1 / 25

Facial Action Units (AU)

Selective Transfer Machine for Personalized Facial Action Unit Detection Wen-Sheng Chu , Fernando De la Torre and Jeffery F. Cohn Robotics Institute, Carnegie Mellon University July 9, 2013. Facial Action Units (AU). AU 6+12. Main Idea. Related Work: Features. Related Work: Classifiers.

jacob-ruiz
Download Presentation

Facial Action Units (AU)

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Selective Transfer Machine for Personalized Facial Action Unit DetectionWen-Sheng Chu, Fernando De la Torre and Jeffery F. CohnRobotics Institute, Carnegie Mellon UniversityJuly 9, 2013

  2. Facial Action Units (AU) AU 6+12

  3. Main Idea

  4. Related Work: Features

  5. Related Work: Classifiers

  6. Feature Bias Person specific!

  7. Occurrence Bias

  8. Selective Transfer Machine (STM) Formulation Minimize distribution mismatch Maximizes margin of penalized SVM

  9. Goal (1): Maximize penalized SVM margin margin penalized loss

  10. Goal (2): Minimize Distribution Mismatch • Kernel Mean Matching (KMM)* * “Covariate shift by kernel mean matching”, Dataset shift in machine learning, 2009.

  11. Goal (2): Minimize Distribution Mismatch Groundtruth Bad estimator for testing data!

  12. Goal (2): Minimize Distribution Mismatch Groundtruth Selection by reweighting training data Better fitting!

  13. Optimization: Alternate Convex Search

  14. Optimization: Alternative Convex Search

  15. Compare with Relevant Work [1] "Covariate shift by kernel mean matching," Dataset shift in machine learning, 2009. [2] "Transductive inference for text classification using support vector machines," In ICML 1999. [3] "Domain adaptation problems: A DASVM classification technique and a circular validation strategy," PAMI 2010.

  16. Experiments • Features • SIFT descriptors on 49 facial landmarks • Preserve 98% energy using PCA

  17. Experiment (1): Synthetic Data

  18. Experiment (2): Comparison with Person-specific (PS) Classifiers • Two protocols • PS1: train/test are separate data of the same subject • PS2: training subjects include test subject (same protocol in [2]) • GEMEP-FERA

  19. Experiment (2): Selection Ability of STM

  20. Experiment (3): CK+ • 123 subjects, 597 videos, ~20 frames/video

  21. Experiment (4): GEMEP-FERA • 7 subjects, 87 videos, 20~60 frames/video

  22. Experiment (5): RU-FACS • 29 subjects, 29 videos, 5000~7000 frames/vid

  23. Summary • Person-specific biases exist among face-related problems, esp. facial expression • We propose to alleviate the biases by personalizing classifiers using STM • Next • Joint optimization in terms of • Reduce the memory cost using SMO • Explore more potential biases in face problems, e.g., occurrence bias

  24. Questions? [1] "Covariate shift by kernel mean matching," Dataset shift in machine learning, 2009. [2] "Transductive inference for text classification using support vector machines," In ICML 1999. [3] "Domain adaptation problems: A DASVM classification technique and a circular validation strategy," PAMI 2010. [4] “Integrating structured biological data by kernel maximum mean discrepancy”, Bioinformatics 2006. [5] “Meta-analysis of the first facial expression recognition challenge,” IEEE Trans. on Systems, Man, and Cybernetics, Part B, 2012. http://humansensing.cs.cmu.edu/wschu/

More Related