1 / 43

Linear Tracking

Linear Tracking. Jan-Michael Frahm COMP 256. Some slides from Welch & Bishop. Tracking . Tracking is the problem of generating an inference about the motion of an object given a sequence of images.

kapila
Télécharger la présentation

Linear Tracking

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. Linear Tracking Jan-Michael Frahm COMP 256 Some slides from Welch & Bishop

  2. Tracking Tracking is the problem of generating an inference about the motion of an object given a sequence of images. The key technical difficulty is maintaining an accurate representation of the posterior on object position given measurements, and doing so efficiently.

  3. Examples of tracking

  4. Model for tracking • Object has internal state • Capital indicates random variable • Small represents particular value • Obtained measurements in frame i are • Value of the measurement

  5. General Steps of Tracking • Prediction: What is the next state of the object given past measurements • Data association: Which measures are relevant for the state? • Correction: Compute representation of the state from prediction and measurements.

  6. predict correct Tracking

  7. Independence Assumptions • Only immediate past matters • Measurements depend only on current state • Important simplifications Fortunately it doesn’t limit to much!

  8. Linear Dynamic Models • State is linear transformed plus Gaussian noise • Relevant measures are linearly obtained from state plus Gaussian noise • Sufficient to maintain mean and standard deviation

  9. A really simple example We are on a boat at night and lost our position • We know: • star position

  10. Constant Velocity p is position of boat, v is velocity of boat state is We only measure position so

  11. Marc makes a measurement Conditional Density Function , -2 0 2 4 6 8 10 12 14

  12. Jan makes a measurement Conditional Density Function , -2 0 2 4 6 8 10 12 14

  13. Combine measurements & variances Conditional Density Function -2 0 2 4 6 8 10 12 14 Online weighted average!

  14. Rudolf Emil Kalman • Born 1930 in Hungary • BS and MS from MIT • PhD 1957 from Columbia • Filter developed in 1960-61 • Now retired

  15. Kalman filter • Just some applied math. • A linear dynamic system: f(a+b) = f(a) + f(b) • Noisy data in  hopefully less noisy out. • But delay is the price for filtering...

  16. predict correct Predict  Correct KF operates by • Predicting the new state and its uncertainty • Correcting with the new measurement

  17. What is it used for? • Tracking missiles • Tracking heads/hands/drumsticks • Extracting lip motion from video • Fitting Bezier patches to point data • Lots of computer vision applications • Economics • Navigation

  18. A really simple example We are on a boat at night and lost our position • We know: • move with constant velocity • star position

  19. -2 0 2 4 6 8 10 12 14 But suppose we’re moving • Not all the difference is error. Some may be motion • KF can include a motion model • Estimate velocity and position

  20. Process Model • Describes how the state changes over time • The state for the first example was scalar • The processmodel was “nothing changes” • A better model might be constant velocity motion

  21. Measurement Model “What you see from where you are” not “Where you are from what you see”

  22. Constant Velocity p is position of boat, v is velocity of boat state is We only measure position so

  23. State and Error Covariance • First two moments of Gaussian process Process State (Mean) Error Covariance

  24. The Process Model Process dynamics Uncertainty over interval State transition Difficult to determine

  25. Measurement Model Measurement relationship to state Measurement uncertainty Measurement matrix

  26. Predict (Time Update) a priori state, error covariance, measurement

  27. Kalman gain Measurement Update (Correct) a posteriori state and error covariance Minimizes posteriori error covariance

  28. The Kalman Gain Weights between prediction and measurements to posteriori error covariance For no measurement uncertainty State is deduced only from measurement

  29. The Kalman Gain Simple univariate (scalar) example a posteriori state and error covariance

  30. Summary PREDICT CORRECT

  31. Estimating a Constant The state transition matrix The measurement matrix Prediction

  32. Measurement Update

  33. Setup/Initialization

  34. State and Measurements = 0.1

  35. Error Covariance

  36. State and Measurements = 1

  37. State and Measurements = 0.01

  38. Example camera pose estimation

  39. Kalman Filter Web Site • Electronic and printed references • Book lists and recommendations • Research papers • Links to other sites • Some software • News http://www.cs.unc.edu/~welch/kalman/

  40. Java-Based KF Learning Tool • On-line 1D simulation • Linear and non-linear • Variable dynamics http://www.cs.unc.edu/~welch/kalman/

  41. KF Course Web Page http://www.cs.unc.edu/~tracker/ref/s2001/kalman/index.html ( http://www.cs.unc.edu/~tracker/ ) • Java-Based KF Learning Tool • KF web page

  42. Closing Remarks • Try it! • Not too hard to understand or program • Start simple • Experiment in 1D • Make your own filter in Matlab, etc. • Note: the Kalman filter “wants to work” • Debugging can be difficult • Errors can go un-noticed

  43. Relevant References • Azarbayejani, Ali, and Alex Pentland (1995). “Recursive Estimation of Motion, Structure, and Focal Length,” IEEE Trans. Pattern Analysis and Machine Intelligence 17(6): 562-575. • Dellaert, Frank, Sebastian Thrun, and Charles Thorpe (1998). “Jacobian Images of Super-Resolved Texture Maps for Model-Based Motion Estimation and Tracking,” IEEE Workshop on Applications of Computer Vision (WACV'98), October, Princeton, NJ, IEEE Computer Society. • http://mac-welch.cs.unc.edu/~welch/COMP256/

More Related