computer Graphics Computer Graphics can help to solve Computer Vision! | G (p) - I | 2 = min Parameters G ( image ) Parameters V ision ( image ) parameters image G raphics ( parameters )

3

4

5 Analysis by Synthesis 3D World Image Analysis Synthesis Image Model Image Description model parameter

6 Synthesis of Faces Input Image Modeler Result Database Face Analyzer 3D Head Morphable Face Model

7 Approach: Example based modeling of faces 2D Image 3D Face Models = w 1 * + w 2 * + w 3 * + w 4 * +...

8 Cylindrical Coordinates red(h,  ) green(h,  ) blue(h,  ) red(h,  ) green(h,  ) blue(h,  )  h radius(h,  ) h 

9 Morphing 3D Faces 3D Blend 3D Morph 1 __ 2 1 2 + =

10 Correspondence: A two step process! Correspondence between 1. two examples ( Optical Flow like algorithms). 2. many examples ( Morphable Model ) Reference Example 2 nd Example

11 = a 1 * + a 2 * + a 3 * + a 4 * +... b 1 * + b 2 * + b 3 * + b 4 * +... Vector space of 3D faces. A Morphable Model can generate new faces.

12 Manipulation of Faces Modeler

13 Modelling in Face Space Caricatur Original Average

14 Modelling the Appearance of Faces A face is represented as a point in face space. Which directions code for specific attributes ?

15 Learning from Labeled Example Faces Fitting a (linear) regression function

16 Facial Attributes WeightWeight OriginalOriginal Subjective Attractiveness

17 Transfer of Facial Expressions = Smile - - Originals: + Smile = Novel Face:

18 Facial Expressions OriginalOriginal

19 3D Shape from Images Face Analyzer 3D Head Input Image

20 Matching a Morphable 3D-Face-Model = R Optimization problem! a 1 * + a 2 * + a 3 * + a 4 * +.. b 1 * + b 2 * + b 3 * + b 4 * +..

21 Error Function Image difference Plausible parameters Minimize

22 Optimization Strategies Stochastic Gradient Decent Difference Decomposition

23 Future Challenges Which Object Classes are linear ? How to built them automatically?


Download ppt "Learning the Appearance of Faces: A Unifying Approach for the Analysis and Synthesis of Images. Thomas Vetter Germany University of Freiburg"

Similar presentations


Ads by Google