1. Face and Pose Tracking Kat Bradley Kaylin Spitz

2. General Layout (Detection) Right Image Left Image Face Detection Feature Detection Feature Correspondence Pose Detection Face Location (Left) Feature Location (Left) 3D Points

3. General Layout (Tracking) Right Image Left Image Face Tracking Feature Tracking Feature Correspondence Pose Detection Feature Location (Left) 3D Points Previous Face Previous Features Feature Filtering Feature Location (Left) Face Location (Left) Re-Detection If unsuccessful

4. Face Detection Performed on left frame initially & every 20 frames Uses Haar classifiers Slow (~350 ms for one frame)

5. Face Detection Performance Requires frontal face Occasionally (about 5%) misidentifies Possible Improvements

6. Feature Detection SURF features Detects in a single frame Takes about 130 ms

7. Face Tracking: Original Mean-shift color procedure proposed by Comaniciu. et al Target: histogram of color distribution from initial frame Tracking by comparing distribution to target distribution (mean-shift)

8. Face Tracking: Improvements Using two color spaces (robustness) Sampling (speed) Takes about 40 ms per frame

9. Face Tracking Performance Highly dependent on initial face Robust to changes in size and expression changes Issues with lighting changes

10. Feature Tracking Optical Flow (Lucas- Kanade) Performed in left frame

11. Feature Filtering Finds mean and standard deviation of offset (for points in face) Filters away points many standard deviations away from mean Filters away points far from face Signals if few points are in the face (to trigger re- detection)

12. Feature Correspondence Optical Flow (Lucas- Kanade) Gives matched points for pose detection Filters out points with high error

13. Optical Flow Performances Both moderately reliable without filters Without filters, problems with occlusions With filter, highly reliable

14. Pose Approximation Fits a plane to 3D points Normal of plane = approximate direction of face

15. Pose Performance Best on well-distributed features Issues with poorly-distributed features Possible improvements

16. Efficiency/Speed Face Detection:350 ms Face Tracking:40* ms Feature Detection:130 ms Feature Tracking:20 ms Feature Correspondence:25 ms Pose Approximation:<1 ms *easily parallelizable Tested on 800x640 image with face about 100x100.

17. Acknowledgements Thanks to: Ruzena Bacjsy, Gregorij Kurillo, and everybody at the Tele-Immersive Lab for help, support, and lots of explaining. Distributed Research Experiences for Undergraduates for funding.